Issuu

JSES The Journal of Sport and Exercise Science, Vol. 10, Issue 2, 1-9 (2026)

ISSN: 2703-240X

www.jses.net

An emergent athlete development framework for Swiss Triathlon Cameron Lamont1, Andrew J. Martin1*, Geoff Watson1 1

School of Sport, Exercise & Nutrition, Massey University, New Zealand ARTICLE INFO Received: 01.20.2026 Accepted: 11.08.2026 Online: 14.10.2026 Keywords: Athlete development Athlete-centred Case-study Auto-ethnography Coach education

ABSTRACT Issues surrounding selection, de-selection, and non-selection within National Sport Organizations (NSOs) have been shown to negatively affect athletes' self-esteem and longterm retention in sport (Edwards & Brannagan, 2023; Wrang, 2023). In New Zealand, these concerns and stakeholder complaints have prompted recent reviews of high-performance sport, resulting in policy changes aimed at safe-guarding athlete well-being (Rew, 2025; Roberts et al., 2021, Winther et al., 2025). This article examines how a practitioner addressed limitations in existing athlete development pathways while maintaining alignment with established frameworks. The lead author, a former professional athlete from New Zealand and current Swiss Duathlon’s National Coach within Swiss Triathlon, critically reflects on their practice through an auto-ethnographic lens. Two widely applied athlete development frameworks are examined from a practitioner perspective, the LongTerm Athlete Development Model (LTAD; Balyi & Hamilton, 2004) and Foundation, Talent, Elite, and Master Framework (FTEM; Gulbin et al., 2013, 2014). Guided by Schön’s (1983) concept of the ‘reflective practitioner’, the study’s trustworthiness is enhanced through triangulation, cross referencing to a bricolage of secondary sources, strengthening credibility and confirmability. Findings informed the creation of the LEAD framework, an emergent athlete-centred framework subsequently accepted by Swiss Olympic, that provides practical protocols for selection, de-selection, and non-selection. The LEAD framework supports resource management, talent identification and transfer, and is adaptable and transferable to other triathlon and broader sport contexts.

Introduction

1.1. Athlete development frameworks: Policy, selection and athlete well-being Recent research on National Sport Organization (NSO) policy highlights that athlete development frameworks operate as strategic governance tools shaped by political, financial, and performance pressures (Ives et al., 2023; Gale et al., 2025; O’Gorman et al., 2021). Coaches and performance staff are responsible for enacting these frameworks while managing accountability demands and athlete welfare, often within constrained funding environments. As a result, development

systems may prioritise efficiency and short-term outcomes over holistic athlete care. However, contemporary literature increasingly highlights the ethical responsibility of sport systems to balance performance outcomes with athlete well-being, particularly during transitions such as selection, non-selection, and deselection (Brown et al., 2018; Henriksen et al., 2018). Historically, selection criteria have focused primarily on physical testing and performance data, despite evidence that early success poorly predicts long-term elite performance (Till & Baker, 2020). Such practices risk overlooking maturational variations, late entry into sport and psycho-social contexts. Selection-related transitions are recognised as risk points impacting self-esteem and retention, particularly in high-

Corresponding Author: Andrew J. Martin, School of Sport, Exercise & Nutrition, Massey University, Nutrition, a.j.martin@massey.ac.nz

JSES | https://doi.org/10.36905/jses.2026.02.01

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026) performance and youth contexts (Edwards & Brannagan, 2023; Wrang et al., 2023). Recent recommendations call for policy frameworks to incorporate psychological support to safeguard athlete well-being (Rew, 2025). Similarly, studies of New Zealand elite athletes indicate selection pressures contribute to psychological distress, and hence the need for greater organizational support (Winther et al., 2025). Coaches play a central role in talent identification, athlete selection, development, and welfare; however, practitionerinformed research remains limited (Stolz & Pill, 2014). While policy frameworks provide structure, coaches operationalise these policies in daily practice, making athlete-coach relationships critical to achieving both performance and holistic development (Kidman, 2005; Pill, 2018; Vinson & Bell, 2019). Despite widespread adoption of structured athlete development models, there is limited applied research demonstrating how NSO frameworks can integrate athletecentred welfare principles, particularly related to selection, nonselection, and deselection, without compromising performance objectives. This study addresses this gap by presenting a practitioner-informed case study illustrating how athlete-centred development practices can be embedded within existing NSO frameworks. The paper draws on the lived experience of a highperformance athlete and coach, supported by contemporary scholarship and practitioner insights. The practical applications of this are encapsulated in the Lamont Emergent Athlete Development (LEAD) Framework, which has been accepted by Swiss Triathlon. 1.2. Theoretical and conceptual concepts Talent identification (TID) programs are important strategic initiatives for NSOs, but must be aligned with long-term development, ethical responsibility, and athlete welfare (Collins et al., 2022; Reeves et al., 2018). NSOs often prioritize early selection and short-term success, undermining long-term development pathways (Taylor & Collins, 2020). Research emphasises that athlete progression is non-linear and more closely associated with developmental stage than chronological age (Gagné, 2020). The Long-Term Athlete Development (LTAD) model (Balyi & Hamilton, 2003) and the Developmental Model of Sport Participation (DMSP; Côté & Fraser-Thomas, 2007) have influenced global development pathways. However, critiques suggest limited empirical support and insufficient consideration of psychosocial and contextual factors (Côté & Gilbert, 2009; Ford et al., 2011). The Australian Institute of Sport’s FTEM (Foundation, Talent, Elite, Mastery) framework provides greater flexibility by allowing athlete progression based on readiness rather than age (AIS, n.d.). Despite this adaptability, FTEM and its international derivatives remain predominantly physically focused. Notably, the removal of psychological components from some adaptations contrasts with growing evidence emphasising the need for greater athlete care and wellbeing support (Campbell et al., 2021). Collectively, these models highlight a persistent gap between pathway structure and athlete experience, particularly in relation to selection, non-selection, and deselection processes. Calls for athlete-centred coaching, supported by mentoring and JSES | https://doi.org/10.36905/jses.2026.02.01

practitioner-led development, continue to grow (Kidman & Lombardo, 2010; Lascu et al., 2024). However, research rarely demonstrates how these principles can be systematically integrated within existing NSO frameworks, an issue this current study seeks to address. 2.

Methods

2.1. Epistemology framework This research is situated within a social constructivist epistemology (Cresswell, 2007), emphasizing the role of social interaction and shared meaning in shaping individuals' understanding of the world. This perspective enabled a practical examination of change management in real time from a practitioner’s experiences, informing policy within a highperformance National Sport Organization (NSO). This approach culminated in an emergent athlete development framework for Swiss Triathlon with potential adaptability and transferability to other sport contexts. The transferability of sport coaching research from scholarly to practical domains has been questioned (Lyle, 2018), noting a disconnect between practice and therefore being largely absent from coach education. Stolz and Pill (2014, p. 28) argue for a shift from purely “empirical-scientific research to practitionerreferenced research” where reflective practice enables coaches to critically evaluate and learn from their own experiences. Auto ethnographical approaches within sport research offer a means to address this gap by illuminating lived experiences of high performance environments while situating them within broader cultural and institutional structures (Anderson, 2006; Ellis et al., 2011; Sparkes & Smith, 2014). 2.2. Research design: Case study and auto-ethnography This study utilizes an auto-ethnographic approach (Ellis 2012), combined with an instrumental case study design (Merriam, 1998; Yin, 2009). The lead author, as a ‘reflective practitioner’ (Schön, 1983), drew on experiences across regional, national, and international coaching roles whilst also contributing to athlete development policy within Swiss Triathlon. This dual positioning enabled critical examination of athlete progression and the implementation of system change within a high-performance environment. Swiss Triathlon was selected as a case study, due to its governance being situated within an established Olympic sport and its affiliation with international federation structures. Triathlon’s global participation base ensures that developmental, coaching, and governance challenges observed in this context are potentially comparable to those encountered in other NSOs. 2.3. Reflective, analytical, and evaluative processes Auto-ethnography enabled systematic examination of coaching experiences and decision-making processes, providing a rich, reflective narrative, incorporating an iterative approach integrating experience, theory, and dialogue. To enhance methodological rigor and trustworthiness, a multi-method bricolage approach (Denzin & Lincoln, 2011) was employed. This included auto-ethnographic reflection from the lead author’s 2

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026) athlete and coach perspectives, document analysis of Swiss Triathlon policies (e.g. testing and selection) from 2020-2024, and discussions with two external international coaches and two internal Swiss Triathlon staff. The reflective and analytical process followed Slade et al.’s (2020) practitioner auto-ethnography continuum, comprising three stages to enhance the qualitative research’s credibility, confirmability, and analytical transparency: 1. Life-history biography or reflective journaling (Angier, 2010), documenting (30,000 words) 25 years of experiences (from 2020-2024), as an elite athlete and coach. 2. Identifying transformative career stages (Ellis, 2012), highlighting critical incidents and recurring themes informing the LEAD framework

3. Analytical ethnography (Anderson, 2006), triangulating personal insights with scholarly literature, archival documents, and practitioner perspectives. This process facilitated progression from descriptive reflection (in Stage 1), to identifying key themes (in Stage 2), to triangulation, analytical interpretation, and evaluative judgment (in Stage 3), allowing assumptions (by the lead author) to be challenged (by co-authors, external coaches, and internal staff) and refined. Reflective data from the lead author’s doctoral thesis (Lamont, 2025), documented three phases, (1) athletic development and high-performance pathway (2000-2009), (2) coach development (2010-2019), and (3) high-performance coach pathway (2020-2024). Figures 1 and Figure 2 illustrate the emergent research process and reflective analysis and evaluation process.

Figure 1: An emergent research process.

Figure 2: An emergent evolving model of auto-ethnographic narrative (Slade et al., 2020). JSES | https://doi.org/10.36905/jses.2026.02.01

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026) 2.4. Practitioner perspectives and theoretical positioning The research is underpinned by an ‘athlete-centred’ coaching philosophy, which empowers athletes to take responsibility for their sporting behaviours and decision-making (Kidman & Lombardo, 2010). Vinson and Bell (2019, p. 40) note that “at its simplest level, the term means putting the needs of the athlete at the forefront of the coaching process.” Kidman and Lombardo (2010) highlight that mentoring and delegating responsibility enables athletes to assume greater ownership, thereby maximizing development, performance, and overall enjoyment. This approach aligns with contemporary critiques of dominant athlete development models such as LTAD and FTEM, which have been challenged for their linearity and limited responsiveness to individual, contextual, and welfare-related considerations (Bjørndal et al., 2018; Sieghartsleitner et al., 2019). While auto-ethnography inherently involves subjectivity, potential limitations such as confirmation bias were mitigated through systematic triangulation, peer dialogue, and theoretical critique, as suggested by Onwuegbuzie et al. (2024). Prior to undertaking the research, the project was evaluated by the University’s research ethics committee and judged to be low risk. 3.

Discussion

Example: Supporting athletes through periods of injury or de-selection. Recognizing athlete’s education, mental health, and self-development. Fostering team culture and leadership. 3.3. Coach education The framework adopts a “person-first, athlete-second” philosophy, recognizing that coaching fundamentally involves guiding people through change. For meaningful changes, coaches must act as catalysts for positive purposeful transformation. Their responsibilities extend beyond preparing athletes for competition and testing, as well as helping them deal with other stresses in their lives. However, robust support structures are essential.

3.1. The LEAD Framework The study resulted in the Lamont Emergent Athlete Development (LEAD) Framework (Figure 3), a practitioner-informed model designed to operate alongside the Foundations, Talent, Elite, and Mastery (FTEM) framework commonly mandated by National Sport Organizations (NSOs). While FTEM provides a structural classification of athlete stages, the LEAD framework addresses limitations relating to athlete welfare, selection processes, coaching practice, and system adaptability. Drawing on practitioner experience within Swiss Triathlon, the LEAD framework integrates selected elements of LTAD and FTEM while challenging their linear assumptions and limited responsiveness to individual and contextual factors. The findings highlight how athlete development is shaped not only by performance metrics, but by coaching philosophy, leadership practices, and institutional constraints. As an emergent framework, LEAD is cognizant of a coach’s duty of care for athletes’ physical, psychological, and emotional development (Bouchard et al., 1994), offering practical guidance for policy and practice within Swiss Triathlon and similar high-performance systems. 3.2. Athlete-centred philosophy The LEAD framework is grounded on an athlete-centred coaching philosophy, which although developed within the context of triathlon is potentially applicable to other sports. Its core principles are: 1.

‘Comfort the troubled and trouble the comfortable’: A commitment to athlete-centred care, prioritizing physical and mental well-being, whilst applying best practice.

JSES | https://doi.org/10.36905/jses.2026.02.01

‘Create an environment where individuals can grow, as an athlete and a person.’ A shift from, “We create a place where champions grow.” Reflects a commitment to long-term personal and holistic growth. Transformational leadership. Demonstrating ethical coaching practice, strong work-ethic, whilst having fun, to empower athletes to exceed expectations. Bass and Anolio (1994) argue that such leaders and followers advance together toward a higher level of morale and motivation.

Example: Integrating education on managing athlete emotions and stakeholder relationships during selection processes at all coaching qualification levels. Over time, this approach evolves into a caring team culture, grounded by an athlete-centred coaching philosophy, comprising the following components: 1. 2. 3. 4. 5. 6. 7. 8.

Effective use of guided learning (Billett, 2000): Encouraging athletes to take primary responsibility for their development. Support infrastructure: Providing consistent communication, encouragement, and resources for athletes. Role modelling and mentoring: Adapting coaching styles to suit individual athletes and groups. Transformational and situational leadership: Building personal connection and contextual understanding. Scientific coaching influence: Designing effective training plans informed by evidence-based practice rather than tradition. Deliberate, meaningful practice (Ericsson et al., 1993): Establishing a high-performance culture which becomes self-regulating over time. Leading and managing change: Being a catalyst for purposeful system improvement. Mind management (Hermansson, 2011): Teaching athletes’ coping strategies to develop psychological resilience. Example: Athletes respond positively to clear, consistent guidance with improvement evident, reinforcing confidence in both athlete decision-making and coach judgement. 4

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026)

Giftedness/TID

Athlete-Centred Focus

National Coach Junior involved with TID and athlete development. Helps in the transition from different race series and when moving into Elite. Supports the TID process and a constant person as athletes move forward in the system.

National Coach U23 / Elite works mainly with the Elite athletes, but also starts to work with Juniors to become familiar with them and help guide development to have athletes ready for transitions.

Assistant Coaches Area of Responsibility

National Coach Junior Area of Responsibility

All coaches directly responsible for clear, open communication, care and for their athletes’ wellbeing.

As many athletes come into the early stage of Nachwuchs system. Assistant and National Coach Junior looking at TID and long-Term development.

Assistant coaches help with TID and the early stages of Youth development. Very important to help support National Coach Junior with the athlete development and camps.

Nachwuchs Kader to become a place for development and the main focus for National Coach Junior

Total athletes attending Kader tests and Racing

FOUNDATION (FTEM) Active Start; FUNdamentals; Learning to Train (LTAD)

National, Regional, Local: Levels of Nachwuchs (Youth) Kader Note: National & Regional have talent cards and are limited by Swiss Olympic. Local talents do not have a talent card and therefore are not limited in number of athletes in this group.

TALENT (FTEM) Training to Train (LTAD)

Managing Staff Resources

Managing Athlete Resources

The transition from Junior to U23 is a major step and Total number of A, B and C Kader athletes remain requires time for the athlete to adjust. This is a difficult and similar, with more attention from National Coach challenging time. Having support for the athlete is important U23 and Elite. to allow them develop to this level. As the number of places in C, B and A Kader are limited, in the past de-selection and non-selection National Coach U23 / Elite have been high here. With D Kader this number could be reduced. Also, any athletes de-selected or nonArea of Responsibility selected given clear pathways back into the system.

MASTERY (FTEM) Training to Win; Active for Life (LTAD) A, B & C Levels Elite Kader

Elite D Squad

Elite D squad to allow transfer between C and D, some athletes may not hit limits for C, but are still in National system, which helps the total squad and gives more time to develop.

Talent Transfer Creating a robust Talent Transfer System would be important, offering time with a highly qualified coach, to harness athletes from other sports.

Selection Protocols: De-selection, Non-selection Clear selection criteria for National & Regional. Deselection, non-selection protocols and support. Local can be more holistic as not controlled by Swiss Olympic. More athletes can be selected and developed using Local in this way.

Coach Education Talent Transfer System

ELITE (FTEM) Training to Compete (LTAD)

Coach education is designed to meet the requirements of athletes and the sport of triathlon. Coaches understand the systems in place and the ramifications on athletes and other stakeholders of selection, de-selection and non-selection

LEAD Framework: Swiss Triathlon

Figure 3: Swiss Triathlon LEAD framework (Lamont, 2025). JSES | https://doi.org/10.36905/jses.2026.02.01

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026) 3.4. Selection, de-selection, and non-selection protocols Reflections from the lead author highlight several key insights regarding athlete selection: 1. 2.

Junior performance is not a reliable predictor of senior success: Early results often fail to indicate long-term potential. Selection should not be based on physical testing alone: Triathlon is a dynamic sport, and athletes may perform differently in controlled tests compared to race conditions. Talent identification requires a holistic approach: Personal attributes such as attitude and resilience should be considered alongside physical metrics to support long-term development. Example: Providing structured feedback and clearly defined re-selection pathways to retain motivated athletes within the system.

Current practices around deselection and non-selection vary significantly between countries, with many athletes receiving little or no feedback on why they were not selected or what steps they could take to re-enter the pathway. This lack of communication potentially leads to disengagement and attrition from the sport. Addressing these issues requires systemic change, which is inherently a long-term process demanding careful planning and management. As the lead author reflected: My professional cycling experiences taught me selection is not always fair nor understandable. I had always wanted to wear the NZ silver fern as an athlete and have subsequently been in both NZ Duathlon and Triathlon age group teams, but not as a professional cyclist. The desire to achieve selection in teams can be a powerful motivator but should not be the only one. Continual improvement of yourself as an athlete is more important than any selection or non-selection. The lead author’s experiences as an athlete development coach underscored the importance of designing robust selection, nonselection, and de-selection processes linked to TID and athlete pathways. Based on these reflections, key characteristics which contribute to athlete success are: 1. 2. 3. 4. 5.

Appreciating sport as a source of personal development: Providing social capital (Bourdieu, 1984), where status and position influence opportunities (Claridge, 2015). Foundational values: Understanding the importance of work ethic, honesty, and integrity. Self-motivation, self-improvement, self-accountability, and self-reliance: Fostering independence and responsibility. Resilience and well-being: The ability to learn from mistakes and strive for continuous improvement. Perseverance and mental health awareness: Recognizing and managing challenges such as stress and depression.

JSES | https://doi.org/10.36905/jses.2026.02.01

These characteristics reinforce the LEAD framework’s emphasis on an athlete-centred philosophy and the need for development systems that prioritize holistic growth over shortterm outcomes. The findings have also highlighted that there is a need for much greater training and education of coaches and administrators around impacts of de-selection and non-selection. As Edwards and Brannagan (2023) highlight, de-selection can negatively affect an athletes' self-esteem and personal status, as well as leading to distrust of coaches and organizations. It is important to highlight clear reasons and answer questions about the athlete's de-selection. Wrang (2023) suggest questions like “why didn’t they choose me”, and “why am I no longer part of the team?” and others are common after a youth selection process. Roles of parents and coaches need to be considered within the selection and de-selection process, as these have social implications for youth athletes (Wrang, 2023). Incorporating greater support for athletes during and after de-selection and nonselection is an important addition to the LEAD framework. This includes ensuring selection-related procedures are clearly documented and understood by athletes, coaches, and administrators, thereby promoting transparency and athlete welfare. 3.5. Giftedness, talent Identification, and talent transfer Historically, Swiss Triathlon has relied on clubs and coaches to identify and develop gifted athletes. The LEAD framework proposes that talent identification and development (TID) become part of national coaching roles responsibilities. Gagné (2020) emphasized the distinction between giftedness (natural abilities) and talent, noting that giftedness evolves into talent through structured training and development. Example: National Junior Coaches actively assessing giftedness at sanctioned events and guiding athletes into supportive club environments. Within the FTEM framework, Swiss Triathlon highlights talent transfer and a potential pathway for athletes entering the development system. Yet, if an athlete does transfer across from another sport they appear to be expected to immediately meet the same requirements as all other athletes. As a result, within the current Swiss Triathlon system, athlete transfer is very difficult, and is mainly reliant on the clubs assessing an athlete’s potential to be converted, and then taking athletes to races, hoping they can perform. A good example of how a talent transfer system could work is from the USA (Newland & Kellet, 2012) where USA Triathlon actively recruits possible athletes to transfer, and then offers them a year of training with an experienced coach to evaluate their development potential. Talent transfer should therefore play a greater role and be actively supported by Swiss Olympic, and Swiss Triathlon. Both organizations need to promote talent transfer and implement systems that facilitate their success. 3.6. Managing athlete and staff resources: Structural constraints & system alignment

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026) A major finding concerns the tension between funding priorities and development needs. Swiss Olympic funding is closely tied to international performance outcomes, incentivising short-term success. While effective in securing immediate results, this focus risks undermining long-term talent development if pathways are not strategically supported. Sustainable success requires alignment across systems, from national policy to club environments, ensuring clubs are supported to develop athletes holistically rather than prioritising early results. Findings indicate that Swiss Triathlon requires greater investment at the club level, not solely when athletes reach national junior squads. Coach education is central to this process, emphasising:

Practical applications of the LEAD framework can potentially be transferred to clubs, teams, and other NSOs. It has been accepted by Swiss Olympic and positively received due to its alignment with the FTEM framework. It is suggested that this compatibility makes the LEAD framework appealing to any sport that needs a structure to support coaches and athletes. Wider research possibilities include examining how the LEAD framework is implemented by different sport organizations, and how the framework is adapted to meet the needs of the athletes, coaches, clubs, and NSOs. This approach could lead to further evolutions of the LEAD framework ensuring its adaptability and scalability for broader application to other organizations.

1. 2. 3.

Implications

Athlete welfare and transition management De-selection and non-selection support Long-term, incremental development strategies Example: Supporting clubs to reduce burnout risk by emphasizing long-term progression rather than early specialisation or short-term podium outcomes.

Conclusions This study bridges the gap between policy and practice, introducing the Lamont Emergent Athlete Development (LEAD) framework as a practitioner-informed contribution to coach education, NSO governance, and athlete development scholarship. Conceptually, LEAD extends existing athlete development models (e.g., FTEM, LTAD) by repositioning coaching practice, athlete welfare, and system adaptability as central mechanisms through which development pathways function effectively. Rather than proposing an alternative pathway structure, LEAD provides a practice-oriented framework that guides how coaches and organizations work within mandated models while responding to individual, contextual, and ethical demands. The LEAD framework emphasizes the importance of an athlete-centred mentoring approach to sport coaching, as advocated by Jones et al. (2009). Such approaches recognize and respond to athlete’s physical, psychological, and personal development needs, particularly during critical transition periods such as de-selection and non-selection. These moments were identified as important in shaping athlete experience, motivation, and long-term retention. Clear, transparent protocols are essential, not only for optimising performance but safeguarding athlete welfare. The LEAD framework reinforces the philosophy of ensuring athletes grow as people first, which is central to long-term success. From the lead author’s experiences, athletes who grow holistically will become more resilient and more likely to progress to elite levels. If they do not make the elite level, they acquire transferable life skills, applicable to other sports, business, or future employment. This process needs to be under continual review to ensure effective implementation (Astle et al., 2018). Athlete pathways are socially-mediated processes shaped by leadership practice, effective communication, and institutional systems. The LEAD framework demonstrates that short-term performance outcomes must be balanced with long-term development responsibility, to allow re-entry and athlete retention. JSES | https://doi.org/10.36905/jses.2026.02.01

Methodologically, this applied research demonstrates the value of practitioner auto-ethnography combined with case study. The lead author integrated unique insights gained from multiple roles, as an athlete, an individual coach, a team coach, and then, more recently, a national coach, including writing sport policy for Swiss Triathlon/Duathlon. These extensive experiences provided a strong foundation to build a credible, rich narrative of athlete development from a practitioner’s perspective, and a contribution to sport coaching research methodology. This multi-method approach enhanced qualitative trustworthiness by combining, case study (Merriam, 1998; Yin, 2009), auto-ethnography (Ellis, 2012; Ellis et al., 2011), and literature review, enabling triangulation (Slade et al., 2020). This approach leveraged the lead author’s coaching role experiences, as an insider researcher (Onwuegbuzie, 2024), to capture realtime examples and drive change in the form of an emergent athlete development framework. Slade et al. (2020) highlighted that practitioner auto-ethnography provides sport coaching scholars with an approach to frame and discuss issues while enhancing problem-solving capability. This approach also addresses Lyle’s (2018) questioning of the transferability of sport coaching research for practitioners and it not being incorporated into coach education. An overarching finding of this research is the importance of being a ‘reflective practitioner.’ Schön (1983) points out that through reflective practice and questioning professionals become more aware of shortcomings, subsequently adapting and/or changing their subsequent behaviours and decision-making processes. This self-reflective iterative process informed the creation of the LEAD framework, supported by literature review and dialogue with other coaches, to further support athlete development. Recommendations 1.

Embed athlete-centred mentoring within coach education: Coach education programs should extend beyond technical and planning competencies to include mentoring, welfare awareness, communication skills, and managing selection-related conversations at all pathway stages. Formalise selection, non-selection, and de-selection protocols: NSOs should ensure that criteria, decisionmaking processes, and re-selection pathways are clearly 7

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026)

documented, transparent, and communicated consistently. Structured feedback should be mandatory following non-selection and de-selection. Strengthen club, NSO alignment: Clubs require targeted support to implement athlete-centred practices aligned with national development objectives. Early investment at club level expands the developmental pool and improves long-term system sustainability. Integrate welfare and mental health considerations into development frameworks: Athlete welfare should be treated as a core performance enabler rather than a parallel concern, particularly during transitions and periods of uncertainty. Recognize coaches as change agents: Coaches should be supported and empowered to lead adaptive change within evolving systems, underpinned by reflective practice and continuous professional learning.

Conflict of Interest The authors declare no conflict of interests. Acknowledgment The authors have no acknowledgments to declare. Cameron Lamont is a National Coach of Swiss Duathlon, Swiss Triathlon and Doctoral Graduate, Massey University, New Zealand. Andrew Martin is a Professor of Sport and Physical Education at Massey University, Palmerston North, New Zealand. Geoff Watson is an Associate Professor in History at Massey University, Palmerston North, New Zealand. References Anderson, L. (2006). Analytic auto-ethnography. Journal of Contemporary Ethnography, 35(4), 373–395. https://doi.org/10.1177/0891241605280449 Angier, C. (2010). Reflections. In S. Cline & C. Angier (Eds.), The Arvon book of life writing: Writing biography, autobiography and memoir (pp. 5–21). Methuen Drama. Astle, A., Leberman, S., & Watson, G. (2018). Sport development in action: Plan, programme and practice. Routledge. Australian Institute of Sport. (n.d.). FTEM framework. https://www.ais.gov.au/ftem Balyi, I., & Hamilton, A. (2004). Long-term athlete development: Trainability in childhood and adolescence. National Coaching Institute British Columbia. Bass, B. M., & Avolio, B. J. (1994). Learning to share the vision: Improving organizational effectiveness through transformational leadership. Sage. Billett, S. (2000). Guided learning at work. Journal of Workplace Learning, 12(7), 272–285. https://doi.org/10.1108/13665620010355570 Bjørndal, C. T., Andersen, S. S., & Ronglan, L. T. (2018). ‘It takes two to tango’: Exploring the role of the coach-athlete relationship in the transition from junior to senior elite sport. Sport, Education and Society, 23(8), 774–787. https://doi.org/10.1080/13573322.2017.1320752 JSES | https://doi.org/10.36905/jses.2026.02.01

Bouchard, C., Shephard, R. J., & Stephens, T. (Eds.). (1994). Physical activity, fitness, and health: International proceedings and consensus statement. Champaign, IL: Human Kinetics. Bourdieu, P. (1984). Homo academicus. Stanford University Press. Brown, G., Webb, T., Robinson, M. A., & Cotgreave, R. (2018). Athletes’ experiences of social support during their transition out of elite sport: An interpretive phenomenological analysis. Psychology of Sport and Exercise, 36, 71–80. https://doi.org/10.1016/j.psychsport.2018.02.003 Campbell, N., Brady, A., & Tincknell-Smith, A. (2021). Developing and supporting athlete well-being: Person first, athlete second. Routledge. Claridge, T. (2015). Bourdieu on social capital: Theory of capital. https://www.socialcapitalresearch.com/bourdieu-on-socialcapital-theory-of-capital/ Collins, D., & MacNamara, Á. (2012). The rocky road to the top: Why talent needs trauma. Sports Medicine, 42(11), 907–914. https://doi.org/10.2165/11635110-000000000-00000 Côté, J., & Fraser-Thomas, J. (2007). Youth involvement in sport. In P. R. E. Crocker (Ed.), Sport psychology: A Canadian perspective (pp. 266–294). Pearson. Côté, J., & Gilbert, W. (2009). An integrative definition of coaching effectiveness and expertise. International Journal of Sports Science & Coaching, 4(3), 307–323. https://doi.org/10.1260/174795409789623892 Creswell, J. W. (2007). Qualitative inquiry and research design: Choosing among five traditions (2nd ed.). Sage. Denzin, N. K., & Lincoln, Y. S. (Eds.). (2011). The Sage handbook of qualitative research (4th ed.). Sage. Edwards, D., & Brannagan, P. M. (2023). Dealing with deselection from youth international football: A case study of English Premier League academies. International Journal of Sports Science & Coaching. https://doi.org/10.1177/17479541231216873 Ellis, C. (2012). The procrastinating auto-ethnographer: Reflections of self on a blank screen. The International Review of Qualitative Research, 5(3), 331–337. Ellis, C., Adams, T. E., & Bochner, A. P. (2011). Autoethnography: An overview. Historical Social Research, 36(4), 273–290. Ford, P., De Ste Croix, M., Lloyd, R., Meyers, R., Moosavi, M., Oliver, J., Till, K., & Williams, C. (2011). The long-term athlete development model: Physiological evidence and application. Journal of Sports Sciences, 29(4), 389–402. https://doi.org/10.1080/02640414.2010.536849 Gagné, F. (2020). Differentiating giftedness from talent. Routledge. Gale, L. A., Ives, B. A., Mitchell, T., Nichol, A. J., Stamp, D., Maskrey, C., Potrac, P. A., & Nelson, L. J. (2025). The enactment of political skills in community sport coaching. Sports Coaching Review, 1–14. https://doi.org/10.1080/21640629.2025.2520110 Gulbin, J., Croser, M., Morley, E., & Weissensteiner, J. (2013). An integrated framework for the optimisation of sport and athlete development: A practitioner approach. Journal of Sports Sciences, 31(12), 1319–1331. 8

Lamont et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 1-9 (2026) Gulbin, J., Croser, M., Morley, E., & Weissensteiner, J. (2014). A closer look at the FTEM framework. Journal of Sports Sciences, 32(8), 796–800. Henriksen, K., Storm, L. K., & Larsen, C. H. (2018). Organizational culture and influence on developing athletes. In C. Knight, C. Harwood, and D. Gould (Eds.). Sport psychology for young athletes (pp. 216-228). Routledge. Hermansson, G. L. (2011). Going mental in sport. Inside-Out Books. Howe, K. R. (2003). Closing methodological divides: Towards democratic educational research. Dordrecht: Kluwer Academic Publishers. Ives, B., Penney, D., O’Gorman, J., Nichol, A. J., Potrac, P., & Nelson, L. (2023). Investigating policy enactment in community sport coaching: Directions for future research. Sports Coaching Review, 15(1), 179–198. https://doi.org/10.1080/21640629.2023.2291253 Jones, R. L., Harris, R., & Miles, A. (2009). Mentoring in sports coaching: A review of the literature. Physical Education & Sport Pedagogy, 14(3), 267–284. Kidman, L. (2005). Athlete-centred coaching: Developing inspired and inspiring people. IPC. Kidman, L., & Lombardo, B. J. (2010). Athlete-centred coaching: Developing decision-makers (2nd ed.). IPC. Lamont, C. (2025). An emergent athlete development framework: Examining the case of Swiss Triathlon from a coach’s perspective Unpublished doctoral thesis, Massey University, Palmerston North, New Zealand. Lascu, A., Wood, M. A., Moulds, K., & Davids, K. (2024). Coach education as ‘leading out with an experienced other.’ Sports Coaching Review, 13(2), 277–292. Lyle, J. (2018). The transferability of sport coaching research: A critical commentary. Quest, 70(4), 419–437. https://doi.org/10.1080/00336297.2018.1453846 Merriam, S. B. (1998). Qualitative research and case study applications in education. San Francisco: Jossey-Bass. Newland, B., & Kellet, P. (2012). Exploring new models of elite sport delivery: The case of triathlon in USA & Australia. Managing Leisure, 17(2-3), 170–181. O’Gorman, J., Ives, B., Nichol, A., Potrac, P., & Nelson, L. (2021). Policy enactment and emotional labour in community sport coaching. Sport, Education and Society, 26(9), 1011–1024. https://doi.org/10.1080/13573322.2021.1886354 Onwuegbuzie, A. J., Abrams, S. S., & Abrams, M. L. (2024). Blending voices and methods: Exploring the frontier of (integrated) mixed methods autoethnography. Journal of Mixed Methods Studies, 9(1), 130–147. Pill, S. (2018). Perspectives on athlete-centred coaching. Taylor & Francis. Reeves, M. J., MacNamara, Á., & Collins, D. (2018). The role of talent identification in the development of expertise: Comparing the processes of two national governing bodies.

JSES | https://doi.org/10.36905/jses.2026.02.01

European Journal of Sport Science, 18(6), 783–793. https://doi.org/10.1080/17461391.2018.1444098 Rew, Q. (2025). Good sport: Developing a policy to safeguard high-performance athletes from psychosocial hazards. New Zealand Journal of Health and Safety Practice, 2(2), 45–58. Roberts, S., Boddy, P., & Fairclough, S. (2021). Talent identification and athlete welfare: Rethinking best practice. European Physical Education Review, 27(3), 514–529. https://doi.org/10.1177/1356336X20984138 Schön, D. (1983). The reflective practitioner: How professionals think in action. Basic Books. Sieghartsleitner, R., Zuber, C., Zibung, M., & Conzelmann, A. (2019). ‘The early specialised bird catches the worm!’ A specialised sampling model in the development of football talents. Frontiers in Psychology, 10, Article 188. https://doi.org/10.3389/fpsyg.2019.00188 Slade, D. G., Martin, A. J., & Watson, G. (2020). Practitioner auto-ethnography: Developing game centred learning. Curriculum Studies Health and Physical Education, 11(3), 252–264. https://doi.org/10.1080/25742981.2020.1778496 Sparkes, A. C., & Smith, B. (2014). Qualitative Research Methods in sport, exercise and health. Routledge. Stolz, S. A., & Pill, S. (2014). Teaching games and sport for understanding: Exploring and reconsidering its relevance in physical education. European Physical Education Review, 20(1), 36–71. https://doi.org/10.1177/1356336x13496001 Vinson, D., & Bell, J. (2019). Athlete-centred coaching. In E. Cope & M. Partington (Eds.), Sports Coaching: A Theoretical and Practical Guide (pp. 40–51). Routledge. Taylor, J., & Collins, D. (2020). “Shoulda, coulda, didnae”: Why don’t high-potential players make it? The Sport Psychologist, 34(2), 89–97. https://doi.org/10.1123/tsp.2018-0090 Till, K., & Baker, J. (2020). Challenges and [possible] solutions to optimizing talent identification and development in sport. Frontiers in Psychology, 11. https://doi.org/10.3389/fpsyg.2020.00664 Walinga, J., Obee, P., Cunningham, B., & Cyr, D. (2021). The role of leadership and team culture in enhancing sport performance outcomes. The International Journal of Sport and Society, 12(2), 81–104. Winther, T., de Terte, I., Wood, W., Beable, S., Wilson, K., & Hamilton, B. (2025). It was the best of times, it was the worst of times: Daily hassles and depressive symptoms in New Zealand elite athletes. Journal of Clinical Sport Psychology, 19(1), 1–15. https://doi.org/10.1123/jcsp.20240008 Wrang, C. M. (2023). Selection and de-selection in youth sport. In S. Agergaard & D. Karen (Eds.), Social Issues in Sport, Leisure, And Health (pp. 146–159). Routledge. Yin, R. K. (2009). Case study research: Design methods (4th ed.). Sage.

JSES The Journal of Sport and Exercise Science, Vol. 10, Issue 2, 17-37 (2026)

ISSN: 2703-240X

www.jses.net

Perceptions of gym staff toward resistance training technique in New Zealand Melissa Cuthbertson-Moon1,2*, Patria A. Hume1,2,3, Bryce Hastings4, Hannah E. Wyatt5 1

Auckland Bioengineering Institute, The University of Auckland, New Zealand Sports Performance Research Institute New Zealand (SPRINZ), Auckland University of Technology, New Zealand 3 Tech & Policy Lab, Law School, The University of Western Australia, Australia 4 Les Mills International, Chicago, United States. 5 Sport, Health and Rehabilitation Research Cluster, Faculty of Health, University of Canterbury, New Zealand 2

ARTICLE INFO Received: 06.08.2024 Accepted: 04.08.2026 Online: 24.10.2026 Keywords: Resistance Training Technique Gym Injury Musculoskeletal Injury Lower Back Shoulder

ABSTRACT Purpose: To gain understanding of New Zealand gym staff’s perceptions towards resistance training exercise technique of gym users using knowledge, attitudes, and behaviour [KAB] related questions, and to understand whether any fitness trends are emerging that are likely to cause injury. Methods: A Qualtrics XM online survey was distributed to staff of a gym chain in New Zealand. Questions probed respondents KAB towards resistance training exercises performed with poor or unsafe technique and those perceived to cause musculoskeletal injuries of the lower back and shoulders. Results: Ninety-nine full and partial responses are reported. Knowledge of respondents highlighted deadlifts, squats, cleans/clean & press, push ups, and chest press are frequently performed with poor and unsafe technique and perceived to cause injury. Ninety-six percent of respondents (n = 53) observed poor technique more than 40% of the time. All respondents perceived gym users were only sometimes or rarely aware of their poor technique, and 64% (n = 35) perceived gym users do not understand the link between poor technique and injury. Attitudes revealed that 85% of gym staff (n = 46) were only sometimes or never asked for help with performing exercises. Behaviours of respondents emphasised 62% (n = 33) would approach someone performing poor technique and 91% (n = 48) would intervene with unsafe technique. Emerging trends likely to cause injury are those that involve high complexity and highvolume training. Conclusions: There is opportunity for Gym Managers to increase knowledge, skills and confidence of Group Fitness Instructors and Personal Trainers for coaching deadlifts, squats, cleans/clean & press, push ups, and chest press technique to novice gym users, through staff training and upskilling. Attitudes can be improved by providing tools to facilitate conversations between gym staff and gym users. Behaviours should focus on building confidence of all gym staff to provide feedback and intervene with poor or unsafe technique.

Introduction

Physical activity helps maintain good health and reduce the risk of chronic disease (Dhuli et al., 2022; Warburton et al., 2006). Resistance training exercise can improve muscular strength and bone density (Gómez-Cabello et al., 2012; Nelson et al., 2007).

Despite benefits of regular resistance training, many people do not engage in resistance training on a regular basis (Harne & Bixby, 2005; Hurley et al., 2018). Perceived barriers to resistance training are time/effort and the physical hardships of weightlifting (Harne & Bixby, 2005; Hurley et al., 2018). A survey of 432 college students identified physical barriers of resistance training to

Corresponding Author: Melissa Cuthbertson-Moon, Auckland Bioengineering Institute, The University of Auckland, New Zealand, mcut256@aucklanduni.ac.nz

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) include looking silly and being too uncoordinated (Myers & Roth, 1997). Additional studies have validated these findings and reported a lack of knowledge as a contributing factor amongst females (Harne & Bixby, 2005; Hurley et al., 2018). A recent systematic review and meta-analysis of 20 papers, highlighted inadequate knowledge as a frequent barrier to resistance training, specifically “not knowing how to lift weights”, “what to do”, “which ones to do”, and not remembering how to perform the exercises or how to use equipment (Vasudevan & Ford, 2022). It has been recommended that weightlifters and bodybuilders receive instruction and supervision to ensure proper technique (Goertzen et al., 1989). However, it is common that recreational resistance training participants and gym users seldom receive formal coaching and often have no formal induction to a gym facility (Crozier et al., 2020). With the rise of 24-hour gyms, gym floor instructor role eliminations with the COVID 19 pandemic, and gym users typically working out alone or with a friend, gym users are often unsupervised and unable to interact with trained gym staff. In a 2014 survey from New Zealand, respondents stated they were actively seeking knowledge and coaching in their sporting endeavours (Sport New Zealand, 2015). It is important that gym users wishing to participate in resistance training are adequately equipped with the knowledge to perform exercises with the correct technique to prevent injury (Haupt, 2001; Mazur et al., 1993). In New Zealand, citizens and visitors can claim treatment compensation and rehabilitation costs through the Accident Compensation Corporation [ACC] following an accidental personal injury. In 2020 ACC reported an 18% increase in gym injuries in New Zealand over the previous five years (2015 to 2020) with the cost of support rising by 43% (Accident Compensation Corporation [ACC], 2020). In response, our analyses of 10 years (2011 to 2020) of ACC’s gym and fitness related injury claims, found the most common cause of injury from gym and fitness activities in New Zealand was lifting/carrying/strain resulting in lower back/spine and shoulder soft tissue injuries (Cuthbertson-Moon et al., 2024). The injury data provided by ACC were deidentified and included a free text “accident description” field which in some cases gave a short description of what the claimant was doing at the time of injury, the words were analysed by word frequency and categorised. The most frequent words relating to exercise were press, squat, bench, pull, deadlift, up, and push (Cuthbertson-Moon et al., 2024). Common resistance training exercises that result in shoulder and lower back injury include bench press, shoulder press, deadlifts, and squats (Faigenbaum & Myer, 2010; Keogh & Winwood, 2017). Resistance training injuries can occur through a variety of mechanisms, including overuse, sudden trauma, poor technique, fatigue and inappropriate loads are some examples discussed in the literature (Grier et al., 2022; Lavallee & Balam, 2010; Martins Utida et al., 2024). Whilst resistance training is inherently safe in comparison with other sports activities (Serafim et al., 2023) this is a risk that the private fitness sector has a responsibility to mitigate including appropriately qualified staff who act in accordance with their training. In order to create effective interventions, it is important that practitioners have a thorough knowledge of the incidence, severity and mechanisms of the injuries that their prevention programme is targeting (Finch, 2006; van Mechelen, 1997; van JSES | https://doi.org/10.36905/jses.2026.02.03

Mechelen et al., 1992). Coaches play an important role in implementation of injury prevention programmes therefore it is crucial gain an understanding of their knowledge, attitudes, and beliefs (Geertsema et al., 2021; McKay et al., 2014; Räisänen et al., 2021). The Knowledge, Attitudes, and Behaviour (KAB) approach is a multi-construct method used to assess and evaluate performance and learning in complex environments, recognizing that learning involves more than just cognitive development (Schrader & Lawless, 2004). KAB studies evaluate not only knowledge, but also attitudinal and behavioural change. The relationship between knowledge, attitudes, and behaviour is complex, dynamic, and potentially reciprocal. Knowledge alone is often not enough to influence behaviour change, interventions should ideally involve all three (Schrader & Lawless, 2004). In concussion research for example an understanding of knowledge, attitudes, and behaviours has proven to identify gaps that can be addresses with targeted initiatives (Theadom et al., 2020). 1.1. Aim We developed a survey to question gym staff on their observations. One aim of the survey was to gain an understanding of New Zealand gym staff (Group Fitness Instructors, Gym Managers, Personal Trainers) observations of gym users resistance training exercise technique and the link to musculoskeletal injury, using questions relating to gym user knowledge of poor technique and injury. Another aim was to understand gym staff knowledge of which resistance training exercises were perceived to contribute the most to shoulder and lower back injuries in the gym environment, and what technique faults the surveyed gym staff commonly observed during gym user practice. Attitudes and behaviours were linked to questions regarding gym staffs’ observation of gym users performing poor and unsafe technique and what the gym staff would do in this scenario. To help prepare for future gym injury types, an additional question was asked to the gym managers to describe any emerging fitness trends they observe to be causing injury amongst gym users. The purpose of gathering the information was to aid the development of an intervention study targeted at providing gym users with exercise technique coaching. 2.

Methods

Ethical consent for the research was obtained from the Auckland University of Technology Ethics Committee [AUTEC] (22/108), and informed consent was collected from each participant at the time of the survey. 2.1. Participants and recruitment A survey link was emailed to the Fitness Managers of a New Zealand gym chain by the project team (BH) and disseminated by the Gym Managers of the 12 gyms to their teams of Personal Trainers and Group Fitness Instructors. The survey was open from 1st June 2022 until 15th September 2022. Two reminder emails were sent during this period. Participation was anonymous and voluntary. 18

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) 2.2. Survey method and questionnaire The online survey developed using Qualtrics XM (Qualtrics, SAP, Seattle, WA, USA) (Supplementary Table 1) consisted of a maximum of 21 questions and took approximately 10 minutes to complete. The survey was used to assess knowledge and behaviours toward gym user resistance training exercise technique and the link to injury. The design of the questionnaire was based on those from previous studies (Murphy, 2015; O’Reilly et al., 2020; Theadom et al., 2020). Questions related to respondents’ experiences, opinions, and observations regarding poor and unsafe resistance training exercise technique and to rank a selection of exercises commonly associated with shoulder and lower back injuries. The Checklist for Reporting Results of Internet E-Surveys [CHERRIES] (Eysenbach, 2004) was used to aid survey design (Supplementary Table 2). Respondents were requested to rank order a set of exercises that they perceived to cause the most risk of shoulder and lower back injuries. Rank order data were extracted from Qualtrics in raw data format and the weighted average calculated in Microsoft Excel. In addition, respondents were asked to select exercises, from a list provided, where they typically observe techniquebased faults, and poor or unsafe technique. Respondents were also asked to state the specific faults commonly observed for the exercises selected in a free text format. Gym managers were requested to rank order specific exercises in terms of their contribution to gym user injuries, “Based on your experience, please reorder the following exercises to rank in terms of their contribution to gym user injuries (1 = most likely to result in injury, 10 = least likely to result in injury).” Gym managers were asked how they train their staff to observe and approach gym users performing unsafe technique, and what additional resources, if any, would be needed to assist with staff training. Gym managers were also asked a question relating to the emerging fitness trends that they perceive to be associated with an increased risk of injury. 2.3. Data analysis

All data were exported into a Microsoft Excel spreadsheet (Microsoft Excel for Mac 16.70, Microsoft Corporation, Redmond, WA, USA). Percentages and weighted averages were calculated using Microsoft® Excel and free text entries were analysed in NVivo® for Mac (NVivo 1.6.2, QSR International Pty Ltd, Burlington, MA, USA.) the top five words were tabulated, and word clouds generated. 3.

Results

Of 122 individuals who clicked the survey link and agreed to participate in the survey, 99 respondents either fully or partially completed the survey (81.1%), while 23 took no further action. There was a median time of 7.7 minutes to complete the survey. Results are reported for the 99 fully or partially completed responses and question specific sub-groups. 3.1. Respondent demographics Responses came from a total of 99 Group Fitness Instructors, Gym Managers and Personal Trainers, located at 12 gyms throughout New Zealand; 61% of respondents (n = 61) were based in Auckland, 16% (n = 16) were based in Wellington, the remaining 23% (n = 23) were distributed throughout Christchurch, Dunedin and Hamilton. The industry experience of respondents (i.e., time worked in the fitness industry) varied from less than one year to over 21 years, with 53% (n = 53) of respondents having industry experience between three and 10 years (Supplementary Table 3). Of the 99 respondents, 53 (53%) identified their main role as Group Fitness Instructor, 26 (26%) Personal Trainer, and 20 (20%) Gym Manager (brief role descriptions are provided in Supplementary Table 4). Of the 53 Group Fitness Instructors, 57% (n = 30) taught resistance training (e.g., Les Mills BODYPUMP®), 40% (n = 21) taught High Intensity Interval Training [HIIT] (e.g., Les Mills GRIT®), and 9% (n = 5) taught functional training (e.g., CROSSFIT®, Les Mills CEREMONY®). A Group Fitness Instructor may teach one or multiple class types (Table 1).

Table 1: Types of group exercise classes taught by respondents. Class Type

Number

Percent (%)

Cardio (e.g., Les Mills BODYCOMBAT , Les Mills BODYATTACK )

Cycling (e.g., Les Mills RPM®, Les Mills SPRINT®, Les Mills THE TRIP®)

HIIT (e.g., Les Mills GRIT® series)*

Yoga/Pilates (including Les Mills BODYBALANCE )

Dance (e.g., Les Mills BODYJAM®, Les Mills SH’BAM®, Les Mills Barre®)

Functional Training (e.g., CROSSFIT , Les Mills Ceremony )*

Other

Strength Training (e.g., Les Mills BODYPUMP®) ®

Core training (e.g., Les Mills CORE and Les Mills TONE ) ®

Step Aerobics (e.g., Les Mills BODYSTEP ) ®

Note: *Les Mills GRIT® and Les Mills Ceremony® include a significant component of resistance training. JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) The 53 Group Fitness Instructor respondents were asked a follow up question whether they coach or had coached resistance training to determine their suitability to answer the questions. If they answered “no” to this question (n = 34) the survey ended, while if they answered yes (n = 19) they survey continued. Two Group Fitness Instructors answered yes but did not complete the remainder of the survey (drop out). There were also one Gym Manager drop out and four Personal Trainers who dropped out. The 17 Group Fitness Instructors with experience in coaching resistance training, along with 19 Gym Managers, and 22 Personal Trainers, resulted in a total of 58 respondents who went on to fully or partially complete the remainder of the survey (Supplementary Figure 1). 3.2. Knowledge of musculoskeletal injuries and resistance training Cleans/clean & press, and overhead shoulder press were ranked joint first as the exercises perceived to have the greatest risk of shoulder injuries, chest press/chest fly was the third highest ranked exercise (Supplementary Table 5). Deadlift was ranked the highest risk exercise for lower back injury, followed by squats and cleans/clean & press (Supplementary Table 6). Analysis of responses to the free text fields to identify “other” exercises perceived to result in shoulder and lower back injuries revealed bench press and shoulder rotator raise to be the most frequently stated for shoulder injury, while the leg press and hip/glute bridge for lower back injury (Figure 1).

Shoulder injury “other” (9 responses)

Lower back injury “other” (10 responses)

Word

Weighted percent (wt.%)*

Word

Weighted percent (wt.%)*

press

bridges

bench

hip

machine

leg

rotators

press

side

abdominals

shoulder stands raises

machine smith biceps barbell bench raise

press rotators curl

side hand

press hip bridges leg

pulldown snatch kettlebell lat swings ups

sit

planks glutes

pulls abdominals machine

Figure 1: Word frequency of the top five words and word cloud for all or free text descriptive entries for “other” exercises that were perceived to put people at risk of shoulder and lower back injury. Note: *percentages do not add up to 100% because only the top five words are presented in the table. JSES | https://doi.org/10.36905/jses.2026.02.03

3.3. Gym user technique of specific exercises Of the respondents, 54 answered the question “what exercises (including free weights and resistance machines) do you typically observe technique-based faults or unsafe technique” Deadlift (91%, n = 49), squat (89%, n = 48), push up (76%, n = 41), cleans/clean & press (72%, n = 39), and chest press (63%, n = 34) were the most frequently selected exercises by respondents observed to be performed with poor or unsafe technique (Figure 2). Common answers for technique faults associated with the five most selected exercises were as follows; deadlift; rounding of the back, squats; knee alignment and forward lean, push up; hand alignment and core control, cleans/clean & press; cleaning the bar away from the body, and chest press; bar above the shoulders. “Other” exercises performed with poor or unsafe technique were selected 13% (n = 7) of the time and named by respondents in the free text field, were kettlebell swing (n = 2), box jumps (n = 2), and lunges (n = 2) (Table 2).

Table 2: Summary of typical technique faults observed by the gym staff for the named exercises. Exercise

Common technique faults observed

Deadlift

Arched back, rounded back, posture.

Squat

Knee alignment, knees caving in, chest tipping forward, performing a ‘good morning’ exercise rather than a ‘squat’.

Push up

Core activation, hand/elbow position.

Cleans/clean & press

Reverse curling the bar, cleaning away from the body.

Chest press

Bar above shoulders/head, grip.

Bent over row

Rounded back.

Overhead shoulder press Overextending, leaning back. Lat pull down

Behind head, rocking.

Lateral raise

Swinging /shrugging the weight.

Biceps curl

Swinging the body.

Chest fly

Range.

Triceps dip

Shoulder positioning.

Upright row

Hand/wrist position, rounded shoulders.

Triceps extension

Elbow flare, rounded shoulders.

Pull up/chin up

Scapular control.

Other

Lunges, kettlebell swings, box jumps. 20

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026)

Figure 2: Percentages (%) of respondents that selected exercises for the question “What exercises (including free weights and resistance machines) do you typically observe technique-based faults or unsafe technique?” (Total respondents n = 54).

JSES | https://doi.org/10.36905/jses.2026.02.03

training

activation

exercise

fast

3 unconditioned complimentary bootcamp

group

explosive band squats

weight

crossfit exercise

bench compound lifts posture

work poor

form intensity style incorrect

wods grit

unaware

CrossFit

training injuries

Emerging fitness trends observed by Gym Manager respondents were coded and analysed using thematic analysis (Braun & Clarke, 2006, 2012). The most frequent words were “CrossFit”, ‘training”, “activation”, “exercise”, and “fast” (Figure 3 provides weighted percentages and a word cloud). Ten of the 16 responses (63%) indicated workouts that incorporated high complexity or highvolume training, including CROSSFIT®, and HIIT. Other comments can be grouped thematically into type of workout, “WODs”, “bootcamp”, “plyometric”; style of training “explosive”, “compound”, “volume”; equipment “Olympic”, “bands”; movement or movement pattern “handstands”, “pressing”, “squats”, “deads”; intensity “tempo”, “max”; body part “shoulder”, “glute”, “core”.

Weighted percent (wt.%)*

glute trend muscle deads hiit activation

3.4. Emerging fitness trends resulting in injury

Word

workouts novice ratio plyometric handstands instagram complex clean core

For the 16 Gym Managers who answered this question, deadlift was the highest ranked exercise most likely perceived to result in injury, followed by squats, cleans/clean & press, chest press, and overhead shoulder press (Supplementary Table 7). Regarding staff training, a variety of responses were obtained from Gym Managers. The most common theme amongst Gym Manager responses was gym staff education through training and upskilling, and effective communication with gym users. A representative selection of responses is summarised in Supplementary Table 7. When asked what additional information Gym Managers needed to train and assist gym staff in observing and intervening to prevent gym user injury, training resources was the most emphasised response (Supplementary Table 8).

lifters heart tempo

fast fitness burn classes movements engagement max volume bodypump

instructor complicated shoulder pressing loading olympic

Figure 3: Word frequency of the top five words and word cloud of most frequent words in free text form for the question “Are there any emerging fitness trends that are resulting in injuries?” Note: *percentages do not add up to 100% because only the top 5 words are presented in the table. 21

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) 3.5. Attitudes towards musculoskeletal injuries and resistance training The survey asked four questions related to observations of technique within the gym (Table 3). All 53 respondents stated they had observed poor technique in the gym and 96% (n = 52) stated they had observed poor technique from gym users moderately often (more than 40% of the time), 57% (n = 31) observed poor technique often (more than 60% of the time), and 30% (n = 16) observed poor technique very often (more than 80%

of the time). All 53 respondents perceived that gym users are rarely or only sometimes aware they are performing poor technique, and no respondents reporting that they perceive gym users to be often or very often aware of their poor technique. Only 36% (n = 19) believed that gym users are aware of the link between poor technique and injury. When questioned about how often they were sought to help with performing exercises by gym users, 9% (n = 5) of gym staff respondents stated they have never been asked for help, 85% (n = 45) reported sometimes being asked for help, and only 6% (n = 3) were often asked for help.

Table 3: Survey questions relating to technique. How often do you see poor technique? (% of time)

Percent of respondents (%)

Number of respondents (n = 54)

Never (0%)

Very Rarely (0-20%)

Rarely (20-40%)

Moderate (40-60%)

Often (60-80%)

Very Often (80%+)

Rarely

Sometimes

Often

Very often

Yes

Never

Sometimes

Often

Very often

How often do you think gym users are aware they have poor technique?

Do you think gym users link poor technique to injury?

Do gym users ask for help with performing exercises?

Gym Staff were asked about their confidence to approach someone performing poor technique who was at risk of injury. Of the 53 respondents who answered this question 47% (n = 25) felt confident to approach someone who is working out with poor technique and at risk of injury. Fifty-three percent (n = 28) of the respondents either did not feel confident, or sometimes felt confident to do so (Table 4). Of the 28 respondents that did not feel confident or sometimes felt confident to approach someone working out with poor technique at risk of injury, 13 were Group JSES | https://doi.org/10.36905/jses.2026.02.03

Fitness Instructors, eight were Personal Trainers, and seven were Gym Managers. Moreover, 53% (n = 28) of gym staff respondents did not feel confident all or some of the time to approach someone who is exercising with poor technique and at risk of injury. They were asked the follow up question, “what additional education or training would be useful to enable you to feel confident to approach someone performing poor technique?” Seven of 28 respondents (25%) felt that approaching a gym user may not be well received. 22

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Table 4: Confidence to approach gym users at risk of injury. Do you feel confident to approach someone working out with poor technique at risk of injury?

Percentage of respondents (%)

Number of respondents (n = 53)

Sometimes

Yes

Nine respondents (32%) stated they need specific training or tools to feel confident to approach gym users, and/or felt that having a culture where any member of staff may approach a gym user for safety reasons would improve their confidence. Four gym staff respondents (14%) stated they felt unqualified or needed more personal knowledge to be able to feel confident to approach someone on the gym floor performing poor technique at risk of injury. Four gym staff respondents (14%) stated that it was not their job, or they did not have the time, three of these respondents identified their role as Group Fitness Instructor. 3.6. Behaviours towards musculoskeletal injuries and resistance training Gym staff respondents (91%, n = 53) answered free text behaviour related questions regarding approaching gym users performing poor or unsafe technique. When observing a gym user performing poor technique 62% (n = 33) of respondents who answered these questions stated they would approach them, 17 of the respondents were Personal Trainers, eight were Gym Managers, and eight were Group Fitness Instructors. When observing unsafe technique 91% (n = 48) would approach, 23 of the respondents were Personal Trainers, 15 Gym Managers, and 10 Group Fitness Instructors. When asked what recommendations they would provide to someone exercising with unsafe technique and at risk of injury, 94% (n = 50) of gym staff responded. Fiftyeight percent (n = 29) of respondents stated they would approach to offer advice about correct form regressions, or alternative exercises, 12 of these respondents were Personal Trainers, 10 were Gym Managers, and seven were Group Fitness Instructors. Thirty-six percent (n = 18) stated they would approach to recommend a personal trainer, of these respondents, eight were Gym Managers, six were Personal Trainers, and four were Group Fitness Instructors. 4.

Discussion

Whilst poor technique is not a sole trigger for injury, the recommendation to learn proper technique (Hume & Marshall, 1994) and to be supervised during resistance training is well established to prevent injuries (Jones et al., 2000; Risser, 1990). This study highlighted resistance training exercises that are commonly performed with poor and unsafe technique, and those that are likely to cause shoulder and lower back injuries as perceived by gym staff, these results are to be intended to add to our findings from the ACC injury analysis (Cuthbertson-Moon et al., 2024). JSES | https://doi.org/10.36905/jses.2026.02.03

When asked to think about poor and unsafe technique that can lead to resistance training injury, only 36% of all gym staff surveyed perceived that gym users actively link poor technique to injury. Gym staff respondents also identify that gym users are rarely or only sometimes aware that they have poor technique. The gym staff questioned revealed that they are only sometimes asked for help when performing an exercise. A brief evaluation of the literature uncovered some potential barriers that gym users may have when asking for help, the fear of being judged, weight stigma (Auster-Gussman et al., 2021), cultural norms in a highly male dominated environment can make particular gym users feel uncomfortable, often causing them avoiding certain areas and times of the day (Cowley & Schneider, 2025). Some gym users report they know they are doing something that might cause injury but carry on regardless (Pelters, 2024). Perceived obstacles can often result in gym users performing risky behaviours, unsafe exercise modification, unsupervised exercise, and a reluctance to ask for help. Moving on to the point that gym staff do not always feel confident to approach someone exercising with poor or unsafe technique in the gymnasium, an interesting observation within this study was that respondents from all three groups felt this way. Common responses as to why this might be uncovered beliefs that this type of feedback might not be well received by the gym user, that it was not within their role, and that they would benefit from further training on how to approach gym users. Perceived barriers to teachers, coaches and instructors providing feedback reviewed in the literature included: workload (Paris, 2022), concern about negatively impacting the relationship; lack of knowledge or training to give feedback (Karimi Moonaghi et al., 2016) and how feedback was received (Mason et al., 2020). Fostering a culture where any staff member, may intervene for safety reasons could benefit the wider gym floor community, along with the provision of training and tools to aid observation, facilitate conversations, and offer effective feedback (Hattie & Timperley, 2007; Jug et al., 2018; Postal, 2019). Gym staff identified deadlifts, squats, push ups, cleans/clean & press, and chest press as the exercises where they observe technique faults most frequently. The same exercises were also perceived to be the most likely to cause lower back or shoulder injuries. Improper exercise technique and unfavourable shoulder positioning in exercises like bench (chest) press and overhead pressing with repetition until failure increase the likelihood of shoulder injury (Kolber et al., 2010). The exercises highlighted in this survey are consistent with those from previous studies. For example, our retrospective epidemiological analysis of injury claims revealed injuries occurring through squats, deadlifts, and bench press to be some of the most frequent injuries amongst gym injuries in New Zealand (Cuthbertson-Moon et al., 2024), while a systematic review focused on injuries occurring in the weight training sports where squats, cleans, and deadlifts are performed (Keogh & Winwood, 2017). Gym Managers gave rich and varied responses about staff training, stating that they train gym staff to observe unsafe exercise technique, and how to approach gym users and communicate with them effectively. Gym staff disclosed that they do not always have the confidence and knowledge to do so. The inclusion of Group Fitness Instructors and Personal Trainers in training sessions on observation of poor and unsafe technique and communication with gym users might be valuable to enable them 23

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) to have the knowledge and confidence to approach gym users. This could be delivered during staff onboarding or as part of continuing professional development. Future research might address the content and frequency of training needs to ensure that all gym staff and contractors have adequate knowledge and confidence to approach gym users who are exercising with poor technique and intervene where they are risk of injury. Gym Managers ranked deadlifts, squats, and cleans/clean & press, as the most likely exercises to result in injury revealing an opportunity to train gym staff to focus on coaching technique in these specific exercises during gym user inductions and personal training sessions, and to run workshops in clubs for novice gym users to gain knowledge of proper technique. All three groups of gym staff (Group Fitness Instructors, Gym Managers, Personal Trainers) consistently observed the same group of exercises to be performed with poor technique and to be at the highest risk of causing injury, giving weight to the design and implementation of an injury prevention strategy focussing on improving education and technique of deadlifts, squats, cleans/clean & press, push ups, and chest press. There is an opportunity to bridge the knowledge gap of gym users between poor technique and injury, through facilitating dialogue between gym staff and gym users, and fostering a culture where all gym staff may approach gym users for wellbeing and safety reasons. Gym inductions and coaching novice gym users to learn proper exercise technique during the early stages of resistance training is important, and providing the necessary training and tools to both gym users and gym staff may help to address this. Empowering gym users to ask for help to perform exercises when they need it, and empowering all gym staff to be confident to provide that help when asked or when they observe an opportunity for intervention.

of deadlift, squat, chest press, clean/clean & press and push ups for gym users. The key findings were: •

•

4.1. Limitations The survey was limited to staff from one gym chain, with potential sampling bias. Since the survey was distributed voluntarily by managers, and completed anonymously by respondents, it is not possible to know who or how many people received the survey or to calculate the overall response rate which may result in some participation or non-response bias. The exercises were provided as a list that included common resistance exercises that are performed in a gym, and those highlighted from the ACC gym injury analysis (Cuthbertson-Moon et al., 2024) the list was not extensive however the survey offered the opportunity for respondents to add exercises in an “other” free text entry field. Nonetheless, the study provides valuable insight into staff perceptions that complement injury data and inform intervention design. 4.2. Benefits The value the current study brings is the focus on observations and perceptions of gym staff towards resistance training technique amongst gym users, giving a fresh perspective on resistance training exercise technique and injury prevention research. The perspectives of gym staff add to our ACC injury epidemiology study to enable us to appropriately design an intervention study aimed at reducing injury risk by focussing on technique coaching JSES | https://doi.org/10.36905/jses.2026.02.03

•

Common injury-prone exercises identified: Gym staff perceive deadlifts, squats, cleans/clean & press, push ups, and chest press as exercises frequently performed with poor technique and most likely to cause shoulder and lower back injuries. Prevalence of poor technique observed: Ninety-six percent of respondents reported observing poor technique more than 40% of the time, with common faults including rounding of the back during deadlifts and knee misalignment during squats. Gym users' awareness and help-seeking behaviour: Staff perceive that gym users are rarely or only sometimes aware of their poor technique, with 64% believing users do not understand the injury risk. Furthermore, 85% of staff reported being only sometimes asked for help. Staff confidence and intervention behaviours: While 62% of staff would approach someone performing poor technique, only 47% felt confident doing so. Barriers to intervention include fear of negative reception and feeling unqualified. However, 91% would intervene when unsafe technique is observed. Training and upskilling needs: Gym managers emphasize the importance of staff education, communication skills, and specific training resources to improve observation and intervention capabilities among staff. Emerging fitness trends for injury: High complexity and high-volume training styles such as CrossFit ® and HIIT are perceived as emerging trends contributing to injuries, particularly involving explosive and compound movements. Opportunities for injury prevention: There is scope for gym managers to enhance staff knowledge and confidence through targeted training, focusing on coaching proper technique in key exercises during inductions and workshops for novice gym users.

Conclusions Understanding gym staff perceptions about resistance training technique is crucial to preventing injuries among gym users. This study surveyed New Zealand gym staff to assess their knowledge, attitudes, and behaviours regarding exercise technique, and its link to musculoskeletal injuries. The knowledge of gym staff toward resistance training exercise technique was consistent amongst the surveyed groups of Group Fitness Instructors, Gym Managers, and Personal Trainers regarding the exercises that they consider to be performed with poor or unsafe technique and those likely to result in shoulder and lower back injuries, and the technique faults typically observed in gym users. Gym staff attitudes indicated that they perceive gym users to only sometimes have an awareness that they have poor exercise technique, and that poor technique can lead to injury. Gym staff are sometimes asked for help by gym users whilst performing exercises. Generally, gym-staff behaviours showed that if they observe someone exercising unsafely, they are likely to intervene and offer 24

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) advice or recommend a personal trainer. However, some gym staff self-reported needing more knowledge and empowerment to approach and intervene with gym users performing exercises with poor technique and at risk of injury. Deadlifts, squats, cleans/clean & press, push ups, and chest press were perceived by gym staff to be the exercises most frequently performed poorly and those most likely to result in lower back and shoulder injuries amongst recreational gym users and group fitness participants. There is an opportunity for Gym Managers to train their Group Fitness Instructors and Personal Trainers to focus on coaching technique of these specific exercises during gym user inductions and personal training sessions, and to schedule periodic workshops for novice gym users to improve their knowledge of resistance training and how to perform the exercises with proper technique. Disclaimer Les Mills New Zealand Personal Trainer and Group Fitness Instructor is contractor role and is not employed by the Les Mills organisation, some contractors may have additional employment roles at Les Mills New Zealand or Les Mills International. Gym Managers are employed by Les Mills New Zealand. The views expressed in this manuscript are the respondents own and do not represent those of Les Mills New Zealand or Les Mills International. Contributors statement According to the definition given by the International Committee of Medical Journal Editors (ICMJE), the authors listed qualify for authorship based on making one or more of the substantial contributions to the intellectual content of the manuscript: Conceptualization PH, MCM, HW, BH; Ethics PH, MCM, HW; Methodology MCM, PH; Data collection MCM; Data analysis MCM; Data interpretation MCM, PH, HW, BH; Writing— original draft preparation MCM; Writing—review and editing MCM, PH, HW, BH; Project administration MCM, PH. All authors have read and agreed to the published version of the manuscript. Conflict of Interest MCM is a Group Fitness Instructor contractor at Les Mills New Zealand, MCM started contracting for Les Mills New Zealand after the survey was distributed. BH is a contractor for Les Mills International. There was no financial compensation for conducting this research. There are no conflicts of interest for the other co-authors. Acknowledgments We thank Ish Cheyne, Tauvaga Siolo, and Geena Pannett, for their support to distribute the survey and for providing additional information. The Group Fitness Instructors, Personal Trainers, and Gym Managers for their engagement in the survey. Isaac Carlson for helpful discussions during the design of the survey. Andrew Moon for helpful discussion during the analysis of the JSES | https://doi.org/10.36905/jses.2026.02.03

survey, and for proofreading and providing feedback on the manuscript. References Accident Compensation Corporation [ACC]. (2020, February, 3). Spike in minor gym injuries could be a good thing. https://www.acc.co.nz/newsroom/stories/spike-in-minorgym-injuries-could-be-a-good-thing/ Auster-Gussman, L. A., Crim, J., & Mann, T. L. (2021). The soulless cycle: Social physique anxiety as a mediator of the relation between body mass index and exercise frequency. Stigma Health, 6(2), 192–199. https://doi.org/10.1037/sah0000291 Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2),77–101. https://doi.org/10.1191/1478088706qp063oa Braun, V., & Clarke, V. (2012). Thematic analysis. American Psychological Association. Cowley, E. S., & Schneider, J. (2025). "I sometimes feel like I can't win!": An exploratory mixed-methods study of women's body image and experiences of exercising in gym settings. PLoS ONE, 20(1), e0316756. https://doi.org/10.1371/journal.pone.0316756 Crozier, A., Porcellato, L., Buckley, B. J. R., & Watson, P. M. (2020). Facilitators and challenges in delivering a peersupport physical activity intervention for older adults: A qualitative study with multiple stakeholders. BMC Public Health, 20(1), 1904. https://doi.org/10.1186/s12889-02009990-x Cuthbertson-Moon, M., Hume, P. A., Wyatt, H. E., Carlson, I., & Hastings, B. (2024). Gym and fitness injuries amongst those aged 16–64 in New Zealand: Analysis of ten years of Accident Compensation Corporation injury claim data. Sports Medicine - Open, 10(1), 53. https://doi.org/10.1186/s40798-024-006949 Dhuli, K., Naureen, Z., Medori, M. C., Fioretti, F., Caruso, P., Perrone, M. A., Nodari, S., Manganotti, P., Xhufi, S., Bushati, M., Bozo, D., Connelly, S. T., Herbst, K. L., & Bertelli, M. (2022). Physical activity for health. Journal of Preventive Medicine and Hygiene, 63(2 Suppl 3), E150–E159. https://doi.org/10.15167/2421-4248/jpmh2022.63.2S3.2756 Eysenbach, G. (2004). Improving the quality of Web surveys: The Checklist for Reporting Results of Internet E-Surveys (CHERRIES). Journal of Medical Internet Research, 6(3), e34. https://doi.org/10.2196/jmir.6.3.e34 Faigenbaum, A. D., & Myer, G. D. (2010). Resistance training among young athletes: Safety, efficacy and injury prevention effects. British Journal of Sports Medicine, 44(1), 56–63. https://doi.org/10.1136/bjsm.2009.068098 Finch, C. (2006, May). A new framework for research leading to sports injury prevention. Journal of science and medicine in sport, 9(1-2), 3–9. https://doi.org/10.1016/j.jsams.2006.02.009 Geertsema, C., Geertsema, L., Farooq, A., Harøy, J., Oester, C., Weber, A., & Bahr, R. (2021). Injury prevention knowledge, beliefs and strategies in elite female footballers at the FIFA Women's World Cup France 2019. British Journal of Sports 25

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Medicine, 55(14), 801–806. https://doi.org/10.1136/bjsports2020-103131 Goertzen, M., Schöppe, K., Lange, G., & Schulitz, K. P. (1989). Injuries and damage caused by excess stress in body building and power lifting. Sportverletz Sportschaden, 3(1), 32–36. https://doi.org/10.1055/s-2007-993630 Gómez-Cabello, A., Ara, I., González-Agüero, A., Casajús, J. A., & Vicente-Rodríguez, G. (2012). Effects of training on bone mass in older adults. Sports Medicine, 42(4), 301–325. https://doi.org/10.2165/11597670-000000000-00000 Grier, T., Brooks, R. D., Solomon, Z., & Jones, B. H. (2022). Injury risk factors associated with weight training. The Journal of Strength & Conditioning Research, 36(2), e24–e30. https://doi.org/10.1519/jsc.0000000000003791 Harne, A. J., & Bixby, W. R. (2005). The benefits of and barriers to strength training among college-age women. Journal of Sport Behavior, 28(2). Hattie, J., & Timperley, H. (2007). The power of feedback. Review of Educational Research, 77(1), 81–112. https://doi.org/10.3102/003465430298487 Haupt, H. A. (2001). Upper extremity injuries associated with strength training. Clinics in Sports Medicine, 20(3), 481–490. https://doi.org/https://doi.org/10.1016/S02785919(05)70264-7 Hume, P., & Marshall, S. (1994). Sports injuries in New Zealand: Exploratory analyses. New Zealand Journal of Sports Medicine, 22, 18–22. Hurley, K. S., Flippin, K. J., Blom, L. C., Bolin, J. E., Hoover, D. L., & Judge, L. W. (2018). Practices, perceived benefits, and barriers to resistance training among women enrolled in college. International Journal of Exercise Science, 11(5), 226–238. Jones, C. S., Christensen, C., & Young, M. (2000). Weight training injury trends: A 20-year survey. The Physician and Sportsmedicine, 28(7), 61–72. https://doi.org/10.3810/psm.2000.07.1086 Jug, R., Jiang, X. S., & Bean, S. M. (2018). Giving and receiving effective feedback: A review article and how-to guide. Archives of Pathology & Laboratory Medicine, 143(2), 244– 250. https://doi.org/10.5858/arpa.2018-0058-RA Karimi Moonaghi, H., Vafaee, M., Ildarabadi, E.-h., & Rice, S. (2016). Experiences of medical teachers about methods, types, and barriers of giving feedback. Future of Medical Education Journal, 6(3), 26–31. Keogh, J. W. L., & Winwood, P. W. (2017). The epidemiology of injuries across the weight-training sports. Sports Medicine, 47(3), 479–501. https://doi.org/10.1007/s40279-016-0575-0 Kolber, M. J., Beekhuizen, K. S., Cheng, M.-S. S., & Hellman, M. A. (2010). Shoulder injuries attributed to resistance training: A brief review. The Journal of Strength & Conditioning Research, 24(6), 1696–1704. https://doi.org/10.1519/JSC.0b013e3181dc4330 Lavallee, M. E., & Balam, T. (2010). An averview of strength training injuries: Acute and chronic. Current Sports Medicine Reports, 9(5), 307–313. https://doi.org/10.1249/JSR.0b013e3181f3ed6d Martins Utida, K. A., Cruz Carvalho Gomes, D., de Souza Ajala, F., Felippe Martinez, P., & Assis de Oliveira-Junior, S. (2024). Sports injury prevalence and associated factors in functional JSES | https://doi.org/10.36905/jses.2026.02.03

fitness: A cross-sectional study. Human Movement, 25(3), 62– 71. https://doi.org/10.5114/hm/190500 Mason, R. J., Farrow, D., & Hattie, J. A. C. (2020). Sports coaches’ knowledge and beliefs about the provision, reception, and evaluation of verbal feedback. Frontiers in Psychology, 11, 571552. https://doi.org/10.3389/fpsyg.2020.571552 Mazur, L. J., Yetman, R. J., & Risser, W. L. (1993). Weighttraining injuries - Common injuries and preventative methods. Sports Medicine, 16(1), 57–63. https://doi.org/10.2165/00007256-199316010-00005 McKay, C. D., Steffen, K., Romiti, M., Finch, C. F., & Emery, C. A. (2014). The effect of coach and player injury knowledge, attitudes and beliefs on adherence to the FIFA 11+ programme in female youth soccer. British Journal of Sports Medicine, 48(17), 1281–1286. Murphy, K. J. (2015). What do secondary school rugby players think about concussion? [Masters, University of Waikato]. Hamilton, New Zealand. https://hdl.handle.net/10289/10099 Myers, R. S., & Roth, D. L. (1997). Perceived benefits of and barriers to exercise and stage of exercise adoption in young adults. Health Psychology,, 16(3), 277–283. https://doi.org/10.1037//0278-6133.16.3.277 Nelson, M. E., Rejeski, W. J., Blair, S. N., Duncan, P. W., Judge, J. O., King, A. C., Macera, C. A., & Castaneda-Sceppa, C. (2007). Physical activity and public health in older adults: Recommendation from the American College of Sports Medicine and the American Heart Association. Circulation, 116(9), 1094. https://doi.org/https://doi.org/10.1161/CIRCULATIONAHA. 107.185650 O’Reilly, M., Mahon, S., Reid, D., Hume, P., Hardaker, N., & Theadom, A. (2020). Knowledge, attitudes, and behavior toward concussion in adult cyclists. Brain Injury, 34(9), 1175– 1182. https://doi.org/10.1080/02699052.2020.1793386 Paris, B. (2022). Instructors’ perspectives of challenges and barriers to providing effective feedback. Teaching and Learning Inquiry, 10. https://doi.org/10.20343/teachlearninqu.10.3 Pelters, P. (2024). The good, the bad and the ugly – a Swedish qualitative interview study about the landscape of meaningimbued, exercise-related physical pain, as experienced by ‘normal’ gym-users. BMC Public Health, 24(1), 1167. https://doi.org/10.1186/s12889-024-18623-6 Postal, K. (2019). Effective feedback conversations. In K. M. Sanders (Ed.), Physician's Field Guide to Neuropsychology: Collaboration through Case Example (pp. 575–586). Springer New York. https://doi.org/10.1007/978-1-4939-8722-1_34 Räisänen, A. M., Owoeye, O. B. A., Befus, K., van den Berg, C., Pasanen, K., & Emery, C. A. (2021). Warm-ups and coaches' perceptions: Searching for clues to improve injury prevention in youth basketball. Frontiers in Sports and Active Living, 3, 619291. https://doi.org/10.3389/fspor.2021.619291 Risser, W. L. (1990). Musculoskeletal injuries: Caused by weight training guidelines for prevention. Clinical pediatrics, 29(6), 305–310. https://doi.org/10.1177/000992289002900602 Schrader, P. G., & Lawless, K. (2004). The knowledge, attitudes, & behaviors approach how to evaluate performance and learning in complex environments. Performance Improvement, 43, 8–15. https://doi.org/10.1002/pfi.4140430905 26

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Serafim, T. T., de Oliveira, E. S., Maffulli, N., Migliorini, F., & Okubo, R. (2023). Which resistance training is safest to practice? A systematic review. Journal of Orthopaedic Surgery and Research, 18(1), 296. https://doi.org/10.1186/s13018-023-03781-x Sport New Zealand. (2015). Sport and active recreation in the lives of New Zealand adults. 2013/14 Active New Zealand Survey Results. https://sportnz.org.nz/media/1542/sport-andactive-recreation-in-the-lives-of-new-zealand-adults.pdf Theadom, A., Reid, D., Hardaker, N., Lough, J., & Hume, P. A. (2020). Concussion knowledge, attitudes and behaviour in equestrian athletes. Journal of Science and Medicine in Sport, 23(11), 1055–1061. van Mechelen, W. (1997). Sports injury surveillance systems. Sports Medicine, 24(3), 164–168. https://doi.org/10.2165/00007256-199724030-00003

JSES | https://doi.org/10.36905/jses.2026.02.03

van Mechelen, W., Hlobil, H., & Kemper, H. C. G. (1992). Incidence, severity, aetiology and prevention of aports injuries. Sports Medicine, 14(2), 82–99. https://doi.org/10.2165/00007256-199214020-00002 Vasudevan, A., & Ford, E. (2022). Motivational factors and barriers towards initiating and maintaining strength training in women: A systematic review and meta-synthesis. Prevention Science, 23(4), 674-695. https://doi.org/10.1007/s11121-02101328-2 Warburton, D. E., Nicol, C. W., & Bredin, S. S. (2006). Health benefits of physical activity: The evidence. Canadian Medical Association Journal, 174(6), 801–809. https://doi.org/10.1503/cmaj.051351

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Appendix Supplementary Table 1: Survey questions. Q1. Thank you for agreeing to take part in this survey. Your views are important to our project and will be kept in strict confidence. (Consent provided at this question – details deleted) Demographics Q2. Please select the gym at which you work (if you work at more than one location, please select the main gym at which you work). 1. Auckland City 2. Britomart 3. Christchurch 4. Dunedin 5. Hamilton 6. Howick 7. Hutt City 8. Lambton Quay 9. New Lynn 10. Newmarket 11. Takapuna 12. Taranaki Street Q3. How long have you worked in the fitness industry in one capacity or another? 1. Up to 1 year 2. 1-2 years 3. 3-5 years 4. 6-10 years 5. 11-20 years 6. 21 years and over Q4. Which of the following best describes your current role? If you have more than one role, please select the one that you spend the most time. 1. Group exercise instructor 2. Personal trainer 3. Gym manager If Q4 - answer 1 (group exercise instructor). Q4a. You selected that one of your roles is a group exercise instructor. What group exercise classes do you coach? (Please select all that apply). 1. Strength Training (e.g. BODYPUMP) 2. Functional Training (e.g. Crossfit, Ceremony) 3. HIIT (e.g. GRIT series) 4. Core training (e.g. Les Mills CORE and Les Mills TONE) 5. Cardio (e.g. BODYCOMBAT, BODYATTACK) 6. Cycling (e.g. RPM, SPRINT, THE TRIP) 7. Dance (e.g. BODYJAM, SHBAM, Barre) 8. Step Aerobics (e.g. BODYSTEP) 9. Yoga/Pilates (including BODYBALANCE) 10. Other (please describe) (if Q4 1. group exercise instructor only selection, and Q4a only 3-9 selected) Q4b. Have you ever coached strength based, weight training, resistance training, functional training classes or worked as a personal trainer / gym instructor prior to your current role? 1. Yes 2. No (If no – end screen and thank for participation). JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Personal trainer/Group exercise instructor/Gym manager This part of the survey focusses on resistance training technique and injury. Please use any experience (past and present) in coaching to answer these questions. Attitudes Q5. Thinking about both the gym floor and group exercise classes, how often do you see a participant (either supervised or unsupervised) performing poor technique? 1. Never (0%) 2. Very Rarely (0-20%) 3. Rarely (20-40%) 4. Moderate 40-60% 5. Often (60-80%) 6. Very Often (80%+) Knowledge Q6. What exercises most commonly put people at risk of SHOULDER injury? (Please reorder the following exercises, 1 = most likely to result in injury, 7 = least likely to result in injury) 1. Chest press/chest fly 2. Lat pull down 3. Cleans/clean & press 4. Overhead shoulder press 5. Upright row 6. Triceps dip 7. Other (please name) Knowledge Q7. What exercises most commonly put people at risk of LOWER BACK injury? (Please reorder the following exercises, 1 = most likely to result in injury, 5 = least likely to result in injury) 1. Deadlift 2. Squats 3. Bent over row 4. Cleans/clean & press 5. Other (please name) Knowledge Q8. What exercises (including free weights and resistance machines) do you typically observe technique-based faults or unsafe technique? For each selection, please indicate any specific faults or unsafe technique you observe. 1. Deadlift 2. Squats 3. Chest press 4. Chest Fly 5. Lateral raise 6. Lat Pull down 7. Overhead shoulder press 8. Bicep Curl (BB/DB) 9. Triceps extension (BB/DB) 10. Triceps dip 11. Bent over row 12. Upright row 13. I do not experience poor technique Behaviours Q9. When walking through the gym and you see someone exercising with POOR technique what do you do? (free text) Behaviours Q10. When walking through the gym and you see someone exercising with UNSAFE technique what do you do? (free text) JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Behaviours Q11. What are the recommendations that you would give to a participant that is exercising unsafely and in risk of injury? (free text) Attitudes Q12. If you see someone in the gym performing an exercise with poor technique and is at risk of injury, do you feel confident to approach them? 1. Yes 2. No 3. Sometimes If yes/sometimes selected display question 12a Attitudes Q12a. You indicated that you do not always feel confident to approach someone who you see in the gym performing poor technique. What additional education or training would be useful to enable you to feel confident to approach someone performing poor technique? (free text) Attitudes Q13. How often do you think gym participants are aware they have poor technique? 1. Rarely 2. Sometimes 3. Often 4. Very often Attitudes Q14. Do you think participants link poor technique to injury? 1. Yes 2. No Attitudes Q15. Do participants ask for help with performing exercises? 1. Never 2. Sometimes 3. Often 4. Very often Gym manager additional questions Knowledge Q16. How do you train the staff to observe and approach gym users and class participants who are exercising with unsafe technique? (Please specify – free text) 1. Exercise specific 2. Injury prevention 3. Other Knowledge Q17. What additional information or resources, if any, do you need to train and assist staff to observe and intervene to prevent injury? (free text) Knowledge Q18. Based on your experience, please reorder the following exercises to rank in terms of their contribution to gym user injuries. (1 = most likely to result in injury, 16 = least likely to result in injury). 1. Deadlift 2. Squat JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) 3. Chest press 4. Chest fly 5. Lateral raise 6. Lat pull down 7. Cleans/clean & press 8. Push ups 9. Overhead shoulder press 10. Biceps Curl (DB/BB) 11. Triceps extension (DB/BB) 12. Triceps dip 13. Bent over row 14. Upright row 15. Pull ups/chin ups 16. Other (please name) Knowledge Q19. Are there any emerging fitness trends that are resulting in injuries? (free text)

Note: The coding for knowledge, attitudes and behaviours (KAB) and observation (O) did not appear for the survey respondents but is provided for the reader to show the analysis by KABO.

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Supplementary Table 2: Checklist for Reporting Results of Internet E-Surveys (CHERRIES) (Eysenbach, 2004). Item Category

Checklist Item

Explanation

Describe survey design

The target population was Group Fitness Instructors, Personal Trainers and Gym Managers at Les Mills New Zealand.

Design

IRB (Institutional Review Board) approval and informed consent process Ethics Approval

The study was approved by Auckland University of Technology Ethics Committee (AUTEC) (#22/108).

Informed consent

Informed consent was obtained prior to commencement of the survey. Participant information sheet and informed consent described in supplementary table 1.

Data protection

The survey was anonymous.

Development and pre-testing Development and testing

The survey questions were designed by the project team. Functionality of the electronic questionnaire was tested by the project team and third year students at AUT SPRINZ.

Recruitment process and description of the sample having access to the questionnaire Open survey versus closed survey

The survey was an open survey where anyone with a direct link was able to participate.

Contact mode

Initial contact of participants was made by BH sending the survey advertisement email to the National Fitness Manager of Les Mills New Zealand. This email was sent to the managers at the 12 Les Mills New Zealand gyms for distribution to their teams of Personal Trainers and Group Fitness Instructors. A reminder was sent

Advertising the survey

The survey was not advertised, the survey was sent directly to staff at Les Mills New Zealand.

Survey administration Web/E-mail

The survey was an electronic survey hosted by Qualtrics XM (Qualtrics, SAP, Seattle, WA, USA). The link to complete the survey was sent to potential respondents by email.

Context

Not applicable.

Mandatory/voluntary

The survey was voluntary. Participants chose whether or not to click the link on the email they received from their manager.

Incentives

No incentives were offered.

Time/Date

Responses were collected between 1st June 2022 until 15th September 2022.

Randomization of items or questionnaires

No randomisation was used.

Adaptive questioning

Adaptive questioning (logic) was used, certain questions were displayed based on previous responses, (e. g. If yes/sometimes selected display question 12a) Supplementary Table 2 (continued)

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Item Category

Checklist Item

Explanation

Number of Items

The number of questions varied between one and three per page. The survey comprised of 21 questions although due to the adaptive nature of the survey not all questions were displayed.

Number of screens (pages)

The survey consisted of a maximum of 16 pages.

Completeness check

All survey questions were deemed to be mandatory, and respondents prompted to complete outstanding items before leaving the survey page (clicking next) on which the item was contained.

Review step

Respondents were unable to change their responses due to the logic used to screen participants and display the relevant questions to their roles. Answers could be changed prior to clicking the next button to move to the next section. Respondents were unable to change their responses once submitted.

Unique site visitor

Not applicable, we were unable to determine this.

View rate (Ratio of unique survey visitors/unique site visitors)

Not applicable, respondents were invited to follow a link on an email.

Participation rate (Ratio of unique visitors who agreed to participate/unique first survey page visitors)

We are unable to calculate the participation rate because we are unable to determine how many people were distributed the survey.

Completion rate (Ratio of users who finished the survey/users who agreed to participate)

122 participants agreed to participate in the survey after reading the participant information sheet on the first page. 99 participants completed the survey either partially or fully, giving an 81.1% completion rate. 58 Participants answered all survey questions.

Response rates

Preventing multiple entries from the same individual Cookies used

None used.

IP check

Due to the participants being from a single organisation, many responses came from the same IP address, (WiFi router).

Log file analysis

Not used

Registration

The survey was open to all that received the link.

Handling of incomplete questionnaires

Partially completed responses were included in the analysis.

Questionnaires submitted with an atypical timestamp

No respondents were removed for completing the survey too quickly, the shortest time taken to complete the survey was just over 6 mins.

Statistical correction

A satisfactory number of responses was achieved for the sub-groups surveyed.

Analysis

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Supplementary Table 3: Respondent time working in the fitness industry. How long have you worked in a gym in one capacity or another?

Count (n)

Percentage (%)

Up to 1 year

1-2 years

3-5 years

6-10 years

11-20 years

21 years and over

Total

100

Supplementary Table 4: Role descriptions. Contractor

Role description

Group Fitness Instructor

Deliver group exercise classes as designed by Les Mills International in accordance with the choreography and instructions. Adhere to the health and safety requirements of the Les Mills New Zealand clubs.

Personal Trainer

Design and deliver bespoke workouts to clients. Adhere to the health and safety requirements of the Les Mills New Zealand clubs.

Supplementary Table 5: Rank order data for the question “What exercises most commonly put people at risk of SHOULDER injury?” (n = 58 respondents). Rank order (1= most, 7= least)

Weighted average

Cleans (including clean & press)

2.7

Overhead shoulder press

2.7

Chest press/Chest fly

3.3

Upright row

3.9

Triceps dip

4.6

Lat pull down

5.8

Other (please describe)

7.0

Supplementary Table 6: Rank order data for the question “What exercises most commonly put people at risk of LOWER BACK injury?” (n = 58 respondents). Rank order (1= most, 5= least)

Weighted average

Deadlift

1.6

Squats

2.3

Cleans (including clean & press)

3.0

Bent over row

3.2

Other (please describe)

4.9

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Supplementary Table 7: Managers response to the exercises most to least likely to result in injury (n = 16 respondents). Exercise

Mean

Median

Minimum

Maximum

Deadlift

1.75

Squats

2.81

Cleans / clean & press

5.00

Chest press

5.81

6.5

Overhead shoulder press

6.19

5.5

Chest fly

7.81

Push ups

8.63

8.5

Bent over row

8.81

Upright row

9.31

11.5

Lateral raise

9.38

Pull ups / chin ups

10.19

9.5

Lat pull down

10.50

10.5

Triceps dip

10.69

Biceps curl

12.25

Triceps extension

12.44

12.5

Other

14.44

Notes: Exercises were reordered in response to the question “Based on your experience, please reorder the following exercises to rank in terms of their contribution to gym user injuries. (1 = most likely to result in injury, 10 = least likely to result in injury).”

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Supplementary Table 8: Selection of responses from Gym Managers (n = 17) related to staff training. How do you train the staff to observe and approach gym users and class participants who are exercising with unsafe technique? Exercise specific

Injury prevention

Other

C - Tell them the muscles targeted and the execution of the exercise. T - Our trainers know what correct technique looks like and will use the introduce, build rapport, offer advice funnel. C - Always introduce themselves to new group fitness participants and check in with them post class.

T - Safety cues, range.

T - Do a demo of the exercise.

C - In these situations, it is important to explain WHY you are intervening and that it is totally for their benefit

C - Engaging with people around what they are training for is often the key to achieving a positive outcome.

C - Be polite. If you start with the member's safety as the concern, you will get the best response. O - Open door policy to PT manager for PTs to speak about client issues/ share effective client approach techniques/ encourage networking with remedial specialists/ flag risk potential with PTs when I see it / teach-exercise regressions and alternatives.

T - Teach PTs and staff to screen client movement well.

T - 1 on 1 coaching through wide variety of exercises to gain confidence to approach and offer advice to members. U - Short regular technique training sessions on problematic lifts / mentor training with senior staff or PTs /advise PTs of upcoming courses of all types / run workshops several times a year run by specialists’ coaches / one on one talk when I see room for improvements / encourage culture of ongoing education and product improvement.

U - Upskill sessions and one-on-ones.

C - Communication approach and turning it into positive situation of member interaction.

T - Group fitness instructors are trained to give cues to the entire class during a session that are prompted by seeing an individual doing something wrong to address it without singling out. PTs are just trained to coach safe technique and to use that knowledge to approach people if they are willing.

Key: T = Staff training, C = Communication, U = Upskill, O = Other What additional information or resources, if any, do you need to train and assist staff to observe and intervene to prevent injury? Selection of Gym Managers responses (n = 17) I would like to see the return of the gym instructor role (employed position) to assist with this. Potentially a simplified "chart" of recommended exercises to correct certain exercise specific technique issues Adding a more specific injury prevention component into the PT Module would raise the general awareness of risk in the gym and give more confidence for PTs to approach situations they see. More available or practical education resources around injury management and preventive exercise. Conversation techniques. Most staff know what to do a lot of members don't want intervention even if they are doing something wrong.

JSES | https://doi.org/10.36905/jses.2026.02.03

Cuthbertson-Moon et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 17-37 (2026) Supplementary Figure 1: Summary of responses.

JSES | https://doi.org/10.36905/jses.2026.02.03

JSES The Journal of Sport and Exercise Science, Vol. 10, Issue 2, 38-48 (2026)

ISSN: 2703-240X

www.jses.net

Rugby sevens is not a ‘mini version’ of rugby fifteens: An integrated performance analysis framework for representative practice design in rugby sevens Zhanyu Huang1,2*, Keith Davids1, Martyn Rothwell3, Joseph Antony Stone1 1

School of Sport and Physical Activity, Sheffield Hallam University, UK JTSportech Research Institute of Sport Sciences (JTRISS), China 3 The Constraints Collective, UK 2

ARTICLE INFO Received: 26.09.2025 Accepted: 05.05.2026 Online: 24.10.2026 Keywords: Ecological Dynamics Performance Analysis Rugby Sevens Department of Methodology Representative Practice Design

ABSTRACT Representative practice design in team sports, like rugby (union) sevens is based on effective sampling of task and environment constraints, elucidated via performance analysis. However, performance analysis of rugby sevens has conventionally been undertaken with the tendency to follow performance methodologies from the sport of rugby (union) fifteens, which may lead to mismatching performance demands because of the different task constraints between the two codes. In this position paper, we consider how performance analysis to facilitate practice design in rugby sevens need to understand how specific task and environment constraints shape the way that skills are performed and collective patterns of behaviour emerge in that specific performance context. We aimed to analyse the self-organising performance tendencies in rugby sevens gameplays from an Ecological Dynamics framework, discussing: (i) how rugby sevens players perceive invitations for action (i.e. affordances) available in their surrounds, (ii) how game environment and task constraints constitute a landscape of affordances for performers to perceive and act upon, and (iii) why attacker-defender interpersonal transactions form the fundamental layer of nested interdependencies in competitive performance. We propose an integrated conceptual model for analysing performance of rugby sevens based on key concepts in Ecological Dynamics which may help performance analysts better extract representative and collective information regarding decision making and emergence of actions representative of the specific performance context of rugby sevens.

Introduction

Rugby sevens is a stand-alone variation of the traditional 15-aside game of rugby union, played for shorter duration and at a higher intensity level (Hogarth et al., 2016). A game of rugby sevens is played by two teams of seven players on a standard rugby field for two halves of seven minutes (World Rugby, 2023a). The game follows the general laws of rugby fifteens with several modifications, such as no place kicking, no rolling mauls and scrums formed by three players (World Rugby, 2023a). Since rugby sevens was admitted into the Olympics Games in 2016, it has gained increasing popularity. The success of the World Rugby Sevens Series and World Rugby Challenger Series in recent years

has also developed rugby sevens as a professional team sport, rather than merely a variation of its rugby fifteens counterpart. To achieve higher levels of competitiveness, high performance rugby sevens teams have started adopting sport science methodology, such as performance analysis, to facilitate their performance preparation and practice design, following precedents from rugby fifteens (Higham et al., 2014; Higham et al., 2013). Effective practice design in team sports, such as rugby union, include specific training activities in practice that seek to simulate performance contexts, representing task constraints of the competitive environment, using bona fide data from competitive contexts for athletes to submerge themselves in task constraints representing the competition environment (O’Sullivan et al., 2026;

Corresponding Author: Zhanyu Huang, Sheffield Hallam University, UK; JTSportech Research Institute of Sport Sciences (JTRISS), China, z.huang@shu.ac.uk

JSES | https://doi.org/10.36905/jses.2026.02.04

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) Woods et al., 2020; Woods et al., 2019; Woods et al., 2018). These practice tasks are representative of actual competition demands for improving performance by providing athletes with contextualised learning experiences (Pinder et al., 2011). Designing such representative practice constraints relies on effective sampling and analysis of statistical performance data from interactions between players and task constraints from competitive performance environment (Robertson et al., 2019; Robertson & Joyce, 2015). This approach ensures that interactions of players and task constraints are context-dependent (Davids et al., 2005) and specific to the competitive performance environment and ways of performing (Browne et al., 2019). To achieve this aim in rugby sevens, performance analysis should be applied to sample data on the continuous interactions between players and task constraints specifically in the contexts of rugby sevens. Recent work in other team sports has begun to align performance analysis more explicitly with representative learning design. For example, studies in Australian football have examined how informational constraints and their interactions differ across competition and training tasks, showing how performance analysis can be used, not only to describe outcomes, but also to evaluate the representative nature of practice designs (Browne et al., 2019; Browne et al., 2020). These developments are important because they demonstrate that contextual performance analysis can support the design and monitoring of training tasks intended to facilitate transfer from preparation to competition. However, despite this broader methodological direction, rugby sevens has still largely lacked a contextualised performance analysis approach that is specific to its own task constraints and performance demands. Despite varied ways of playing the game, performance in rugby sevens game play has traditionally been analysed as a mini version of the different team sport of rugby fifteens. Rugby sevens has not yet developed its own methodology of performance analysis (Barkell et al., 2017a; Barkell et al., 2018; Barkell et al., 2017b; Carreras et al., 2013; Hendricks et al., 2020; Higham et al., 2014a, 2014b; World Rugby, 2023b). Conventional performance preparation in rugby sevens tends to “copy and paste” a methodology from rugby fifteens (e.g., O’Sullivan et al., 2024, p.1) relying heavily on analysing performance indicators, statistical relationship of match variables, and interpretations of rugby fifteens (Finnegan et al., 2025). Regardless, conventional analysis in rugby sevens and fifteens relies extensively on identifying and describing isolated and discrete performance actions during distinct events and game phases, ignoring contextual performance variables which might play a significant role in shaping performance behaviours in rugby union (Colomer et al., 2020; Finnegan et al., 2025). This typical analytic methodology simplifies performance of rugby sevens as a process of inter-correlations of successful or unsuccessful performance outcomes (e.g., points scored, tries scored), highlighting frequencies of actions or events such as tackles completed, kicks in possession of the ball, and defensive line breaks with the ball (Colomer et al., 2020; Henderson et al., 2019; Ross et al., 2016). The underlying assumption of conventional performance analysis in rugby sevens is that to achieve a desired outcome in competitive performance, a rugby sevens team must improve their acquisition of a given set (dashboard) of conventional performance indicators. Typically, the specific values (e.g., frequencies, ratios, percentages) of key performance indicators have been associated with successful outcomes in competitive performance (McGarry, 2009). This mode JSES | https://doi.org/10.36905/jses.2026.02.04

of assessing performance is outcome driven and tends to guide coaches to design prescriptive practices to facilitate rugby seven players to acquire set values of isolated performance indicators, based around performance analysis data (Barkell et al., 2017; Henderson et al., 2019; Hendricks et al., 2020). This conventional approach does not adequately attend to differences in specific task constraints between rugby sevens and rugby fifteens, nor to implications of those differences for performance preparation, as outlined in Table 1. Table 1 combines: (i) rule- and structure-based differences between rugby sevens and rugby fifteens; (ii) implications supported by available literature where possible; and (iii), conceptually reasoned practice implications derived from an Ecological Dynamics conceptualisation of interacting task constraints. Where empirical support remains limited, implications are presented cautiously as context-sensitive propositions for performance preparation rather than as established universal effects. 1.1. The need for a contextualised performance analysis in sampling and evaluating task constraints Sport performance can be framed through multiple theoretical and conceptual approaches, for example informed by dynamical systems ideas, ecological psychology, and related models of behaviour. Over recent decades, concepts from dynamical systems theory have been especially influential in helping researchers understand the non-linear and self-organising properties of sport performance (e.g., Davids et al., 2005; McGarry et al., 2002). However, in applications, performance analysis in team sports has often remained centered on frequency analyses of discrete events, action counts, and outcome variables, which can make it difficult to account for how behaviours emerge through interactions among performers, tasks, and environments (O’Sullivan et al., 2026). In this paper, we adopt Ecological Dynamics as the primary theoretical rationale to provide a framework for understanding performer–environment transactions, affordances, interpersonal couplings, and how such contextual analyses may inform representative practice design. Our critique is directed at the persistent de-contextualisation of analytical practices that do not adequately account for how performance emerges from interacting constraints. These criticisms imply that, instead of focusing predominantly on frequencies of isolated behaviours in competition, performance analysts may usefully attend to the specific contexts in which performance behaviours emerge (O’Sullivan et al., 2026; Travassos et al., 2013; Vilar et al., 2012). Performance contexts mean everything, providing informational backdrops that channel how players perceive, decide, and act to solve performance problems (Seifert et al., 2016). Information on the task constraints under which behaviours emerge are immensely helpful for practitioners to understand not only what happened, but also how and why particular behaviours became functional within specific competitive scenarios. Effective sampling and contextual analysis of such information is important because it may help coaches recreate performance scenarios in practice that more closely reflect the informational demands of rugby sevens competition. Embedding practice designs in informational constraints revealed by contextualised performance analysis is critical for this change in focus. 39

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) Table 1: The perceived differences in task constraints between rugby sevens and rugby fifteens and implications for performance preparation. Rule differences

Perceived task constraints

Implications for performance preparation

A team formed by seven players instead of fifteen

Increased relative playing area (from 233m 2 per player in rugby fifteens to 500 m2 per player in rugby sevens).

Reduced availability of the supporting players for the ball carrier in the offensive team during offensive opportunities.

- Players have more space to cover, and they may need to perform in any area of the pitch during rugby sevens competition. - The ball carrier has more space to exploit and may face defensive challenges from different angles. - Forming structured plays become difficult during offensive opportunities for the attacking team. - Maintaining the defensive shape of the players becomes difficult for the defending team during defensive scenarios. - Reduced passing or decoying options for the ball carrier during offensive opportunities. - Ball carriers may more frequently encounter locally constrained 1 vs. 1 interactions with a primary defender, particularly when immediate support options are reduced.

Increased opportunity for 1 vs. 1 tackles and reduced opportunities for the tackler to form dominant tackle or 2 vs. 1 tackle.

- Defensive systems in rugby sevens are different than in rugby fifteens. Dominant tackles or 2 vs. 1 tackles are less seen and side-way tackles and grabbing tackles are more prevalent. - Tacklers will need to be effective in the 1 vs. 1 dyads.

Defenders may experience greater pressure in direct 1 vs. 1 situations when immediate cover is limited.

- Defenders faced higher defensive pressure during direct 1 vs. 1 dyads because they have no supporting defensive player to cover the space from behind.

Reduced time for the ball to be circulated out of the scrum.

- Props may enter the scrum under higher physical exertion and there are more tactical scrum options (e.g, inside prop up or outside prop up).

Increased interpersonal gaps in both the inside and outside channel after ball is out of the scrum.

- Scrum half can attack both channels with a ball carry once the ball is out of the scrum.

A game lasts fourteen minutes instead of eighty minutes

Increased temporal pressure for both teams and accelerated pace of the game.

- Players may be required to make faster information-regulated decisions in highly variable performance contexts under considerable physical, tactical, and psychological exertion.

No placed kick

Players are constrained to either play the ball using quick tap, call for scrum or kick out of touch for lineout after penalty is given.

- Penalty and free-kick restarts may increase ball-in-play demands by reducing opportunities for prolonged stoppages (Pollard et al., 2018). - Direct score from kicking becomes rare.

No rolling maul

Team cannot have a group of players bonded together to progress the attack during competition.

- The game may become more dynamic, with synergies between players being formed and decoupled more rapidly during physical contests.

Increased interpersonal gaps during competition.

A scrum formed by three players instead of eight

JSES |

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) In the remainder of this paper, we discuss how an ecological dynamics rationale may underpin performance analysis to benefit practitioners (e.g., performance analysts, sport scientists and coaches collaborating together)) in analysing and understanding the specific performance constraints. Key insights from the theory of Ecological Dynamics may provide an appropriate conceptual framework for examining performance constraints of rugby sevens, conceiving how players may co-adapt to each other’s action to achieve intended performance goals (Rothwell et al., 2021; Woods et al., 2020; Woods et al., 2020). In the following sections, we: (i) analyse the emergence of performance in rugby sevens through an ecological dynamics lens; and (ii) propose how performance analysis may be integrated with ecological dynamics to frame the self-organising and emergent nature of rugby sevens performance in a way that is more useful for representative practice design. 2.

An Ecological Dynamics lens for analysing performance in rugby sevens

Ecological Dynamics is a contemporary theoretical framework for understanding the non-linear and context-dependent emergence of sport performance behaviours (Araújo et al., 2006). Rooted in ecological psychology and dynamical systems theory, it explains how performers regulate action through continuous transactions with task-relevant information in their performance environments (Vilar et al., 2012). The performer-environment transactional relationship underpins the co-adaptation of each individual to specific task constraints in performance contexts (Araújo et al., 2020). Performers actively seek and perceive relevant information to support their actions in contexts, bringing unique characteristics (e.g., abilities, capacities, and knowledge) to meet that challenge (Passos et al., 2009). Tasks and environmental properties of the competitive environment act as constraints that channel their actions, as players form goal-oriented synergies during performance (Silva et al., 2013). Performance preparation from an Ecological Dynamics perspective, aims to improve players’ capacity to produce accurate, timely, efficient and functional movement solutions to solve specific performance problems in various rugby sevens scenarios, under specific competitive performance constraints (Bernstein, 1967; Woods et al., 2019). The essence of effective performance preparation is to construct a niche in which athletes are free to explore the specific constraints presented by coaches, creating their self-organised performance solutions (Rothwell et al., 2021). Performance niches in practice that fail to represent competitive performance demands may not channel athletes towards desirable action possibilities (Woods et al., 2020). To improve athlete coupling of perception and action, coaches need to ensure high ‘training specificity’ by integrating task constraints, and information on player-environment relations and inter-player relationships, sampled from performance data in actual competitions (Headrick et al., 2015). In this way, practice activities can offer players a rich landscape of information that act as affordances to fine-tune their performance solutions, for example, helping them to form synergies with teammates in rugby sevens-specific scenarios (Rietveld & Kiverstein, 2014). Effective constraint sampling, therefore, is a key prerequisite to construct a niche for effective performance preparation (Woods JSES |

et al., 2020). Constraints sampled need to be: (i) relevant to achieving desired performance goals (e.g. constraints to score a try or successfully regain possession of the ball), (ii) representing the scenarios in which performance challenges emerge (e.g. a winger always making poor decisions during 1 vs. 1 dyad from a chaotic turnover), and (iii), effectively perceived by the stakeholders (e.g. athletes perceive information available in task constraints via direct experience (gaining knowledge of the environment; Gibson, 1979), while coaches may provide this information to athletes indirectly, via augmented information (providing knowledge about the environment; Gibson, 1979). Guided by an Ecological Dynamics framework, performance analysts of rugby sevens may support their knowledge about the performance environment through conventional statistics-based analyses by understanding how rugby sevens player perceive relevant informational constraints within a performance context (identifying key task and environmental constraints). These key ideas can help analysts attend to how a team of rugby players may form synergies on field to collectively achieve performance goals during competition (Woods et al., 2020). 2.1. Differentiating knowledge about and knowledge of the performance context The quality of performance analysis used to support preparation for competition depends not only on what information is delivered, but also on how that information relates to the ways in which performers perceive and act in context (Fajen et al., 2008). Conventional performance analysis often provides stakeholders with knowledge about the performance environment through statistics, reports, video clips, and visual summaries. Such augmented information can be useful for coaches and analysts to support guidance and feedback to performers. However, from an ecological dynamics perspective, performers also require knowledge of the environment: information that is directly perceived through continuous engagement with task-relevant affordances in practice and competition (Gibson, 1979). Importantly, the utilisation of available affordances is not uniform across performers. In the same performance environment, players with different functional capacities, tactical roles, and skill histories may perceive different invitations for action and act upon them in different ways (Rietveld & Kiverstein, 2014). For this reason, the challenge for coaches and performance analysts is not only to describe competitive performance, but also to ensure that representative practice designs preserve the perceptual information needed for players to attune to relevant opportunities for action. In this sense, augmented information may support learning, but it does not replace the value of direct engagement with the informational dynamics of performance (O’Sullivan et al., 2023; O’Sullivan et al., 2026). In the same performance environment, performers of varied skill-levels, with different functional capacities, may pick up different affordances to guide their decisions and actions (Woods, et al., 2019; Rietveld & Kiverstein, 2014). For example, in a defensive scenario, a shoulder-dropping tackle posture from a defender may invite an experienced ball carrier to commit with a strong fend and drive past the defender at speed. Whereas, for an inexperienced player, a shoulder dropping tackle may invite them to brace for contact and then to fall to the ground to avoid losing 41

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) ball possession. In this case, the richness of affordances in the competitive rugby sevens performance landscape shapes the discursive space of possibilities that represent the possible behaviours which can be supported in individual performers (Aggerholm et al., 2011). The current abilities, capacities, and tendencies (termed effectivities by Gibson, 1979) of performers shape which affordances may be used to solicit actions by an individual. Skilful performers, with a wider repertoire of foundational skills and abilities, are increasingly able to select more relevant affordances from a wider range, exploiting more meaningful resources and adjusting their actions to fulfil their specific performance intentions. Performance analysis for rugby sevens, therefore, could focus on helping coaches and skill acquisition specialists to design a rich landscape of affordances specific to individual recipients for them to pick up and use in practice in individualised performance preparation (Fajen et al., 2008). 2.2. Environmental and task constraints incubate a rich landscape of affordances in rugby sevens Representative design of constraints in practice influences the space of challenges, problems faced and movement solutions available in practice activities. Distinct features of rugby sevens, including the modified rules and enlarged relative playing area available for the players set up regimes of constraints packed with information for players to perceive (Rothwell et al., 2018). Accurately sampling such informational constraints is key to effectively reconstructing a representative niche for performance preparation. Task and environmental constraints can collectively create a rich landscape of affordances for players to perceive and act upon during preparation for rugby sevens match play. For instance, reduced numbers of players and vast relative areas of the field available to explore and exploit may create opportunities for action in the potential for overloads in inter-personal space between defending players and availability of covering teammates (Granatelli et al., 2014). Larger relative space availability serves as a significant performance constraint, guiding the defensive team to find ways to adopt a more compact defensive line in the game flow. Performance analysts, therefore, could sample information from statistical patterns that emerge from key task and environmental constraints, such as space, player positioning and the like, that reflect high levels of specificity to rugby sevens game plays to facilitate coaches to design representative training activities. In addition, the referee should not be viewed only as an external arbiter of the game, but also an informational constraint of the competitive environment in which performance emerges. The decision-making of individual referees can shape restart tempo, the continuation or interruption of advantage, the thresholds of acceptable contact, and the extent to which attacking or defensive play is allowed to flow. In this sense, referees may contribute to the evolving informational landscape shaping the behaviours of players during competition. This view aligns with work in association football suggesting that referees co-construct the game with players and that referee decision making is itself shaped by interacting constraints nested within task goals (Russell et al., 2019). JSES |

Constraints should also be sampled in a way that, in combination, incubate the synergies formed between teammates during rugby sevens competition. Synergy formation is a fundamental property of complex adaptive interactive systems in sport (Araujo & Davids, 2016). Synergy formation in teams during competition is a process of behavioural self-organisation emerging from complex and dynamical interactions between subsystems (i.e., individuals or combinations of players) during competition (Araujo & Davids, 2016). The transactional relationships formed between performers and teammates, opponents and referee during competition constantly shape the collective system dynamics from which affordances emerge (Araújo et al., 2013). Since game play in rugby sevens is dynamical and open, changes in energy distribution patterns (e.g., information from gaps, locations, distances, movements) around performers rapidly emerge and dissipate. Players constantly find themselves in transient game scenarios, with minimal time to pick up the most relevant invitations for action available in the surrounding affordance landscape. The affordances of a performance context emerge and disappear quickly, shaped under continuous transactions between players and environment. Practice designs should seek to simulate this dynamical nature of competition to support players in their problem-solving and decision-making capacities. Coaches could support this aspect of skilled performance in rugby sevens by constructing a constraining performance niche (founded on performance data) that submerges players in continuous game-specific transactions. The aim is to help performers learn to form functional synergetic relationships on field, to achieve task goals in performance. 2.3. Temporal constraints place perception-action coupling under pressure in rugby sevens competition Rugby sevens players may constantly be required to switch from attack to defence or vice versa in a fraction of a second. This temporal pressure in rugby sevens performance is a constant constraint that accelerates the coupling of perception and action during competition, soliciting rapid decision making under chaotic and complex competitive scenarios. Compared with rugby fifteens, a game lasting for 80 minutes, with both teams having many opportunities to contest ball possession, rugby sevens game play is much shorter. Short game duration and temporal pressure invite a higher game pace, soliciting different physical and mental demands on players. They are required to be continuously active, moving with high intensity during a sevens game, with a small number of breaks in between (Furlan et al., 2015; Latham-Coyle, 2024). In rugby sevens, each team typically has 6 to 12 possession phases that last from several seconds to one or two minutes (World Rugby, 2023b). Players from an attacking team need to either quickly identify space available to attempt a line break or patiently circulate the ball under high defensive pressure; whereas players from a defensive team need to constantly shift to reposition themselves to cover potential gaps and pressurise attacking players. Possession turnovers in ‘ball-in-play’ time are relatively more frequent in rugby sevens than rugby fifteens (Higham et al., 2014b). Adding temporal constraints into analyses may help performance analysts and coaches better understand how perception and action coupled under the constraints of rugby 42

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) sevens competition. This innovation may influence how different landscapes of informational constraints may shape the formation of different synergies between players. For a broadly similar set of spatial and interpersonal relations, different levels of temporal pressure may invite quite different decisions and performance outcomes. In a 1 vs. 1 scenario close to the touchline, for example, a ball carrier with more time available may approach the interaction with greater control, potentially committing the defender with a fend or an evasive step. By contrast, when time is compressed, the state space of possible actions may become restricted, and the opportunities available to the player may emerge and disappear more rapidly. Relatedly, performance observed in lower-pressure or more self-paced practice tasks may not transfer directly to more time-compressed, externally paced competitive contexts. Research in rugby union has shown that manipulating practice task constraints can reshape emergent decision-making behaviours by altering the informational relations available to players (Correia et al., 2012). This finding supports the view that temporal pressure should not be treated as a background condition, but as a rate-modifying constraint that can substantially change the affordances available in 1 vs. 1 dyadic interaction. At the same time, direct evidence on this issue in rugby sevens remains limited, highlighting an important area for future study. 2.4. Focusing attacker-defender interpersonal transactions in rugby sevens practice and performance Performance in rugby sevens emerges extensively through interpersonal transactions between competing performers (Araújo et al., 2020). Previous research in rugby union and other invasion games has shown that eco-physical variables like interpersonal distance, relative velocity, player co-positioning, and goal-related information can shape the coordination tendencies that emerge in attacker-defender dyadic systems (Passos et al., 2008; Correia et al., 2016; Vilar et al., 2014). These findings suggest that attacker-

defender interactions can be treated as meaningful local units of analysis when attempting to understand how performance behaviours emerge under specific task constraints. For rugby sevens, such work is especially relevant because it supports a move away from treating performance as a sequence of isolated actions and toward understanding how behaviour is shaped by context-sensitive couplings between performers. Interpersonal transactions are context-dependent couplings that emerge between players over time, whether relatively stable or rapidly changing (Woods et al., 2024). Rugby sevens contains multiple forms of interpersonal relations, including linked teammate interactions, multi-player configurations at scrums and rucks, and attacker-defender couplings in open play. However, the larger relative playing space, reduced player density, and frequent exposure to direct defensive confrontation in rugby sevens suggest that attacker-defender 1 vs. 1 dyads may be especially important localised interactions in many open-play scenarios. This does not mean that they are the only meaningful level of analysis, nor that other collective relations are unimportant. Rather, attacker-defender dyads appear to warrant analytical priority because they often provide a proximal layer through which broader task and environmental constraints become behaviourally expressed. On this basis, attacker-defender couplings may be understood as a primary local layer within the nested interpersonal organisation of rugby sevens performance, particularly when analysts seek to support representative practice design around ball-carrying, line-breaking, or first-defender contests. 3.

An integrated conceptual model for performance analysis of rugby sevens

Summarising the key messages of this position statement so far, a conceptual model to specifically support the contextual analysis of performance in rugby sevens may be conceived, predicated on an Ecological Dynamics framework (Figure 1):

Figure 1: Conceptual model of rugby sevens performance analysis based on an Ecological Dynamics framework. JSES |

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) This model acknowledges first that rugby sevens game play is constituted by a continuous, context-dependent sequence of actions and events between the onset and the end of competitive games. In this model, the central focus is on performers and their continuous transactions with changing environmental and task constraints, which they rely on to organise their actions in varied game scenarios. The model suggests that the continuous, information-based, transactional relations that emerge from the dynamics of the environment, tasks and performers alone shape the emergence of performance outcomes. A rugby sevens game may contain 20 to 40 game scenarios which can be grouped according to different types of player-environment transactions, such as attacking build up, defensive build up, breakdown and setpieces, amongst others (World Rugby, 2023b). The model proposes that performance analysis of rugby sevens should focus on the overlapping of performance behaviours in varied fields of affordances, with performance outcomes assessed in specific competitive contexts. This model clarifies that performance analysts should consider performance behaviours in context to better understand how and why performers couple their perception and action in a field of affordances available in different rugby union formats. Intended performance outcomes constitute the attractor wells of the dynamic performance system transitioned towards in practice. Performance analysts may extract performance information from every scenario regarding the continuous relationship between interacting constraints and performance behaviours emerging between teammates to achieve desired outcomes. This comprehensive set of performance information not only shows the emergence of performance behaviours but also indicates the collective constraints regime regulating the perception and action of the players and their potential responses in each game scenario. It provides a rich source of information for other practitioners, including coaches, trainers, players and sport scientists with abundant information to guide performance preparation. This model resonates with another contemporary professional support model, a Department of Methodology (DoM), which facilitates a transdisciplinary approach to sport science support in elite sporting teams (Davids et al., 2023; Otte et al., 2020; Rothwell et al., 2020). The DoM advocates that, to achieve effective collaboration in a sport science support team with many subdisciplines, communications between practitioners with different backgrounds should be coordinated through shared principles and coherent theoretical concepts on performance, learning and development. Performance preparation should be approached by infusing rich information seeking to guide the emergence of multi-dimensional behaviours (Rothwell et al., 2020). Performance information extracted from this model provides solid support for the practical application of DoM in sport science support teams (e.g., Hydes et al., 2026). Our model for the future of performance analytics in rugby union provides rich performance information related to the emergence of successful and unsuccessful performance outcomes. Therefore, other practitioners may identify context-specific variables and nested relations between them. They may add their own interpretation to the information based on shared principles to reflect team or individual strengths and weaknesses (Rothwell et al., 2020). Information related to potential performance behaviours in JSES |

competition also provides a reference for the design of different practice tasks (Correia et al., 2012). 3.1. Example: Ecological Dynamics performance analysis of ball carries in rugby sevens Consistent with the previous analysis, we again focus on ball carrying scenarios to elaborate the functioning of the conceptual model. The large space availability and single layer of defending players can act as key affordances inviting ball carriers to attack available space behind defenders with speed. To analyse ball carrying performance behaviours, therefore, key environmental and task constraints need to be identified, including space availability, time availability, current game state, defenders’ interpersonal distances and positioning, body orientation and posture of the primary defender(tackler), supporting teammates’ availability, as well as the individual capacities of the ball carrier, The latter include skill proficiency, speed, visual focus, estimated fatigue states, and choices available. In this scenario, analysts may summarise a set of alternative performance behaviours for ball carriers. For instances, ball carriers may attempt to (i) commit with strong fend, (ii) deceive the tackler with deceptive body orientation, (iii) pass to the teammates, (iv) commit the defender and offload to a teammate, (v) drop kick for points, and more. After listing constraints and performance behaviours, analysts could then consider desired outcomes in this scenario to evaluate nested affordances available from different combinations of constraints and performance behaviours to identify specific emergent coordination patterns as reference. The application of the model is shown in Figure 2. To illustrate how the conceptual model may be operationalised in practice, Figure 2 outlines the analysis of ball-carrying performance in rugby sevens, while Figure 3 presents a hypothetical 1 vs. 1 example showing how interacting task, environmental, and interpersonal constraints may shape an emergent performance outcome. Figure 3 presents a hypothetical example of how contextual information may be extracted from a ball carrier–primary defender 1 vs. 1 scenario in rugby sevens using the conceptual model outlined in Figure 2. We approached the scenario from the ball carrier perspective. First, a detailed description and illustration were presented from a notational analysis of this performance scenario. Environmental and task constraints, as well as personal constraints, were also noted exclusively by practitioners for this scenario. Potential behaviours, listed from the most to the least desired, were annotated according to previous analyses and coaches’ experiences. This information presented practitioners with clear views of how the winger typically performed in this performance context, instead of assessing their performance merely from abstract statistical data analysis. The new analytic method reflects the interpersonal transactional relationship in the dyad by showing how an opponent’s movements, and co-adaptive positioning afforded the ball carrier opportunities to perceive and act in a competitive performance scenario. The analysis also offers coaches a rich landscape of contextual information for how practice games and activities may be designed to simulate a competitive game scenario for players to improve their decision making and guide emergence of better performance. 44

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026)

Figure 2: Conceptual model of ball-carry performance analysis in rugby sevens based on an ecological dynamics framework, illustrating how interacting constraints may shape gain advantage line/field progression and try-scoring opportunities.

Figure 3. Hypothetical analysis of a ball carrier–primary defender 1 vs. 1 scenario in rugby sevens using the conceptual model shown in Figure 2. JSES |

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) 3.2. From contextual performance analysis to representative practice design in rugby sevens The practical value of contextual performance analysis lies not only in explaining past performance, but also in informing the design of future practice tasks. In rugby sevens, this approach signifies that training activities and games should be built around the preservation or purposeful manipulation of the key informational constraints that shape competitive performance. Rather than prescribing isolated technical actions, coaches may design representative tasks that: (i) retain the relevant performeropponent relations that emerge in competition; (ii) manipulate space, support availability, and practice tempo to reproduce specific game scenarios; (iii) expose players to forms of temporal pressure that shape affordance perception in match play; and (iv), evaluate practice success, not only through outcomes, but through the extent to which players become better attuned to contextuallyrelevant informational constraints. In this way, contextual performance analysis may support a more direct translation from competition-derived information to representative practice design in rugby sevens. 4.

Conclusion

Conventional action-centred analysis of performance in rugby sevens, is largely inherited from rugby fifteens, and may be insufficient for understanding how performance behaviours emerge under the distinct task constraints of the sevens game. In this paper, we have argued that ecological dynamics offers a suitable conceptual rationale for analysing rugby sevens performance in a way that foregrounds prominence of performer– environment transactions, affordances, and interpersonal couplings. Rather than treating performance as the accumulation of isolated actions or the attainment of predefined KPI values, the proposed framework encourages analysts to examine how behaviours emerge within specific game scenarios shaped by interacting constraints. Within this broader performance landscape, attacker-defender couplings may be understood as a particularly important local layer for analysis in many open-play situations, especially when the aim is to inform representative practice design around ball-carrying and defensive confrontation. This position piece proposes that contextual performance analysis may help practitioners develop more representative and practically useful understanding of rugby sevens competition, while also providing a stronger basis for future empirical work on the informational constraints that shape performance in this game. Conflict of Interest The authors declare no conflict of interests. Acknowledgment The authors declare no acknowledgement. References Araújo, D., Davids, K., & Hristovski, R. (2006). The ecological dynamics of decision making in sport. Psychology of Sport and JSES |

Exercise, 7(6), 653–676. https://doi.org/10.1016/j.psychsport.2006.07.002 Araújo, D., Davids, K., & Renshaw, I. (2020). Cognition, emotion and action in sport: An ecological dynamics perspective. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook of sport psychology (4th ed., pp. 535–555). Wiley. https://doi.org/10.1002/9781119568124.ch25 Bernstein, N. A. (1967). The co-ordination and regulation of movements. Pergamon Press. Browne, P. R., Sweeting, A. J., Davids, K., & Robertson, S. (2019). Prevalence of interactions and influence of performance constraints on kick outcomes across Australian Football tiers: Implications for representative practice designs. Human Movement Science, 66, 621–630. https://doi.org/10.1016/j.humov.2019.06.013 Browne, P. R., Woods, C. T., Sweeting, A. J., & Robertson, S. (2020). Applications of a working framework for the measurement of representative learning design in Australian football. PLoS ONE, 15(11), e0242336. https://doi.org/10.1371/journal.pone.0242336 Correia, V., Araújo, D., Duarte, R., Travassos, B., Passos, P., & Davids, K. (2012). Changes in practice task constraints shape decision-making behaviours of team games players. Journal of Science and Medicine in Sport, 15(3), 244–249. https://doi.org/10.1016/j.jsams.2011.10.004 Correia, V., Passos, P., Araújo, D., Davids, K., Diniz, A., & Kelso, J. A. S. (2016). Coupling tendencies during exploratory behaviours of competing players in rugby union dyads. European Journal of Sport Science, 16(1), 11–19. https://doi.org/10.1080/17461391.2014.915344 Davids, K., Araújo, D., & Shuttleworth, R. J. (2005). Applications of dynamical systems theory to football. In T. Reilly, J. Cabri, & D. Araújo (Eds.), Science and football V: The proceedings of the Fifth World Congress on Science and Football (pp. 537– 550). Routledge. https://doi.org/10.4324/9780203412992-204 Davids, K., Button, C., & Bennett, S. J. (2008). Dynamics of skill acquisition: A constraints-led approach. Human Kinetics. Davids, K., Renshaw, I., & Glazier, P. (2005). Movement models from sports reveal fundamental insights into coordination processes. Exercise and Sport Sciences Reviews, 33(1), 36–42. Davids, K., Rothwell, M., Hydes, S., Robinson, T., & Davids, C. (2023). Enriching athlete–environment interactions in youth sport: The role of a Department of Methodology. Children, 10(4), 752. https://doi.org/10.3390/children10040752 Fajen, B. R., Riley, M. A., & Turvey, M. T. (2009). Information, affordances, and the control of action in sport. International Journal of Sport Psychology, 40(1), 79–107. Finnegan, C., Scriney, M., Goss, H., Walsh, J. C., Bezbradica, M., & O’Hagan, A. D. (2025). Key performance indicators in rugby sevens: A systematic review. International Journal of Sports Science & Coaching, 20(6), 2831–2851. https://doi.org/10.1177/17479541251351294 Headrick, J., Renshaw, I., Davids, K., Pinder, R. A., & Araújo, D. (2015). The dynamics of expertise acquisition in sport: The role of affective learning design. Psychology of Sport and Exercise, 16(1), 83–90. https://doi.org/10.1016/j.psychsport.2014.08.006 Henderson, M. J., Fransen, J., McGrath, J. J., Harries, S. K., Poulos, N., & Coutts, A. J. (2019). Situational factors affecting rugby sevens match performance. Science and Medicine in 46

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) Football, 3(4), 275–280. https://doi.org/10.1080/24733938.2019.1609070 Henderson, M. J., Harries, S. K., Poulos, N., Fransen, J., & Coutts, A. J. (2018). Rugby sevens match demands and measurement of performance: A review. Kinesiology, 50(Suppl. 1), 49–59. Hendricks, S., Sin, D. W., van Niekerk, T., den Hollander, S., Brown, J., Maree, W., Treu, P., & Lambert, M. (2020). Technical determinants of tackle and ruck performance in international rugby sevens. European Journal of Sport Science, 20(7), 868–879. https://doi.org/10.1080/17461391.2019.1675764 Higham, D. G., Hopkins, W. G., Pyne, D. B., & Anson, J. M. (2014). Relationships between rugby sevens performance indicators and international tournament outcomes. Journal of Quantitative Analysis in Sports, 10(1), 81–87. https://doi.org/10.1515/jqas-2013-0095 Hogarth, L. W., Burkett, B. J., & McKean, M. R. (2016). Match demands of professional rugby football codes: A review from 2008 to 2015. International Journal of Sports Science & Coaching, 11(3), 451–463. https://doi.org/10.1177/1747954116645209 Hydes, S., Strafford, B. W., Rothwell, M., Stone, J., Davids, K., & Otte, F. (2026). A department of methodology: A feasible framework to integrate the applied practice of multidisciplinary support teams. International Journal of Sports Science & Coaching. Advance online publication. https://doi.org/10.1177/17479541251409729 McGarry, T. (2009). Applied and theoretical perspectives of performance analysis in sport: Scientific issues and challenges. International Journal of Performance Analysis in Sport, 9(1), 128–140. https://doi.org/10.1080/24748668.2009.11868469 McGarry, T., Anderson, D. I., Wallace, S. A., Hughes, M. D., & Franks, I. M. (2002). Sport competition as a dynamical selforganizing system. Journal of Sports Sciences, 20(10), 771– 781. https://doi.org/10.1080/026404102320675620 O’Sullivan, M., Manna, M., & Davids, K. (2026). Re-framing performance analysis in sport science and psychology from an ecological dynamics perspective: Toward a corresponsive, relational, and culturally situated practice. In T. Teo (Ed.), The Palgrave encyclopedia of theoretical and philosophical psychology (pp. 1–11). Palgrave Macmillan. https://doi.org/10.1007/978-3-031-70581-6_553-1 O’Sullivan, M., Vaughan, J., Woods, C. T., & Davids, K. (2025). There is no copy and paste, but there is resonation and inhabitation: Integrating a contemporary player development framework in football from a complexity sciences perspective. Journal of Sports Sciences, 43(1), 99–108. https://doi.org/10.1080/02640414.2023.2288979 Passos, P., Araújo, D., Davids, K., Gouveia, L., Serpa, S., Milho, J., & Fonseca, S. (2009). Interpersonal pattern dynamics and adaptive behavior in multiagent neurobiological systems: Conceptual model and data. Journal of Motor Behavior, 41(5), 445–459. https://doi.org/10.3200/35-08-061 Paul, L., Naughton, M., Jones, B., Davidow, D., Patel, A., Lambert, M., & Hendricks, S. (2022). Quantifying collision frequency and intensity in rugby union and rugby sevens: A systematic review. Sports Medicine - Open, 8(1), 12. https://doi.org/10.1186/s40798-021-00398-4 Pinder, R. A., Davids, K., Renshaw, I., & Araújo, D. (2011). Representative learning design and functionality of research JSES |

and practice in sport. Journal of Sport & Exercise Psychology, 33(1), 146–155. https://doi.org/10.1123/jsep.33.1.146 Pollard, B. T., Turner, A. N., Eager, R., Cunningham, D. J., Cook, C. J., Hogben, P., & Kilduff, L. P. (2018). The ball in play demands of international rugby union. Journal of Science and Medicine in Sport, 21(10), 1090–1094. https://doi.org/10.1016/j.jsams.2018.02.015 Rietveld, E., & Kiverstein, J. (2014). A rich landscape of affordances. Ecological Psychology, 26(4), 325–352. https://doi.org/10.1080/10407413.2014.958035 Robertson, S., Spencer, B., Back, N., & Farrow, D. (2019). A rule induction framework for the determination of representative learning design in skilled performance. Journal of Sports Sciences, 37(11), 1280–1285. https://doi.org/10.1080/02640414.2018.1555905 Rothwell, M., Davids, K., Stone, J., O’Sullivan, M., Vaughan, J., Newcombe, D., & Shuttleworth, R. (2020). A department of methodology can coordinate transdisciplinary sport science support. Journal of Expertise, 3(1), 55–65. Rothwell, M., Stone, J., & Davids, K. (2023). Exploring niche construction in sport coaching: An ecological dynamics analysis. Sports Coaching Review, 12(2), 209–231. https://doi.org/10.1080/21640629.2021.1974692 Russell, S., Renshaw, I., & Davids, K. (2019). How interacting constraints shape emergent decision-making of national-level football referees. Qualitative Research in Sport, Exercise and Health, 11(4), 573–592. https://doi.org/10.1080/2159676X.2018.1493525 Seifert, L., Komar, J., Araújo, D., & Davids, K. (2016). Neurobiological degeneracy: A key property for functional adaptations of perception and action to constraints. Neuroscience & Biobehavioral Reviews, 69, 159–165. https://doi.org/10.1016/j.neubiorev.2016.08.006 Travassos, B., Davids, K., Araújo, D., & Esteves, T. P. (2013). Performance analysis in team sports: Advances from an ecological dynamics approach. International Journal of Performance Analysis in Sport, 13(1), 83–95. https://doi.org/10.1080/24748668.2013.11868633 Vilar, L., Araújo, D., Davids, K., & Button, C. (2012). The role of ecological dynamics in analysing performance in team sports. Sports Medicine, 42(1), 1–10. https://doi.org/10.2165/11596520-000000000-00000 Vilar, L., Araújo, D., Davids, K., & Travassos, B. (2012). Constraints on competitive performance of attacker–defender dyads in team sports. Journal of Sports Sciences, 30(5), 459– 469. https://doi.org/10.1080/02640414.2011.627942 Vilar, L., Araújo, D., Davids, K., Travassos, B., Duarte, R., & Parreira, J. (2014). Interpersonal coordination tendencies supporting the creation/prevention of goal scoring opportunities in futsal. European Journal of Sport Science, 14(1), 28–35. https://doi.org/10.1080/17461391.2012.725103 Woods, C. T., Araújo, D., & Davids, K. (2024). On a corresponsive sport science. Sports Medicine, 54(5), 1071–1084. https://doi.org/10.1007/s40279-023-01981-3 Woods, C. T., McKeown, I., O’Sullivan, M., Robertson, S., & Davids, K. (2020). Theory to practice: Performance preparation models in contemporary high-level sport guided by an ecological dynamics framework. Sports Medicine - Open, 6(1), 36. https://doi.org/10.1186/s40798-020-00268-5 47

Huang et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 38-48 (2026) Woods, C. T., McKeown, I., Rothwell, M., Araújo, D., Robertson, S., & Davids, K. (2020). Sport practitioners as sport ecology designers: How ecological dynamics has progressively changed perceptions of skill “acquisition” in the sporting habitat. Frontiers in Psychology, 11, 654. https://doi.org/10.3389/fpsyg.2020.00654 Woods, C. T., McKeown, I., Shuttleworth, R. J., Davids, K., & Robertson, S. (2019). Training programme designs in professional team sport: An ecological dynamics exemplar.

JSES |

Human Movement Science, 66, 318–326. https://doi.org/10.1016/j.humov.2019.05.015 World Rugby. (n.d.). Laws of the game. World Rugby Passport. Retrieved August 19, 2026, from https://passport.world.rugby/laws-of-the-game/ World Rugby. (2014, November 11). Sevens game analysis. Retrieved August 19, 2026, from https://www.world.rugby/the-game/gameanalysis/reports/sevens-men

JSES The Journal of Sport and Exercise Science, Vol. 10, Issue 2, 49-53 (2026)

ISSN: 2703-240X

www.jses.net

The effect of different swim cap shapes and surfaces on drag Peter Olsen1*, Lindsey Alton2, Helen Marshall1, Nay Lin Oo2, Michael J. Hamlin3, and Graeme Harris2 1

Department of Applied Sciences and Social Practice, Ara Institute of Canterbury, New Zealand Department of Engineering and Architectural Studies, Ara Institute of Canterbury, New Zealand 3 Department of Tourism, Sport and Society, Lincoln University, New Zealand 2

ARTICLE INFO Received: 18.06.2025 Accepted: 27.09.2026 Online: 24.10.2026 Keywords: Swim Caps Swim Cap Shape Swim Cap Surface Equity Drag

ABSTRACT Swim cap research has typically focused on male swimmers with short hair that produces a dome swim cap shape. However, elite female backstroke competitors usually wear their hair in a low bun or high bun, which does not produce a dome shape. It was also observed swimmers in the Paris Olympics wearing the Arena Aquaforce wave swim cap that has an undulating rather than a smooth surface, which might affect drag. Consequently, the effect of different swim cap shapes and surfaces on aerodynamic drag are investigated. Drag was measured in a wind tunnel set to a wind speed of 10 𝑚𝑠 −1 with a Reynolds number similar to swimming at 1.76 𝑚𝑠 −1 in water. Aerodynamic drag was measured on a model head with long hair in six conditions: No Cap (NC), Low Bun (LB), High Bun (HB), Pineapple Bun (PB), Messy Bun (MB) and Aero Bun (AB), in two swim cap surface conditions - smooth and undulating. A model head was at a fixed, horizontal position equivalent to the head position in backstroke. The NC condition for LB, HB, and PB shapes produced considerably more drag (> 0.147 N, p < 0.001) than all swim cap shapes and surface conditions. The LB cap produced more drag (> 0.098 N, p <0.001) than other shapes and independent of surface conditions. There was no significant difference (p = 0.06) between the lowest undulating and smooth swim cap conditions (PB undulating vs HB smooth, 0.298N vs 0.3N). The LB shape produced approximately 28% more drag than all other swim cap shapes. Long hair competitive backstroke swimmers should not wear a low bun when using a swim cap. In contrast, swim cap surface (smooth or undulating) appears to have minimal impact on drag on a model head in a backstroke position.

Introduction

Research on swim caps has found the composition and structure of the cap has a significant effect on drag (Gatta et al., 2013, 2015; Harris et al., 2024; Marinho et al., 2011; Olsen, Alton, et al., 2024). Gatta et al. (2013) found a smooth rigid silicone swim cap without seams caused less passive drag than silicone or Lycra caps with seams. Further research by Gatta et al. (2015), found a wrinkled silicone swim cap caused more speed-specific drag than a dimpled or smooth swimming cap when swimmers had their arms alongside the body. Consequently, Gatta et al. (2015) indicated a wrinkled swim cap could potentially be detrimental to swimming performance. Recent research by Harris et al. (2024) investigated the effect of different types and sizes of swim caps on

aerodynamic drag in a wind tunnel. As in research by Gatta et al. (2013, 2015), the presence of wrinkles was found to increase drag, and a size effect was also observed, with drag progressively increasing in larger swim caps due to wrinkling and a looser fit. Based on their findings, Gatta et al. (2015) and Harris et al. (2024) suggested that swimmers should ensure there are minimal ripples and a close adherence of a swim cap to the head to reduce drag, as even a slight decrease in drag could make a difference due to a very competitive winning time margins in elite competitions. Nevertheless, despite the small margins, there has been no research undertaken to analyze the effect of different shapes of swim caps as suggested by Marinho et al. (2012) and Harris et al. (2024). In contrast, there has been extensive research on drag and helmet shape in cycling (Alam et al., 2010, 2019; Blair & Sidelko,

Corresponding Author: Peter Olsen, Department of Applied Sciences and Social Practice, Ara Institute of Canterbury, New Zealand, Peter.Olsen@ara.ac.nz.

JSES | https://doi.org/10.36905/jses.2026.02.05

Olsen et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 49-53 (2026) 2010; Crouch et al., 2017; Novak et al., 2019; Underwood et al., 2015). Therefore, the aim of this study was to examine the effects of different shapes of swim caps and hair styles on drag. Research undertaken on swim caps has typically focused on male swimmers where swim caps had a dome shape (Gatta et al., 2013, 2015). However, a dome shape may not represent all swim cap shapes of swimmers, for example, pilot work commenced by Olsen, Alton, et al. (2024) found elite female backstroke competitors in Olympic finals wore their hair in a low or high bun (87.5% and 12.5%, respectively). Therefore, research using a dome shape swim cap on swimming performance or drag may not represent the swim cap shapes used in elite female swimming competitions. Olsen, Alton, et al. (2024) also observed that an elite backstroke swimmer had a swim cap resembling the shape of an aero-helmet or Time Trial (TT) cycle helmet. It is known in cycling that a TT helmet can cause less drag than road or standard helmets due to their aerodynamic shape. To date, there are very limited studies on swim cap shapes. Moreover, Olsen, Oo, et al. (2024) also observed that swimmers in the Paris Olympics were wearing Arena Aquaforce wave swim caps. According to Aquaforce (Arena Sport, n.d.), the wave swim caps have an undulating surface that assists in reducing friction and allows the swimmer to slide more easily through the water (Arena Sport, n.d), where the design reduces water drag significantly for perfect glide (Arena Swimming, n.d.). The first study to examine different swim cap surfaces was conducted by Gatta et al. (2015), who investigated the effect of swim cap composition and structure on passive drag in male swimmers being towed underwater. Overall, swim cap surface (i.e. smooth, dimpled or irregular small wrinkles) did not affect performance, the exception was that the wrinkled swim cap surface showed significantly larger speed-specific drag (4.4%) compared to dimples when swimmers kept their arms alongside the body (Gatta et al., 2015). Pilot work by Olsen, Oo, et al. (2024) in a wind tunnel with a model head fitted with either an Arena Aquaforce (undulating) or Arena 3D (smooth) swim cap in a dome shape, also found a small difference where the undulating swim cap decreased drag by 1.2%. Olsen, Oo, et al. (2024) concluded that the undulating Aquaforce swim cap may confer a slight advantage as winning margins in elite swimming are very small (milliseconds). Nevertheless, there is no research on the effect of different shapes and surfaces of swim caps on drag. Consequently, the aim of this study was to investigate the effect of different swim cap shapes on aerodynamic drag in either a (smooth) Arena high speed 3D or (undulating) Arena Aquaforce surface swim cap condition. 2.

Methods

2.1. Procedure A long-haired manikin head was set up in an open circuit wind tunnel, connected to a load cell (PT Transducers shear beam load cell [3kg], https://www.ptglobal.com/) where PT transducers adjustable software was used to measure the aerodynamic drag at a fixed, horizontal position to replicate the head position in backstroke. The tested open circuit wind tunnels has a crosssection of 0.75 m × 0.75 m, and has a 1.5 m long test section, where the maximum wind speed can reach up to 25 ms −1 . To JSES | https://doi.org/10.36905/jses.2026.02.05

keep a realistic operating condition between the wind and water, the wind speed was set at 10 ms −1 , equivalent to a Reynolds number of 1.32x106, comparable to a 1.98 m swimmer at 1.76 ms −1 in the water (Wei et al., 2014). Since the Reynolds number was matched to provide a comparable flow regime between the wind-tunnel and swimming conditions, the wind-tunnel measurements were used primarily to compare the relative effect of swim cap shape and surface on drag rather than to directly reproduce the hydrodynamic conditions experienced during swimming. Therefore, the flow regime in the wind tunnel is considered similar to what would be observed by swimmers in the water, proven by (Milne et al., 2015; Milne et al., 2010). Prior to testing in the wind tunnel, the long hair on the model head was fashioned into the following hairstyles: Low Bun (LB), High Bun (HB), Pineapple Bun (PB) in Figure 1. Furthermore, to verify the repeatability of the test and the load cell measurements, the reference model was tested 3 times, where the variation was found to be negligible. For consistency, the head was repositioned on a machine spigot and the hair style was commenced by the same person.

LB HB PB Figure 1: Hairstyles in no cap (NC) condition: Low Bun (LB), High Bun (HB) and Pineapple Bun (PB). Figure 2 shows the two swim cap surfaces in the LB hairstyle, where the camera angle in Figure 2b was tilted slightly to show the undulating surface (ripples) of the Aquaforce cap.

2a. Smooth surface 2b. Undulating surface (high speed 3D) (Aquaforce) Figure 2: LB in smooth and undulating swim cap surfaces. Figure 3 shows the surface of the Aquaforce or undulating surface swim cap, in HB, PB, Messy Bun (MB) and Aero Bun (AB). Note: MB and AB involved the researchers molding and smoothing the hair on the model to achieve a rounded-smooth shape over the head or a shape resembling a TT cycle helmet respectively. Once this shape was achieved, a swim cap was placed over the model head. Due to the molding of the hair prior to swim cap placement there was not a NC condition in MB or AB. 50

Olsen et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 49-53 (2026)

HB PB MB AB Figure 3: Undulating cap condition in HB, PB, MB and AB shapes. For every test model, drag data was sampled for 45 s in each swim cap condition at a sampling frequency of 10 Hz. The order of the test conditions was randomised. To ensure consistency among the data, the first 5 s of each dataset were eliminated, and the subsequent 400 measurements (approximately 40 s at 10 Hz) were extracted for comparison between conditions. Each test condition was measured once. To verify the repeatability of the experimental setup and load-cell measurements, the reference model was tested three times, with negligible variation observed. Between test conditions, the head was repositioned on the same machined spigot, and the hairstyles were prepared by the same person to maintain consistency. All data were collected during the same test session, and the research had institutional ethical approval. 2.2. Statistical approach

Results

In the NC condition, HB, LB and PB shapes produced considerably more drag (> 0.147 N, p < 0.001) than all the swim cap shapes and surface conditions (Figure 4). The shape of the swim cap also had a significant effect on drag, with the LB condition producing significantly more drag (> 0.098 N, p < 0.001) than the other swim cap shapes and surface conditions (Figure 4). In contrast, differences between other shapes (HB, PB, MB, and AB) were smaller (0.00147 N to 0.00166 N) in the different surface conditions. Overall, there were conflicting findings in the different shape and surface conditions, for example, the lowest overall drag (0.298 N) was observed in the PB undulating/Aquaforce condition, which had significantly less drag (0.0186, p < 0.05) than the PB smooth condition. Nevertheless, there was no significant difference (0.00147 N, p = 0.06) between the lowest undulating and smooth swim cap conditions (PB undulating vs HB smooth, 0.298 N vs 0.3 N, respectively, Figure 4). The HB and AB smooth conditions produced significantly less drag compared to the HB and AB undulating conditions (approx. 0.0147 N, p < 0.05), whereas there was no significant difference (9.8 × 10-3 N, p = 0.15) between the MB smooth and undulating conditions.

0.769 0.678

0.601

0.426

0.315

0.310

0.325

Messy Bun (undulating)

Aero Bun (smooth)

Aero Bun (undulating)

Low Bun NoCap

0.314

Messy Bun (smooth)

0.408

Low Bun (undulating)

HighPineapple Bun(undulating)

Low Bun (smooth)

0.299

HighPineapple Bun NoCap

0.316

HighPineapple Bun (smooth)

0.315

HighBun NoCap

0.300

HighBun (undulating)

1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00

HighBun (smooth)

Mean drag (N)

The data was exported from the factory provided loadcell software (PT transducers adjustable software) into a readable data sheet and transferred to the Statistical Analysis System v 9.4 (SAS Institute; Cary, NC, USA) for further analysis. The data was then further inspected for any outliers and inaccuracy by investigating the distribution and probability plots. Mean ± SD (standard

deviation) were calculated for drag measurements (N) in each condition for the 400 sample points using a mixed modelling procedure (Proc Mixed) in the Statistical Analysis System. The differences in drag variables were compared between conditions using a repeated measures analysis. Statistical significance was set at p < 0.05. Finally, a Bonferroni correction was applied to account for multiple comparisons.

Swim Cap Condition

Figure 4: Mean (±SD) drag (N) in different swim cap conditions. JSES | https://doi.org/10.36905/jses.2026.02.05

Olsen et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 49-53 (2026) 4.

Discussion

The low bun shape produced approximately 28% more drag than all other swim cap shapes. While there are no other studies on swim cap shape, Cortesi and Gatta et al. (2015) found head position had a substantial effect on passive drag in a towed swimmer, where head up positions increased drag by 5% to 18%. Cortesi and Gatta et al. (2015) suggested the decreased drag in a lower or middle head position in gliding (arms along sides) is due to a smoothing of the flow effect, where the boundary-layer flow shifts from laminar to turbulent or separates from the body. In our study, it was observed that the more horizontal streamlined swim cap shape in the HB, HP, AB and MB conditions resulted in smoother flow, reducing drag, compared to the perpendicular shape of the LB hairstyle/swim cap. In contrast, there was minimal drag difference (0.01N) between the other swim cap shapes. Under the present wind-tunnel conditions, the high, pineapple, messy and aero-bun cap shapes were associated with lower measured drag than the low-bun condition. These findings indicate that the low-bun configuration may be less aerodynamic; however, further in-water testing is required before recommending a specific hairstyle for competitive backstroke. There was a minimal (0.5%) insignificant difference between the swim cap shapes that produced the lowest drag in the undulating or smooth surface conditions (HB and PB respectively). Consequently, the surface of a swim cap appears to have little effect on drag when a model head is positioned as in backstroke. Olsen, Oo et al. (2024) found a small (1.2%) decrease in a dome shaped swim cap with an undulating surface compared to a smooth surface, and Gatta et al. (2015) found smooth, dimpled and irregular small wrinkles on swim cap surfaces had minimal effect on performance. The exception was that the wrinkled swim cap caused significantly larger speed-specific drag (4.4%) compared to dimples. Gatta et al. (2015) and Olsen, Oo et al. (2024) suggested a non-wrinkled and undulating swim cap surface, respectively, may confer a slight advantage, as winning margins in elite swimming are typically very small (i.e. milliseconds). However, in our present study, it is unlikely an undulating surface would give a competitive advantage relative to a smooth surface, in swim cap shapes produced by long hair. An advantage of using a wind tunnel is that it is easier to manipulate head position, drier and probably a more consistent testing environment compared to a swimming pool. Nevertheless, several factors may affect the external validity of our findings. For example, in our study, the long hair on the model head was fashioned into the different hairstyles by the researchers, this process was unwieldly, and it is likely that a swimmer with long hair would be more proficient at producing the different hairstyles or shapes, with a closer adherence of hair to the head, which may create a smoother cap surface. Therefore, we cannot exclude the possibility that a swimmer may be able to create a smoother messy or aero bun, which might produce similar drag to the more streamlined high and pineapple bun shapes. We also cannot discount the possibility that a whole-body model or a swimmer in an aquatic environment may produce different results. Future research could use computer modelling, which will enable a less cumbersome manipulation of different swim cap shapes or test different swim cap shapes in swimming. The researchers also acknowledge that other factors such as body position, training, JSES | https://doi.org/10.36905/jses.2026.02.05

limb coordination, stroke and technical skill have a substantial effect on swimming performance (Cortesi & Gatta, 2015; Zaparo et al., 2019). In conclusion, it is recommended that long hair competitive backstroke swimmers do not wear a low bun when using a swim cap. It is likely that other shapes such as high, pineapple, messy or aero buns produce a smoother and more streamlined swim cap, which will reduce drag. In contrast, swim cap surface (smooth or undulating) appears to have minimal impact on drag on a model head in a backstroke position. Conflict of Interest The authors declare no conflict of interests. Acknowledgment This research was funded by Ara Institute of Canterbury. References Alam, F., Chowdhury, H., & Moria, H. (2019). A review on aerodynamics and hydrodynamics in sports. Energy Procedia, 160, 798–805. Alam, F., Chowdhury, H., Moria, H., & Subic, A. (2010). Aerodynamics of racing bicycle helmets—An experimental study. Proceedings of the 17th Australasian Fluid Mechanics Conference, 1–4. Arena Sport. (n.d.). Aquaforce wave swim cap. Retrieved June 12, 2025, from https://www.arenasport.com/en_row/005371aquaforce-wave-swim-cap.html Arena Swimming. (n.d.). Aquaforce wave cap (white). Retrieved June 12, 2025, from https://arenaswimming.com.au/product/aquaforce-wave-capwhite-101 Blair, K. B., & Sidelko, S. (2008). Aerodynamic performance of cycling time trial helmets. The Engineering of Sport 7, 371– 377. Cortesi, M., & Gatta, G. (2015). Effect of the swimmer's head position on passive drag. Journal of Human Kinetics, 49, 37– 45. https://doi.org/10.1515/hukin-2015-0106 Crouch, T. N., Burton, D., LaBry, Z. A., & Blair, K. B. (2017). Riding against the wind: A review of competition cycling aerodynamics. Sports Engineering, 20, 81–110. Gatta, G., Zamparo, P., & Cortesi, M. (2013). Effect of swim cap model on passive drag. Journal of Strength and Conditioning Research, 27(10), 2904–2908. Gatta, G., Zamparo, P., & Cortesi, M. (2015). Effect of swim cap surface roughness on passive drag. Journal of Strength and Conditioning Research, 29(11), 3253–3259. Harris, G., Alton, L., Hamlin, M., Macdonald, A., & Olsen, P. (2024). The Tokyo Olympics had no soul: A swimming cap controversy. The Journal of Sport & Exercise Science, 8(2), 21–25. Marinho, D. A., Mantha, V. R., Rouboa, A., Vilas-Boas, J. P., Machado, L., Barbosa, T. M., & Silva, A. (2012). The effect of wearing a cap on the swimmer's passive drag. Portuguese Journal of Sport Sciences, 11(Suppl. 2), 319–322. 52

Olsen et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 49-53 (2026) Milne, I., Day, A., Sharma, R., & Flay, R. (2015). Blade loading on tidal turbines for uniform unsteady flow. Renewable Energy, 77, 338–350. Milne, I., Sharma, R., Flay, R., & Bickerton, S. (2010). The role of onset turbulence on tidal turbine blade loads. In Proceedings of 17th Australasian Fluid Mechanics Conference 2010 (pp. 444–447). https://researchrepository.uwa.edu.au/en/publications/the-role-of-onsetturbulence-on-tidal-turbine-blade-loads/ Morales, A. T., Tamayo Fajardo, J. A., & González-García, H. (2019). High-speed swimsuits and their historical development in competitive swimming. Frontiers in Psychology, 10, 2639. https://doi.org/10.3389/fpsyg.2019.02639 Novak, J., Burton, D., & Crouch, T. (2019). Aerodynamic test results of bicycle helmets in different configurations: Towards a responsive design. Proceedings of the Institution of Mechanical Engineers, Part P: Journal of Sports Engineering and Technology, 233(2), 268–276.

JSES | https://doi.org/10.36905/jses.2026.02.05

Olsen, P., Alton, L., Marshall, H., Macdonald, A., Oo, L., Simcic, A., & Harris, G. (2024). The effect of swim cap shape on aerodynamic drag. In J. Smiler, D. Bentley-Gray, J. Drysdall, & J. L. Savage (Eds.), ITP Symposium 2024. Te Pūkenga. https://doi.org/10.34074/proc.2412 Olsen, P., Oo, N. L., Marshall, H., Harris, G., & Alton, L. (2024). The effect of swim cap surface on aerodynamic drag. In J. Smiler, D. Bentley-Gray, J. Drysdall, & J. L. Savage (Eds.), ITP Symposium 2024. Te Pūkenga. https://doi.org/10.34074/proc.2412 Underwood, L., Jermy, M., Eloi, P., & Cornillon, G. (2015). Helmet position, ventilation holes and drag in cycling. Sports Engineering, 18, 241–248. Wei, T., Mark, R., & Hutchison, S. (2014). The fluid dynamics of competitive swimming. Annual Review of Fluid Mechanics, 46, 547–565. Zamparo, P., Cortesi, M., & Gatta, G. (2020). The energy cost of swimming and its determinants. European Journal of Applied Physiology, 120(1), 41–66. https://doi.org/10.1007/s00421019-04270-y

JSES The Journal of Sport and Exercise Science, Vol. 10, Issue 2, 54-60 (2026)

ISSN: 2703-240X

www.jses.net

Physiological and subjective changes in measures of intensity at 60% V̇O₂peak on a lower body positive pressure treadmill Julia Phillips*1, Trace Dominy1, Katherine Tibbs1, Brooke Gilmore2, Thomas Andre1 1 2

Department of Kinesiology and Sport, University of Mississippi, USA Department of Nutrition and Hospitality Management, University of Mississippi, USA ARTICLE INFO Received: 30.04.2026 Accepted: 27.07.2026 Online: 24.10.2026 Keywords: Body-weight support Exercise prescription Intensity Running

ABSTRACT Lower body positive pressure treadmills (LBPPT) offer a variety of benefits and applications within clinical, sport and rehabilitation populations. However, there is limited research examining physiological responses on a LBPPT at a fixed 60% V̇O₂peak at different percent body weight (BW). Twenty aerobically trained participants completed five counterbalanced 5-km runs at various BW conditions: 100% on a regular treadmill (RT) and 100%, 90%, 80% and 70% on a LBPPT. Intensity was fixed at 60% V̇O₂peak and other measures including heart rate, lactate, RPE and rate of energy expenditure were measured across all conditions to examine the effects of different percentages of BW. A linear mixedeffects model with pairwise comparisons was used to examine the changes in all parameters. The results showed no significant differences in blood lactate (p = 0.224). A significant main effect of condition was observed for RPE (p = 0.024) and rate of energy expenditure (p = 0.005); however, no differences were detected between conditions following Tukey adjustment. Significant differences in HR were seen across various conditions (p < 0.001). These findings indicate that relative metabolic intensity appears to be maintained across varying levels of body-weight support despite substantial differences in running velocity and mechanical loading. However, heart rate responses differed between conditions, suggesting that heart rate alone may not accurately reflect exercise intensity during LBPPT running.

Introduction

Several measurement and monitoring tools have been proposed to quantify training stress, intensity, and recovery for runners. Internal load measurements previously utilized are rating of perceived exertion (RPE), blood lactate concentration, heart rate (HR) and training impulse (Halson, 2014). Although weekly mileage is commonly utilized by runners to quantify workload, it fails to account for training intensity. Training load is modulated by exercise intensity, which subsequently influences external loads (pace and duration) and internal loads (HR and RPE) (Paquette et al., 2020). The lower body positive pressure treadmills (LBPPT) offer a unique advantage by allowing runners to exercise while minimizing the stress they would experience

during over-ground or regular treadmill running. Previous research has examined chronic differences in physiological changes across different percentages of body weight (% BW) Notably, only one prior study matched relative intensity across reduced % BW. Runners showed no change in aerobic fitness after 8 weeks of training at 75% or 50% BW compared to the 100% BW, which had slight increases while training at a fixed HR to reflect 60% maximal oxygen consumption (V̇O₂max) (Lee et al., 2021). This suggests that the LBPPT could be a plausible option to maintain fitness during steady-state submaximal running training sessions when appropriately prescribed. A fixed Intensity of 60% V̇O₂peak was selected based on previous research by Lee et al. on the LBPPT and because it represents a moderate-intensity workload that can typically be

Corresponding Author: Julia Phillips, Department of Kinesiology and Sport, University of Mississippi, USA, jpphill4@olemiss.edu

JSES | https://doi.org/10.36905/jses.2026.02.06

Phillips et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 54-60 (2026) sustained for prolonged duration (Lee et al., 2012). It represents a realistic and commonly used intensity for sustained aerobic training and recovery running but also could provide clinical prescription in rehabilitation and return-to-play programs and evaluation physiological responses at differing % BW running. This raises the question of whether heart rate is the most appropriate marker of intensity while running on a LBPPT. Intensity can be quantified using both objective and subjective measures. For example, common aerobic measurements to quantify objective intensity include oxygen consumption (V̇O₂), HR and lactate, while the most common subjective measurement is RPE. Lactate is a common marker of exercise intensity, produced through the glycolytic pathway as a response to exercise (Takeda et al., 2022). At the onset of physical activity, lactate is produced by the working muscles. As exercise intensity increases, blood lactate levels also rise, with a rapid increase typically observed at approximately 50-70% of an individual’s maximal workload (Chmura & Nazar, 2010). When exercising on a LBPPT, the velocity and BW experienced will produce different responses at fixed aerobic intensity, which could lead to varying lactate responses across conditions. An alternative method to measure intensity is utilizing RPE, which is a subjective scale utilized to gauge the strain and effort experienced by the exerciser (Kilpatrick et al., 2020). In elite endurance athletes, RPE has demonstrated strong correlations with heart rate (HR), V̇O₂, and blood lactate, (r = 0.80–0.82) (Losnegard et al., 2021). This understanding of RPE’s relationship with physiological markers may be particularly useful in tracking exercise intensity on a LBPPT, where traditional physiological measures like HR might be less reliable and valid due to the altered loading mechanics and reduced BW. Previous research has shown lactate increases with increases in muscle activation, which increases as a result of velocity on the LBPPT (Mercer et al., 2013). The rectus femoris, biceps femoris and gastrocnemius activation increased with velocity increases on a LBPPT (Mercer et al., 2013). These increases can be attributed to increases in GRF and an increase in impact on landing. Despite the increase in velocity to counter the decreases in BW, some studies report reductions in heart rate and perceived exertion with decreased body weight (Barnes & Janecke, 2017; Farina et al., 2017; Fleckenstein et al., 2020) whereas others suggest that metabolic demand may remain relatively stable when intensity is matched (Gojanovic et al., 2012). Therefore, the purpose of this study was to examine how common measurements of exercise intensity are affected at a fixed 60% V̇O₂peak while varying % BW and velocity across conditions. 2.

Methods

2.1 Participants Twenty aerobically active male (n = 7) and female (n = 13) runners volunteered to participate in the study. Participants were considered aerobically active if they had been running 3 times per week for the past 6 months and not having suffered from any lower-extremity injury in the prior 3 months. All participants signed an informed consent and all procedures were approved by the Institution Review Board (IRB) for human subject research. JSES | https://doi.org/10.36905/jses.2026.02.06

2.2 Procedures Participants arrived at the lab and signed informed consent, filled out a Physical Activity Readiness Questionnaire (PAR-Q) and medical history questionnaire. Participants rested for 5 minutes before baseline heart rate, blood pressure and blood lactate (Lactate Plus; Nova Biomedical, Waltham, MA) were measured. Body weight and height were also collected. Participants were then fitted with a Polar H10 heart rate monitor (Polar, Stamford, CT, USA) and a pair of compression shorts for the LBPPT (Boost 2 Microgravity Treadmill, Boost Treadmills, Russia, OH, USA). Participants were familiarized with the OMNI Pictorial Rate of Perceived Exertion Scale (0-10) (Kilpatrick et al., 2020). For the maximal exercise test, participants wore a Hans Rudolph mask with a two-way nonrebreathing valve for metabolic measurements using a Parvo Medics metabolic measurement system (Parvo Medics, Sandy UT, USA). Participants then underwent a graded exercise test on the LBPPT starting with a 2-minute warm-up at 1.34 m/s (3 mph) and 0% grade, increasing velocity by 0.44 m/s (1 mph) each subsequent minute until reaching 4.02 m/s (9 mph) (Hamlin et al., 2012). Following 1 minute at 4.02 m/s, the treadmill grade increased by 2% per minute until voluntary exhaustion. A valid V̇O₂peak was defined by meeting at least two of the following: RER > 1.10, 95% of age-predicted heart rate, and/or RPE ≥ 9 (Knaier et al., 2019). RPE and HR were recorded every minute during the exercise test. Immediately after the V̇O₂peak test, blood lactate level was measured. Visits 2–6 were performed in a randomized and counterbalanced order of unweighted on a regular treadmill (RT) (Trackmaster treadmill; Full Vision, Inc., Newton, KS, USA), 100%, 90% 80% and 70% BW on the LBPPT. Participants would come into the lab and rest for 5 minutes before blood pressure, heart rate and blood lactate were measured. Participants were equipped with the heart rate monitor, boost shorts for the LBPPT conditions and the Hans Rudolph mask. Participants did a 5minute self-selected warm-up on the treadmill on the condition they were going to perform the 5-km run at. Following the warmup, treadmill velocity was adjusted over a 5-minute period to achieve a target exercise intensity of 60% V̇O₂peak. After 5 minutes, the participants’ 5-km run began. Post-run RPE and lactate were measured and recorded. The average heart rate for 30-second intervals and total caloric expenditure of the 5-km run were recorded from the Parvo metabolic system output (Weir, 1949). To achieve the target intensity of 60% V̇O₂peak, treadmill velocity was individually adjusted for each BW condition. During each trial, participants’ V̇O₂ was monitored throughout the run to ensure they achieved approximately 60% V̇O₂peak under the given BW condition. If the participant fell outside a ± 5% range of 60% V̇O₂peak for four consecutive 30-second average readings, velocity was adjusted. Because BW support alters running mechanics and metabolic cost, running velocity differed across conditions despite the same relative metabolic intensity target. 2.3 Statistical approach All statistical analyses were conducted using R (version 2026.01.2; R Foundation for Statistical Computing, Vienna, 55

Phillips et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 54-60 (2026) Austria). Linear mixed-effect models were used to examine differences in RPE, HR, lactate and energy expenditure per minute across all BW conditions while controlling for V̇O₂ as a covariate and subject included as a random intercept to account for within-subject repeated measures. V̇O₂ was included as a covariate within the linear mixed-effects models to account for small residual variations in oxygen consumption across trials, despite efforts to maintain intensity within ± 5% of 60% VO₂peak. This approach allowed for more precise estimation of condition effects independent of minor fluctuations in V̇O₂. Descriptive statistics, including adjusted means with 95% confidence intervals were calculated for each BW condition. Figures represent the model-adjusted estimated marginal means with 95% confidence for each variable. When there was a significant main effect of the BW condition, pairwise comparisons were performed using Tukey-adjusted post hoc test. Model assumptions were evaluated through residual diagnostics, and no substantial violations were observed. Statistical significance was assessed at an alpha level of 0.05. 3.

Results

Mean V̇O₂ values were consistent across all conditions (29.0–30.7 mL·kg⁻¹·min⁻¹), remaining within approximately ± 1.8 mL·kg⁻¹·min⁻¹ of the target 60% V̇O₂peak (29.0 mL·kg⁻¹·min⁻¹), indicating that relative exercise intensity appeared to be successfully maintained shown in Table 1. Table 1: Running velocity, completion time, and oxygen consumption during 5-km running at 60% V̇O₂peak. Condition

60% V̇O₂peak (mL·kg⁻¹·min⁻¹)

Time (min)

Velocity (m/s)

100% RT

30.7 ± 3.2

33.8 ± 4.8

2.51 ± 0.34

100% LBPPT

30.5 ± 3.8

31.8 ± 4.0

2.77 ± 0.38

90% LBPPT

30.2 ± 4.8

30.3 ± 4.6

2.78 ± 0.42

80% LBPPT

29.6 ± 3.5

28.0 ± 4.5

3.05 ± 0.47

70% LBPPT 29.0 ± 3.6 24.6 ± 3.3 3.44 ± 0.43 Notes: Mean ± SD reported; LBPPT = lower body positive pressure treadmill; RT = regular treadmill. A significant main effect of condition was observed for heart rate (F (4, 72) = 10.43, p < 0.001, ηp2 = 0.367). Additionally, V̇O₂ was a significant covariate (F (1, 72) = 8.63, p = 0.004). β values represent estimated marginal mean differences derived from the linear mixed-effects model. Heart rate was significantly higher for the 100% LBPPT (166  3 bpm) condition compared to the 100% on the regular treadmill (160  3 bpm; : 6 bpm; p = 0.004), 80% LBPPT (159  3 bpm; : 7 bpm; p = 0.002) and 70% LBPPT (158  3 bpm; : 8 bpm; p < 0.001). Heart rate was also significantly higher for the 90% LBPPT condition compared to the 70% LBPPT (90% LBPPT: 164  3 v 70% LBPPT: 158  3 bpm; : 6 bpm; p = 0.016). These differences in heart rate across body-weight conditions are illustrated in Figure 1. JSES | https://doi.org/10.36905/jses.2026.02.06

No significant main effect of condition was observed for blood lactate concentration (F (4, 75) = 1.46, p = 0.224, ηp2 = 0.072) (Figure 2). V̇O₂ was not a significant covariate (F (1,75) = 1.01, p = 0.319). Adjusted mean lactate concentrations ranged from 1.69 ± 0.42 mmol·L⁻¹ in the 70% LBPPT condition to 2.99 ± 0.42 mmol·L⁻¹ in the 100% RT condition. Tukey-adjusted post hoc comparisons revealed no significant pairwise differences between conditions (all p ≥ 0.126). A significant main effect of condition was observed for RPE (F (4, 75) = 2.998, p = 0.024, ηp2 = 0.138). Adjusted mean RPE was highest during the 100% LBPPT condition (5.4 ± 0.3 AU; 95% CI: 4.7–6.1 AU) and lowest during the 70% LBPPT condition (4.3 ± 0.3 AU; 95% CI: 3.6–5.0 AU). V̇O₂ was not a significant covariate (F (1, 75) = 2.02, p = 0.159). Although a significant main effect of condition was observed, Tukey-adjusted post hoc comparisons did not reveal significant pairwise differences between conditions (all p ≥ 0.052). The largest difference was observed between the 100% LBPPT and 70% LBPPT conditions (β = 1.1 AU, p = 0.052) (Figure 3). For rate of energy expenditure, a significant main effect of condition was observed for energy expenditure (F (4, 75) = 3.998, p = 0.005, ηp2 = 0.176); however, no pairwise differences were detected following Tukey adjustment (all p ≥ 0.918). Adjusted rate of energy expenditure was similar across conditions (9.31– 9.45 kcal·min⁻¹), suggesting that relative metabolic intensity appeared to be maintained across body-weight levels (Figure 4). 4.

Discussion

The purpose of this study was to examine the difference in physiological and subjective responses during a 5‑km run performed at a fixed relative metabolic intensity of 60% V̇O₂peak across varying % BW conditions. The primary finding was that blood lactate did not significantly differ across conditions, while significant main effects of condition were observed for RPE and rate of energy expenditure; however, no pairwise differences remained significant following Tukey adjustment. Heart rate was strongly influenced by condition, with decreases in % BW associated with lower HR responses across LBPPT conditions. Although relative metabolic intensity was standardized at approximately 60% V̇O₂peak, mechanical workload differed across conditions because higher levels of body-weight support required progressively faster running velocities to maintain the target intensity. Completion time decreased from on average 33.8 min during the 100% RT condition to 24.6 min during the 70% LBPPT condition, while running velocity increased (Table 1). Despite these differences in external workload, blood lactate concentrations were not significantly different across conditions, while the rate of energy expenditure remained similar across conditions despite a significant main effect, with no pairwise differences detected. These findings support the effectiveness of using V̇O₂-guided velocity adjustments to help maintain a comparable aerobic training stimulus during LBPPT running. When examining lactate across several days, the intraclass correlation coefficients (ICC) have been shown to range from 0.98 to 0.99 for the velocity at lactate threshold and 0.91 to 0.96 for V̇O₂ at lactate threshold (Pfitzinger & Freedson, 1998). During exercise on a LBPPT, the velocity and BW experienced were different despite a fixed aerobic intensity; however, these differences did not 56

Phillips et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 54-60 (2026)

Figure 1: Mean heart rate (bpm) across all body-weight conditions during a 5-km run at 60% V̇O₂peak (n = 20). Values are presented as mean ± 95% confidence intervals. LBPPT = lower body positive pressure treadmill; RT = regular treadmill. *p < 0.05.

Figure 2: Blood lactate concentrations (mmol·L⁻¹) across all body-weight conditions during a 5-km run at 60% V̇O₂peak (n = 20). Values are presented as mean ± 95% confidence intervals. LBPPT = lower body positive pressure treadmill; RT = regular treadmill.

Figure 3: Rating of perceived exertion (RPE; AU) across all bodyweight conditions during a 5-km run at 60% V̇O₂peak (n = 20). Values are presented as mean ± 95% confidence intervals. LBPPT = lower body positive pressure treadmill; RT = regular treadmill.

Figure 4: Rate of energy expenditure (kcal·min⁻¹), across all body-weight conditions during a 5-km run at 60% V̇O₂peak (n = 20). Values are presented as mean ± 95% confidence intervals. LBPPT = lower body positive pressure treadmill; RT = regular treadmill.

JSES | https://doi.org/10.36905/jses.2026.02.06

Phillips et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 54-60 (2026) result in meaningful changes in lactate responses across conditions (Figure 2). Although lactate values did trend slightly downward, there was no significant difference across % BW conditions. These results differ when compared to underwater treadmill running. Garner et al. (2014) observed no significant differences in running velocity and lactate concentration but significantly lower V̇O₂ and HR values. Plausibly their results reflect lower energy expenditure due to the decreased effective BW in water compared to on land (Garner et al., 2014). When comparing the depth of the water (iliac crest vs xiphoid), those in the water immersed to their iliac crest experienced significantly higher V̇O₂ and HR responses by 13.5% and 8.1% while the lactate concentration was not affected by the water depth across all intensities (Benelli et al., 2014). Similar to the previously mentioned findings, reductions in % BW on the LBPPT in the present study did not significantly alter blood lactate concentrations despite substantial differences in running velocity (Fleckenstein et al., 2020). Unlike underwater treadmill running, LBPPT running does not introduce external resistance that the lower extremities must overcome. In contrast, increasing water depth increases resistance and leads to greater activation of the biceps femoris and tibialis anterior (So et al., 2022). When comparing running and walking at the same velocities (1.94 m/s), blood lactate levels were not different between the two conditions, however running did produce greater HR and V̇O₂ and lower muscle oxygenation saturation (Stathopoulos et al., 2024). Uniquely on the LBPPT, HR decreased with a decrease in % BW despite the increase in velocity. Previous research examining RPE as a more cost-effective modality of objective intensity measures found strong correlations between RPE and heart rate, V̇O₂, and blood lactate in elite endurance athletes (r = 0.80-0.82) (Losnegard et al., 2021). At a fixed intensity, the present study showed no significant differences in lactate, while RPE demonstrated a significant main effect of condition; but no pairwise differences were observed across % BW conditions. This potentially reflects the validity of RPE prescribing exercise intensities at a fixed intensity across different % BW conditions since RPE responses remained relatively similar across conditions, despite a significant main effect, similarly to what has been seen on a regular treadmill (Stoudemire et al., 1996). This understanding of RPE's relationship with physiological markers can be particularly useful in tracking exercise intensity on a LBPPT, where traditional measures like HR might be less valid due to the altered mechanics and reduced BW. The observed differences in HR across LBPPT conditions may be partially explained by variations in exercise duration and reductions in effective BW. As % BW decreased, mean HR decreased despite participants maintaining a similar relative metabolic intensity. Furthermore, differences in exercise duration across conditions may have influenced HR responses (Figure 1). One potential explanation is cardiac drift, which commonly occurs during exercise longer than 15 minutes in duration at moderate intensity and is associated with stroke volume gradually decreasing and a compensatory increase in HR to maintain cardiac output (Laginestra et al., 2023). Because completion time decreased from 33.8 min during the 100% RT condition to 24.6 min during the 70% LBPPT condition, differences in exercise duration may have contributed to the lower HR responses observed at decreased % BW. JSES | https://doi.org/10.36905/jses.2026.02.06

The highest heart rate was observed during the 100% LBPPT condition, which was significantly greater than the 100% RT, 80% LBPPT, and 70% LBPPT conditions. Interestingly, the 100% RT and 100% LBPPT conditions involved similar body weight and comparable exercise duration, yet heart rate remained significantly higher during the LBPPT trial. This suggests an additional potential contributing factor may be due to thermoregulation. The LBPPT creates a seal around the lower extremities, which may trap generated heat and water vapour produced from the runners, thereby increasing cardiovascular strain and heart rate relative to the regular treadmill condition. Previous research has reported environmental temperature within the LBPPT treadmill increased by ~8C compared to a RT (Hokanson et al., 2020). As body temperature increases, there will be an increase in blood flow to the skin to facilitate cooling via sweating. This redistribution in peripheral blood flow reduces venous return subsequently leading to a reduction in stroke volume, increasing HR to maintain cardiac output (Lafrenz et al., 2008). Collectively, these mechanisms may partially explain the observed HR responses across conditions. Therefore, heart rate alone may not accurately reflect exercise intensity during LBPPT running and should be interpreted alongside perceptual and metabolic markers. 4.1 Practical application LBPPT treadmills are frequently used during rehabilitation, return-to-play, running programs, and endurance training to reduce mechanical loading while maintaining aerobic fitness. The present findings suggest that relative metabolic intensity appears to be maintained across varying levels of body-weight support when running velocity is adjusted to achieve a target V̇O₂. Despite significant differences in heart rate across conditions, blood lactate concentration, perceived exertion, and the rate of energy expenditure remained similar across conditions. Therefore, practitioners should exercise caution when using heart rate as the sole marker of exercise intensity during LBPPT running and consider incorporating perceptual measures such as RPE alongside heart rate to guide exercise prescription. These findings support the use of LBPPT systems as a means of reducing mechanical loading while preserving a comparable aerobic training stimulus to V̇O₂peak. This study is not without limitations. One limitation is that only aerobically trained individuals were utilized at a moderate intensity (60% V̇O₂peak), which may limit the applicability of these findings to untrained or clinical populations or at lower or higher intensities. Similarly, only one distance was analyzed (5 km) and responses may vary across shorter or longer durations. Additionally, all testing was conducted in a controlled laboratorybased setting, which may not be feasible in a real-world setting. Future research can be expanded by examining these responses at various intensities and across various populations to determine generalizability of these results. Conflict of Interest This study received funding support from Boost Treadmills. However, the sponsor had no role in the study design, data collection, data analysis, interpretation of the results, manuscript 58

Phillips et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 54-60 (2026) preparation, or decision to publish. The authors declare no other conflicts of interest. Acknowledgment The authors would like to thank Boost Treadmills for their funding support for the current investigation. References Barnes, K. R., & Janecke, J. N. (2017). Physiological and biomechanical responses of highly trained distance runners to lower-body positive pressure treadmill running. Sports Medicine - Open, 3(1), 41. https://doi.org/10.1186/s40798017-0108-x Benelli, P., Colasanti, F., Ditroilo, M., Cuesta-Vargas, A., Gatta, G., Giacomini, F., & Lucertini, F. (2014). Physiological and biomechanical responses to walking underwater on a nonmotorised treadmill: effects of different exercise intensities and depths in middle-aged healthy women. Journal of Sports Sciences, 32(3), 268–277. https://doi.org/10.1080/02640414.2013.824601 Chmura, J., & Nazar, K. (2010). Parallel changes in the onset of blood lactate accumulation (OBLA) and threshold of psychomotor performance deterioration during incremental exercise after training in athletes. International Journal of Psychophysiology, 75(3), 287–290. https://doi.org/10.1016/j.ijpsycho.2009.12.011 Farina, K. A., Wright, A. A., Ford, K. R., Wirfel, L. A., & Smoliga, J. M. (2017). Physiological and biomechanical responses to running on lower body positive pressure treadmills in healthy populations. Sports Medicine, 47(2), 261–275. https://doi.org/10.1007/s40279-016-0581-2 Fleckenstein, D., Ueberschär, O., Wüstenfeld, J., & Wolfarth, B. (2020). Physiological and metabolic reaction to lower bodypositive pressure treadmill running. Deutsche Zeitschrift Für Sportmedizin, 71(1), 11–18. https://doi.org/10.5960/dzsm.2019.405 Garner, R., Wagner, D., Bressel, E., & Dolny, D. G. (2014). Land versus water treadmill running: Lactate threshold. International Journal of Aquatic Research and Education, 8(1). https://doi.org/10.25035/ijare.08.01.03 Gojanovic, B., Cutti, P., Shultz, R., & Matheson, G. O. (2012). Maximal physiological parameters during partial body-weight support treadmill testing. Medicine & Science in Sports & Exercise, 44(10), 1935–1941. https://doi.org/10.1249/MSS.0b013e31825a5d1f Halson, S. L. (2014). Monitoring training load to understand fatigue in athletes. Sports Medicine, 44(S2), 139–147. https://doi.org/10.1007/s40279-014-0253-z Hamlin, M. J., Draper, N., Blackwell, G., Shearman, J. P., & Kimber, N. E. (2012). Determination of maximal oxygen uptake using the Bruce or a novel athlete-led protocol in a mixed population. Journal of Human Kinetics, 31, 97–104. https://doi.org/10.2478/v10078-012-0010-z Hokanson, J. F., Schumann, A., & Austin, C. (2020). Stable body temperature in trained runners despite localized heating within alter GTM treadmill. International Journal of Human JSES | https://doi.org/10.36905/jses.2026.02.06

Movement Science, 14(1), 81–93. https://doi.org/10.23949/ijhms.2020.06.14.1.7 Kilpatrick, M., Newsome, A., Foster, C., Robertson, R., & Green, M. (2020). Scientific rationale for RPE use in fitness assessment and exercise participation. ACSM’S Health & Fitness Journal, 24(4), 24–30. https://doi.org/10.1249/FIT.0000000000000587 Knaier, R., Niemeyer, M., Wagner, J., Infanger, D., Hinrichs, T., Klenk, C., Frutig, S., Cajochen, C., & Schmidt-Trucksäss, A. (2019). Which cutoffs for secondary V˙O2max criteria are robust to diurnal variations? Medicine & Science in Sports & Exercise, 51(5), 1006–1013. https://doi.org/10.1249/MSS.0000000000001869 Lafrenz, A. J., Wingo, J. E., Ganio, M. S., & Cureton, K. J. (2008). Effect of ambient temperature on cardiovascular drift and maximal oxygen uptake. Medicine & Science in Sports & Exercise, 40(6), 1065–1071. https://doi.org/10.1249/MSS.0b013e3181666ed7 Laginestra, F. G., Berg, O. K., Nyberg, S. K., Venturelli, M., Wang, E., & Helgerud, J. (2023). Stroke volume response during prolonged exercise depends on left ventricular filling: evidence from a β-blockade study. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology, 325(2), R154–R163. https://doi.org/10.1152/ajpregu.00293.2022 Lee, D. K., Maltz, N. A., Becker, J., & Cotter, J. A. (2012). Training on a lower body positive pressure treadmill with body weight support does not improve aerobic capacity. International Journal of Exercise Science, 14(4), 829839. Lee, D. K., Maltz, N. A., Becker, J., & Cotter, J. A. (2021). Training on a lower body positive pressure treadmill with body weight support does not improve aerobic capacity. International Journal of Exercise Science, 14(7), 829–839. Losnegard, T., Skarli, S., Hansen, J., Roterud, S., Svendsen, I. S., R. Rønnestad, B., & Paulsen, G. (2021). Is rating of perceived exertion a valuable tool for monitoring exercise intensity during steady-state conditions in elite endurance athletes? International Journal of Sports Physiology and Performance, 16(11), 1589–1595. https://doi.org/10.1123/ijspp.2020-0866 Mercer, J. A., Applequist, B. C., & Masumoto, K. (2013). Muscle activity while running at 20%–50% of normal body weight. Research in Sports Medicine, 21(3), 217–228. https://doi.org/10.1080/15438627.2013.792084 Paquette, M. R., Napier, C., Willy, R. W., & Stellingwerff, T. (2020). Moving beyond weekly “distance”: Optimizing quantification of training load in runners. Journal of Orthopaedic & Sports Physical Therapy, 50(10), 564–569. https://doi.org/10.2519/jospt.2020.9533 Pfitzinger, P., & Freedson, P. (1998). The reliability of lactate measurements during exercise. International Journal of Sports Medicine, 19(05), 349–357. https://doi.org/10.1055/s2007-971929 So, B. C. L., Kwok, M. M. Y., Fung, V. C. Y., Kwok, A. H. Y., Lau, C. W. C., Tse, A. L. Y., Wong, M. S. Y., & Mercer, J. A. (2022). A study comparing gait and lower limb muscle activity during aquatic treadmill running with different water depth and land treadmill running. Journal of Human Kinetics, 82, 39–50. https://doi.org/10.2478/hukin-2022-0030 Stathopoulos, A., Petridou, A., Kantouris, N., & Mougios, V. (2024). A comparison of leg muscle oxygenation, 59

Phillips et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 54-60 (2026) cardiorespiratory responses, and blood lactate between walking and running at the same speed. Sports, 12(2), 48. https://doi.org/10.3390/sports12020048 Stoudemire, N. M., Wideman, L., Pass, K. A., McGinnes, C. L., Gaesser, G. A., & Weltman, A. (1996). The validity of regulating blood lactate concentration during running by ratings of perceived exertion. Medicine & Science in Sports & Exercise, 28(4), 490–495. Takeda, R., Nonaka, Y., Kakinoki, K., Miura, S., Kano, Y., & Hoshino, D. (2022). Effect of endurance training and PGC-1α

JSES | https://doi.org/10.36905/jses.2026.02.06

overexpression on calculated lactate production volume during exercise based on blood lactate concentration. Scientific Reports, 12(1), 1635. https://doi.org/10.1038/s41598-022-05593-1 Weir, J. B. de V. (1949). New methods for calculating metabolic rate with special reference to protein metabolism. The Journal of Physiology, 109(1–2), 1–9. https://doi.org/10.1113/jphysiol.1949.sp004363

JSES The Journal of Sport and Exercise Science, Vol. 10, Issue 2, 61-69 (2026)

ISSN: 2703-240X

www.jses.net

Validity of a bat handle-mounted sensor for measuring bat speed and the between-day reliability of a bat speed assessment protocol in professional cricketers Thomas A Brazier1, Zainab Altai1,2, Sean GJ Hardy3,4, James Fleming5, Sam Callaghan6, Louis Howe1, Jamie Tallent1,7* 1

School of Sport, Exercise and Rehabilitation Sciences, University of Essex, UK Institute of Public Health and Wellbeing, University of Essex, UK 3 School of Health Sciences; Faculty of Medicine and Health; The University of Sydney, Australia 4 Cricket New South Wales, Australia 5 Middlesex County Cricket Club, UK 6 School of Biomedical Science and Pharmacy, University of Newcastle, Australia 7 Department of Physiotherapy, Faculty of Medicine, Nursing and Health Science, School of Primary and Allied Health Care, Monash University, Australia 2

ARTICLE INFO Received: 31.03.2026 Accepted: 08.09.2026 Online: 27.10.2026 Keywords: Cricket Power Hitting Inertial Measurement Unit Bat Speed

ABSTRACT Bat speed underpins power-hitting and scoring in cricket. This study assessed the criterion validity of a bat-handle inertial measurement unit (IMU; Blast®) for measuring bat speed and the between-day reliability of an assessment protocol designed to measure maximal bat speed. Experiment 1: five cricketers performed three shots (pull, slog, cover drive), with bat speed measured concurrently using the Blast® sensor and 3D motion capture; sensorto-sensor agreement was also evaluated using two Blast® units mounted on the same bat. Experiment 2: eighteen professional male cricketers completed two testing sessions comprising tee-based and overarm-feed (overarm throw) hitting to quantify between-day reliability. In Experiment 1, agreement between Blast® and 3D motion capture was strong, and sensor-to-sensor agreement was excellent (overall ICC = 0.99; CV = 0.8%). In Experiment 2, between-day reliability was excellent for tee-based testing (ICC = 0.94–0.98; CV = 0.9–1.9%). For the overarm-feed condition reliability was lower (ICC = 0.79–0.86) and variability higher (CV = 1.9–3.9%). The Blast® bat-handle sensor provides a valid measure of bat speed for the slog shot. Despite statistically significant differences for the pull shot and cover drive compared to 3D motion capture, overall agreement and excellent reliability support its use for monitoring bat speed across stroke types. Greater sensitivity is observed during tee-based testing, with a more cautious interpretation required under throwing conditions.

Introduction

Technological advancements have increasingly been applied to quantify key elements of performance in hitting sports (Mizels et al., 2022). TrackMan (Doppler radar) and GCQuad (quadrascopic camera) are widely used to measure clubhead speed, ball speed and launch angle, from which carry distance can be estimated. These variables are strongly associated with golf performance,

including handicap, ranking, and earnings (Read et al., 2013). Validation studies have demonstrated that these radar-based devices provide accurate measurements when compared to gold standard three-dimensional (3D) motion capture systems (Leach et al., 2017). Similar approaches have been employed in other sports involving striking implements. In tennis, inertial measurement units (IMUs) mounted at the racquet base have demonstrated high

Corresponding Author: Jamie Tallent, Sport Rehabilitation and Exercise Science, University of Essex, UK, jamie.tallent@essex.ac.uk

JSES | https://doi.org/10.36905/jses.2026.02.07

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) agreement for racquet velocity during groundstrokes and serves when compared with 3D motion capture and high-speed video (Lin's concordance correlation coefficient = 0.95–0.99) (Delgado García et al., 2020). In baseball, an IMU-based Blast® sensor has been used to quantify bat speed, showing moderate-to-high agreement with 3D motion capture during batting swings (intraclass correlation coefficient [ICC] = 0.67; Pearson's r = 0.91) (Morishita & Jinji, 2022). However, measurement accuracy declined for some devices at swing speeds exceeding 35 m/s (78 mph) (Morishita & Jinji, 2022). While these technologies enable detailed analysis of sport-specific movement patterns, the accuracy of derived metrics may be influenced by sensor placement, sampling frequency, and implement velocity (Lyu & Smith, 2018; Morishita & Jinji, 2022). Therefore, device validation should be sport-specific to support robust data interpretation in applied settings. Bat speed is a key contributor to boundary hitting and overall scoring performance in cricket (Peploe et al., 2019), with winning teams hitting approximately 20% more sixes than losing teams (Bhardwaj & Dwyer, 2022; Petersen et al., 2008). This has led to detailed analyses using 3D motion capture to identify technical determinants associated with bat speed and power-hitting ability (Peploe et al., 2019). While informative, these laboratory-based approaches are time-consuming, not easily portable, expensive, and require specialist expertise to support data collection and processing, making them less suitable for use in applied or fieldbased settings. To overcome these limitations, IMUs paired with user-friendly software applications may offer a practical alternative. For example, an IMU-based Blast® sensor has been validated for use in elite women's cricket, with generally small biases ranging from −0.3 m/s to 0.69 m/s (−0.67 to 1.54 mph) across different cricket shots. Precision errors were also modest, with the cut shot (a horizontal-batted stroke played at hip height towards three o'clock, assuming a right-handed batter facing 12 o'clock) showing the greatest precision (0.75 m/s; 1.68 mph) and the straight drive (a vertical-batted stroke played towards 12 o'clock) the least (1.22 m/s; 2.74 mph), with proportional error increasing with higher bat speeds for all strokes except the pull shot (a horizontal-batted stroke played at shoulder height towards nine o'clock) (Freeston et al., 2026). Although these studies validated the Blast® sensor, they did not include sensor-to-sensor comparisons, and bat speeds exceeding 24.18 m/s (54.1 mph) were not assessed. Given that elite male batters have been shown to generate peak bat speeds of approximately 27.3 m/s (61.1 mph) during range hitting tasks (Peploe et al., 2019), this 24.18 m/s (54.1 mph) threshold falls short of the upper limits of power hitting in cricket. This highlights the need for further validation within cricket to capture higher swing speeds and include elite male cohorts, so the sensor can be used across all populations (Freeston et al., 2026; Morishita & Jinji, 2022; Stewart et al., 2021). Between-session reliability of general physical performance tests has been extensively studied across various athletic populations (Dos'Santos et al., 2018; Harding et al., 2026), with ICCs often exceeding 0.90 for measures such as countermovement jump height and sprint speed (Nuzzo et al., 2011). In contrast, the reliability of tests involving sport-specific movements has received less attention. In golf, Bishop et al. (2023) reported excellent between-day reliability for clubhead speed, with ICCs as high as 0.99 and a standard error of measurement JSES | https://doi.org/10.36905/jses.2026.02.07

(SEM) below 0.89 m/s (2.0 mph) when assessed using radarbased tracking systems. These findings suggest that sport-specific measures, when properly standardised, can yield reliable data across sessions. However, to date, no standardised protocol has been established to assess the between-session reliability of bat speed in cricket; without quantifying this variability, practitioners cannot confidently differentiate between true performance shifts and typical testing noise during longitudinal monitoring. This study had two primary aims and was structured as a twopart experiment. First, we assessed the validity of the Blast® motion sensor for measuring bat speed by comparing it with a gold-standard motion capture system. Second, we assessed the between-session reliability of bat speed in cricket using hitting trials from both a batting tee and an overarm feed (throw). A secondary aim was to provide preliminary normative bat speed data from a small sample of professional cricketers. 2.

Methods

2.1. Overall study design This study consisted of two experimental components. The first experiment evaluated the validity of the Blast® sensor (Blast™, Carlsbad, CA, USA) for assessing bat speed by comparing it with a gold-standard motion capture system. This experiment also assessed sensor-to-sensor agreement by comparing bat speed outputs from two identical (same make and model) Blast® sensors collected during the same testing session. The second experiment examined the between-session consistency of bat speed, with participants completing two sessions separated by at least 72 hours. 2.2. Participants A formal a priori sample size calculation was not performed; rather, a convenience sampling method was applied due to the logistical complexities of data collection and the need to recruit highly skilled cricketers. While the participant sample size was limited for Experiment 1 (n = 5), the primary unit of analysis for device validation was the individual swing, yielding 75 total paired observations. This approach and sample size are consistent with established precedents in biomechanical device validation research (Freeston et al., 2026; Lyu & Smith, 2018; Meghji et al., 2019). In Experiment 1, recruitment was designed to ensure swing data were obtained through a broad range of bat speeds so that validity and agreement could be assessed across the measurement range of interest. In Experiment 2, a convenience sample of professional cricketers within an elite applied setting was recruited. In line with the purpose of the study, emphasis was placed on estimating validity and reliability through complementary metrics, including Bland–Altman analysis, intraclass correlation coefficients, coefficients of variation, and measurement error statistics, rather than on hypothesis testing alone. These factors should, however, be considered when interpreting the precision and generalisability of the findings. Five healthy cricket players (3 men, 2 women; 28 ± 5 years; 74.6 ± 10.5 kg; 172.3 ± 5.7 cm), ranging from recreational to professional, volunteered for Experiment 1. For Experiment 2, 18 professional male cricket players (26 ± 4 years; 83.1 ± 8.3 kg; 62

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) 182.9 ± 9.6 cm) were recruited. All participants were informed of the risks and benefits of the study, and written informed consent was obtained prior to participation. Participants were excluded if they had a current musculoskeletal injury or no previous cricket playing experience. Ethical approval was granted by the University of Essex Ethics Committee (ETH2324-0276). 2.3. Procedures Across both experiments, participants performed three common cricket shots: the pull shot (to square of the wicket), the slog shot (between deep midwicket and long-on), and the cover drive. Shot direction was standardised using a clock-face reference for a right-handed batter: the batter was positioned at six o’clock, with twelve o’clock directly in front of the batter. The pull shot was defined as a cross-batted stroke directed towards the eight to nine o’clock region, typically producing an upward ball trajectory after impact. The slog shot was also a cross-batted stroke, directed towards the ten to eleven o’clock region, again producing an upward ball flight. For the cover drive, the batter maintained the same stance but aimed between two and three o’clock using a more vertical bat position, with the ball travelling along the ground after impact. In both experiments, all shots were performed from a batting tee (Fortress™, Wrexham, UK). Tee height was self-selected for each player and kept consistent across trials for a given shot type but was adjusted between shot types as required. Players were permitted one preparatory step into the shot to better replicate a typical training environment. Bat speed was recorded using a Blast® sensor secured to the bat handle using a silicone holder. For the validity and sensor-tosensor agreement phase, two sensors were attached to the same bat to quantify between-sensor measurement error. Bat speed was recorded and displayed in real time via the Blast® Motion application, connected via Bluetooth to an iPad (Apple Inc., Cupertino, CA, USA). 2.4. Experiment 1: Validation and sensor-to-sensor reliability Each participant performed five swings for each of the three selected cricket shots. Prior to testing, participants completed a standardised 15-swing warm-up comprising five progressive swings per shot type, with effort gradually increasing such that the fifth swing of each set was at or near maximal intensity. Following the warm-up, participants performed maximal-effort swings, striking the ball from a tee into a net. Bat speed at impact was recorded for subsequent analysis. A 30 s rest period was provided between swings. The Blast® sensor is an IMU (mass: 8.1 g) containing a triaxial accelerometer with a sampling frequency of 500 Hz and an output range of ±300 g. Sensors were factory-calibrated by the manufacturer prior to commercial distribution and were not recalibrated during this study. The Blast® sensor uses proprietary algorithms to calculate bat speed based on data captured approximately three samples before impact. Briefly, accelerations measured by the sensor, together with user-input values for bat length and mass, are used to estimate bat speed at the “sweet spot”, defined as 15.24 cm from the toe of the bat, based on rigid-body motion principles (Freeston et al., 2026). JSES | https://doi.org/10.36905/jses.2026.02.07

Validation of the Blast® sensor was conducted by comparing bat speed at impact with measurements from a 14-camera 3D motion capture system (Vicon™, Oxford, UK; 200 Hz). Prior to testing, the laboratory was dynamically calibrated using a fivemarker wand and an L-frame to define the global coordinate system. The wand was also used to define the origin of the capture volume, which remained consistent across all testing sessions. For data collection, four retroreflective markers were placed on a single bat (84 cm length, 1.2 kg; Cutting Edge, London United Kingdom) used by all participants. Two markers were positioned on either edge of the bat, 15.24 cm from the toe, with the remaining two placed at the same height on the back of the bat. The ball (Dukes, British Cricket Balls Ltd, London, United Kingdom) was covered with reflective tape of the same material as the retroreflective markers, effectively functioning as a single large marker to enable precise identification of the bat–ball impact point. 2.5. Data processing Following data collection, all trials were processed in Vicon Nexus (v1.12.1, Vicon™, Oxford, UK). Marker trajectories were filtered using a two-way (zero-lag) Butterworth low-pass filter with a cut-off frequency of 15 Hz. A custom MATLAB script (R2023b; MathWorks, Inc., MA, USA) was used to compute bat speed at impact from the marker trajectories. Briefly, the x-, y-, and z-axis trajectories of the four bat-mounted markers were numerically differentiated to obtain marker velocity. The resultant velocity was calculated for each marker and then averaged across the four markers. Bat speed at impact was defined as the mean resultant velocity at three frames prior to the instant at which the ball exhibited a 20 mm change in resultant position, which was used to operationally define bat–ball contact. 2.6. Experiment 2: Between-session reliability Test–retest sessions (Test 1 and Test 2) were separated by a minimum of 72 hours and a maximum of 14 days, depending on player availability around training and match commitments. Participants completed both tests under standardised conditions. Because participants were based at different sites, testing was conducted in different indoor cricket halls; however, all protocols and verbal cues were standardised across both tests. The same bat and ball were used across Test 1 and Test 2 for each participant, and shot order was kept consistent. Participants were familiarised with the full testing protocol during a standardised warm-up, which included five warm-up shots from a static tee for each stroke type. Participants also faced up to three warm-up deliveries from the live feeder to familiarise themselves with pace and length. All instructions and testing procedures were delivered by the same assessor across both sessions. Prior to testing, players were instructed to ‘swing as hard and fast as possible’. Each participant performed six maximal-effort swings of the pull shot, slog shot, and cover drive from a static tee into a net, with 30 s rest between swings. In the second condition, participants faced a live feed delivered by a professional cricket coach holding an ECB Advanced Coach qualification (or equivalent), positioned approximately 18.3 m (20 yards) from the 63

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) batter, consistent with standard throw-down practice in cricket nets. The shot type was predetermined and communicated to the participant prior to each delivery. The coach aimed to deliver the ball at a consistent overarm pace, with each delivery bouncing once before the shot was played. In this condition, players also performed six maximal-effort swings for each shot type. Bat speed for each swing was recorded using the Blast® motion sensor. 2.7. Statistical analysis: Experiment 1 Descriptive statistics (mean ± SD) were calculated for all shot types. A paired-samples t test was used to assess systematic bias between the Blast® sensor and the Vicon system (Atkinson & Nevill, 1998). Pearson's correlation coefficients (r) were calculated to assess the strength of the linear relationship between the bat speeds recorded by the Blast® sensor and the Vicon system. Pearson's correlation coefficients were interpreted as follows: < 0.30 = negligible, 0.30–0.49 = weak, 0.50–0.69 = moderate, 0.70–0.89 = strong, and ≥ 0.90 = excellent (Mukaka, 2012). Bland–Altman analysis was performed to evaluate agreement between the Blast® sensor and the reference 3D motion capture system, both across all shots and for each shot type. ICCs with 95% confidence intervals (CIs) were calculated using a two-way random-effects model (ICC(2,k)) to assess sensor-tosensor agreement between the two Blast® sensors attached to the bat, overall and by shot type. ICC values were interpreted as follows: < 0.50 = poor, 0.50–0.74 = moderate, 0.75–0.90 = good, and > 0.90 = excellent (Koo & Li, 2016). Coefficients of variation (CV) were also calculated to evaluate relative reliability between the Blast® sensors, with a CV ≤ 10% deemed acceptable. All statistical analyses were conducted using JASP statistical software (v0.17.1, University of Amsterdam, The Netherlands). 2.8. Statistical analysis: Experiment 2 Descriptive statistics (mean ± SD) were calculated for all shots. Values exceeding ± 2 SD from the mean were deemed outliers and removed (Fristrup et al., 2024; Measson et al., 2022; Raymond et al., 2018). The mean of the three highest bat speeds for each shot was used for analysis. Paired-samples t tests were used to assess systematic bias (mean difference) between Test 1 and Test 2. Relative reliability between sessions was calculated using a two-way random-effects model (ICC(2,k)), with interpretation consistent with Experiment 1. Relative reliability was also assessed using the CV (%). Absolute reliability was quantified using the SEM, calculated as follows (Atkinson & Nevill, 1998): SEM = SD × √(1 − ICC) The minimum detectable change (MDC) was calculated to aid practitioner interpretation (Riemann et al., 2018): MDC = SEM × 1.96 × √2 A 3 × 2 repeated-measures ANOVA was used to examine the effects of shot type (pull, slog, cover drive) and condition (tee vs. feed) on bat speed. Where appropriate, pairwise comparisons JSES | https://doi.org/10.36905/jses.2026.02.07

were conducted with adjustment for multiple comparisons. All analyses were conducted using JASP statistical software (v0.17.1, University of Amsterdam, The Netherlands). 3.

Results

3.1. Experiment 1 Table 1 presents bat speed (mean ± SD) across all shot types as measured by Blast® Sensors 1 and 2 and the 3D motion capture (Vicon) system. Table 2 summarises agreement and mean differences between Sensors 1 and 2 and Vicon. The pull and cover drive shots demonstrated significant differences between the Blast® sensors and the 3D motion capture system. No significant difference was found for the slog shot or when all shot types were pooled. Table 1: Means and SD of Blast® sensor 1, 2 and Vicon.

Sensor 1 Sensor 2 Vicon

All Shots (m/s) 22.13 ± 2.82 22.08 ± 2.82 22.17 ± 2.32

Pull Shot (m/s) 23.02 ± 2.32 22.98 ± 2.32 21.86 ± 2.59

Slog Shot (m/s) 22.53 ± 3.26 22.49 ± 3.26 23.02 ± 2.01

Cover Drive (m/s) 19.76 ± 1.03 19.71 ± 1.12 20.30 ± 1.30

Table 2: Statistical comparisons of the Blast® sensor vs Vicon. Shot Type

Pull

0.89

<0.001*

Slog

0.86

0.265

Cover Drive

0.55

0.036*

All shots

0.90

0.074

Note: p-values reflect paired-samples t-tests assessing systematic bias between the Blast® sensor and Vicon. *p < 0.05. All between-sensor agreements were categorised as excellent (ICC > 0.9) and within acceptable thresholds (CV < 10%). When comparing Vicon to the Blast® sensor (Figure 1: Panels A–D), the overall mean bias across all shots was minimal (−0.17 m/s [−0.38 mph]; 95% LoA = −2.35 to 2.01 m/s [−5.27 to 4.50 mph]). However, shot-specific systematic bias was evident: the Blast® sensor overestimated bat speed during the pull shot (mean bias = −1.12 m/s [−2.50 mph]; 95% LoA = −3.44 to 1.20 m/s [−7.70 to 2.69 mph]), demonstrated high accuracy for the slog shot (mean bias = −0.21 m/s [−0.46 mph]; 95% LoA = −2.57 to 2.16 m/s [−5.76 to 4.84 mph]), and underestimated it during the cover drive (mean bias = +0.44 m/s [+0.98 mph]; 95% LoA = −1.77 to 2.65 m/s [−3.96 to 5.92 mph]). In contrast, inter-unit agreement between Blast® Sensor 1 and Sensor 2 (Figure 1: Panels E–H) demonstrated negligible mean bias and consistently narrow limits of agreement across all stroke types. 64

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026)

Figure 1: Bland–Altman plots showing agreement in bat speed at impact between criterion and sensor measures, and between sensor units. Panels A–D show Vicon 3D motion capture vs Blast® comparisons for the Pull (A), Slog (B), Cover Drive (C), and All shots combined (D). Panels E–H show Blast® Sensor 1 vs Blast® Sensor 2 comparisons for the Pull (E), Slog (F), Cover Drive (G), and All shots combined (H). The solid horizontal line represents zero difference, the blue dashed line the mean bias, and the red dotted lines the 95% limits of agreement (mean bias ± 1.96 SD). Bat speed is reported in m/s. JSES | https://doi.org/10.36905/jses.2026.02.07

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) 3.2. Experiment 2 Relative and absolute reliability metrics are presented in Table 4. Systematic bias between Test 1 and Test 2 was observed for the cover drive in both tee and feed conditions, and for the slog shot under feed conditions. Relative reliability was excellent for all tee-based shots (ICC ≥ 0.90) and good for all feed-based shots (ICC = 0.75–0.90), with CV (%) ranging from 0.8% to 2.4%. A significant main effect was found for shot type (F(2,13) = 13.43, p = 0.001) and delivery condition (tee vs feed; F(1,13) = 25.87, p = 0.001). Post hoc analyses indicated that the pull and slog shots were faster than the cover drive (both p < 0.05), and that bat speed was higher in the tee condition than the feed condition. No significant (p < 0.05) Shot Type × Condition interaction was observed. Table 3: Statistical comparisons of Blast® sensor 1 vs 2. Shot Type

ICC (95% CI)

CV%

Pull

0.95 (0.93–0.97)

0.8

0.441

Slog

0.98 (0.98–0.99)

0.8

0.629

Cover Drive

0.99 (0.99–0.99)

0.9

0.654

All shots

0.99 (0.99–0.99)

0.8

0.362

Note: p-values reflect paired-samples t-tests assessing systematic bias between the Blast® sensor 1 Vs 2; ICC = Intraclass Correlation Coefficient; CI = Confidence Interval; CV% = Coefficient of Variation. 4.

Discussion

The primary aim of this study was to determine the validity, interunit agreement, and between-session reliability of the Blast® motion sensor for assessing maximal cricket bat speed. Experiment 1 demonstrated that the validity of the sensor is shotdependent; while statistically significant mean differences were observed for the pull shot and cover drive compared to a gold-

standard motion capture system, acceptable practical agreement and excellent sensor-to-sensor reliability were established across all strokes. Experiment 2 indicated good-to-excellent betweensession reliability for protocols using both a batting tee and an overarm feed, with the tee-based protocol demonstrating superior reliability. Collectively, these findings support the use of the Blast® motion sensor, particularly within a tee-based protocol, as a valid and reliable method for quantifying maximal bat speed in elite cricketers. The Blast® sensor demonstrated strong overall criterion validity for measuring bat speed. Mean bias relative to the 3D motion capture system was close to zero across all shots (−0.17 m/s; −0.38 mph), though the 95% limits of agreement (e.g., −3.44 to 1.20 m/s [−7.70 to 2.69 mph] for the pull shot) indicate that individual swings can exhibit wider variability. This level of agreement appears more favourable than has been reported in baseball, where larger systematic differences between the Blast® sensor and 3D motion capture have been observed (e.g., +3.89 m/s; +8.7 mph), alongside only moderate associations (r = 0.54) (Stewart et al., 2021). The present findings therefore support the use of the Blast® sensor for quantifying bat speed in cricket. When compared with prior cricket-specific validation, Freeston et al. (2026) reported a systematic bias of ~0.54 m/s (~1.2 mph) for the pull and slog shots in elite female cricketers. In the present study, mean differences were 1.12 m/s (2.5 mph) (pull) and 0.21 m/s (0.47 mph) (slog), indicating broadly comparable magnitude of bias across studies. Notably, the mean slog-shot bat speed was ~21.01 m/s (~47.0 mph) in elite female cricketers (Freeston et al., 2026) and ~23.25 m/s (~52.0 mph) in the present study, with peak values up to 27.27 m/s (61.0 mph). Crucially, these peak values closely align with the maximal bat speeds of 27.5 m/s (61.5 mph) previously reported in elite male cricketers (Peploe et al., 2019). This confirms that the current validation protocol successfully captures the upper speed bandwidths practically relevant to elite male performance. Both studies reported a strong association for the slog shot (r = 0.86), which may provide practitioners with greater confidence that the sensor can tolerate higher swing speeds in cricket for this shot type, addressing a previously cited limitation (Freeston et al., 2026).

Table 4: Between-session protocol reliability. Shot Type Pull Slog Cover Drive

ICC (95% CI)

CV%

26.5 ± 1.6

Difference Test 1 vs Test 2 (m/s) -0.4

0.9

SEM (m/s) 0.4

MDC (m/s) 1.0

0.112

0.95 (0.83–0.98)

23.5 ± 2.1

23.3 ± 2.1

-0.2

0.128

0.79 (0.45–0.92)

3.1

0.9

2.6

Tee

26.6 ± 2.2

26.4 ± 2.3

-0.2

0.476

0.94 (0.85–0.98)

1.5

0.5

1.5

Feed

24.9 ± 2.0

24.2 ± 1.5

-0.8

0.019*

0.86 (0.51–0.95)

1.9

0.6

1.7

Tee

23.4 ± 2.5

22.9 ± 3.0

-0.5

0.054

0.98 (0.86–0.98)

1.9

0.6

1.6

Feed

22.5 ± 2.9

21.3 ± 1.7

-1.2

0.031*

0.79 (0.35–0.93)

3.9

1.1

3.0

Condition

Test 1 (m/s)

Test 2 (m/s)

Tee

26.8 ± 1.7

Feed

Note: p-values reflect paired-samples t-tests assessing systematic bias between the test 1 Vs 2; ICC = Intraclass Correlation Coefficient; CI = Confidence Interval; CV% = Coefficient of Variation; SEM = Standard Error of Measurement; MDC = Minimal Detectable Change.; Feed = Overarm throw. *p < 0.05.

JSES | https://doi.org/10.36905/jses.2026.02.07

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) However, significant systematic bias was observed for the pull shot and cover drive, suggesting that the sensor may handle the complex wrist and forearm mechanics of these strokes differently from the criterion measure. This was most evident for the cover drive, which showed both systematic bias and a weaker (moderate) association compared with the excellent relationships observed for the pull and slog shots. The reduced validity for the cover drive may reflect its more technically controlled, hands-dominant characteristics and distinct vertical swing path. Unlike the rotational acceleration and distinct wrist uncocking typical of a baseball swing, which the pull and slog shots closely replicate, the cover drive has a substantially different acceleration profile. This smoother kinematic signature could challenge an event-detection algorithm originally calibrated to identify baseball impacts, explaining the results of the current investigation. Although reduced accuracy and wider limits of agreement compared to the gold-standard motion capture system were present for specific shot types, the Blast® sensor demonstrated excellent sensor-tosensor agreement and is an appropriate tool for field-based assessments of bat speed in cricket. To be a practical tool for squad-based monitoring, the Blast® motion sensor must demonstrate high inter-unit agreement (i.e., sensor-to-sensor reliability). To date, this has not been examined in validation studies of the Blast® sensor in cricket or baseball (Freeston et al., 2026; Morishita & Jinji, 2022; Stewart et al., 2021), despite being important when multiple sensors are used within a squad testing battery. The present study addressed this gap and demonstrated excellent agreement between two sensor units (ICC = 0.99; CV = 0.8%). Combined with the strong criterion validity observed, these findings support the sensor's utility for applied monitoring. Practitioners can therefore be confident that data collected from different units are directly comparable, facilitating team-wide assessment and comparison against device-specific normative data. Both testing protocols demonstrated good-to-excellent between-session reliability. The tee-based protocol produced the most consistent results across all three shots (ICC = 0.94–0.98; CV = 0.9–1.9%), whereas the live-feed protocol was more variable, albeit more ecologically valid (ICC = 0.79–0.86; CV = 1.9–3.9%). Facing a live overarm feed more closely replicates the perceptual-motor couplings of competition, thereby enhancing ecological validity (Pinder et al., 2011). However, the inherent variations in human delivery (i.e., differing speed, line, and length), combined with the batter's increased perceptual-cognitive processing demands, naturally elicit greater variability. This may account for the lower reliability observed during the live-feed protocol compared to the highly constrained tee-based condition. Hence these differences directly affect the sensitivity of each test for detecting meaningful change over time. Comparable reliability has been reported for striking-speed measures in other sports, including clubhead speed in golf (ICC = 0.97–1.00) (Bishop et al., 2023) and bat speed in baseball (ICC = 0.93–0.99; CV ≈ 1.9%) (Bailey et al., 2016). The Blast® sensor appears sufficiently sensitive (i.e., possessing a measurement error smaller than expected physical adaptations) to detect changes in cricket stroke performance, as reflected by the MDC values reported here. MDC was lower for the tee condition (0.98–1.60 m/s; 2.2–3.6 mph) than for the livefeed condition (1.74–3.00 m/s; 3.9–6.7 mph), indicating greater sensitivity under the tee-based protocol. Notably, the tee-based JSES | https://doi.org/10.36905/jses.2026.02.07

MDC values were smaller than the MDC reported for clubhead speed using TrackMan in golf (2.06 m/s; 4.6 mph) (Bishop et al., 2023). In golf, 12-week strength and conditioning interventions have been shown to increase clubhead speed by up to ~2.01 m/s (~4.5 mph) (Coughlan et al., 2019). By extension, the relatively low MDC observed for the pull shot (1.03 m/s; 2.3 mph) suggests that practitioners should be able to detect practically meaningful improvements in bat speed when using the tee-based protocol. Therefore, the tee-based protocol appears highly sensitive for detecting performance improvements within individual athletes. In practice, practitioners must balance the higher precision of the tee-based protocol against the greater ecological validity of the live-feed assessment, which may be preferable for skill-focused applications (Bishop et al., 2023; Bliss et al., 2021; Coughlan et al., 2019; Hardy, Stelzer-Hiller, et al., 2025; Morishita & Jinji, 2022; Read et al., 2013; Szymanski et al., 2007). A key finding was that bat speed was significantly higher when hitting from a tee than during live feed. This may reflect the more standardised and constrained nature of tee batting, which likely reduces cognitive and perceptual-motor demands: with no requirement to respond to ball flight, swings can be self-paced and preprogrammed. In contrast, live feed introduces additional visuomotor processing demands and greater sensory variability, which may limit maximal motor output (Honda et al., 2012). Therefore, when designing a field-based testing battery to profile an athlete's absolute maximal bat speed, practitioners should utilise the tee-based protocol to better isolate bat speed. Furthermore, given the slog shot's high reliability and lack of systematic bias against 3D motion capture, it may represent the single most robust stroke for longitudinal monitoring. The present study builds on previous cricket research assessing bat speed (Freeston et al., 2026) by evaluating higher bat speeds, quantifying between-sensor agreement, and examining between-session reliability using both tee-based and live-feed protocols. A limitation of Experiment 1 is the modest participant sample size (n = 5). While the primary unit of analysis was the individual swing, yielding a robust number of paired observations (75 total swings) for kinematic validation, the small number of discrete individuals may limit the broader generalisability of the systematic bias observed across different batting techniques. Consistent with prior work, the effective "sweet spot" location in cricket may vary between bats and players (Freeston et al., 2026; Peploe et al., 2019). The Blast® sensor estimates bat speed at a fixed point (15.2 cm from the toe); however, elite bats are often customised to player specifications, meaning the true optimal impact region may not align with this assumed location. Importantly, when motion capture estimates were derived at the same location on the bat (15.2 cm from the toe), the Blast® sensor provided a valid and repeatable measure of bat speed relative to Vicon across all strokes, despite the previously discussed shot-specific systematic biases. Although this fixed location may not represent the performance-optimal sweet spot for all players, it provides a standardised reference point for monitoring bat speed using a commercially available system and for detecting changes over time. Collectively, these findings provide practitioners with greater confidence when using the Blast® motion sensor to assess bat speed in elite cricketers. This study examined whether a commercially available sensor can be used to assess cricket bat speed. Overall, the Blast® sensor demonstrated acceptable agreement with 3D motion capture and 67

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) good-to-excellent between-session reliability. The MDC values reported here provide practitioners with a practical threshold for interpreting whether observed changes in bat speed are likely to reflect true performance change rather than measurement error. As such, bat speed can be incorporated into applied testing batteries with increased confidence in its ability to detect longitudinal changes in performance. Future work should assess bat speed in larger samples of elite cricketers to establish normative values and to explore relationships between bat speed and relevant physical qualities (e.g., strength, power, and sprint performance). With appropriate normative data and sport-specific context, bat speed may also have potential utility within talent identification and selection processes; however, this application requires further validation linking bat speed to match performance outcomes. Furthermore, while future research could explore the tracking of in-game bat speed to establish links to match performance outcomes, in-game kinematics are heavily confounded by tactical intent and situational constraints. Therefore, the primary applied value of this sensor likely remains in standardised, controlled testing environments. By reliably quantifying an athlete's maximal bat speed ceiling, practitioners can monitor training adaptations that ultimately allow batters to operate at lower, more efficient relative intensities during competitive match play. Conflict of Interest The authors declare no conflict of interests Acknowledgment We would like to thank all participants for taking part in this study. References Atkinson, G., & Nevill, A. M. (1998). Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Medicine, 26(4), 217–238. Bailey, C., McInnis, T., & Batcher, J. (2016). Bat swing mechanical analysis with an inertial measurement unit: Reliability and implications for athlete monitoring. Journal of Trainology, 5(2), 42–45. Bhardwaj, D., & Dwyer, D. B. (2022). Team technical performance in elite men's and women's T20 cricket: Determinants of performance within a match and across a season. International Journal of Performance Analysis in Sport, 22(2), 277–290. Bishop, C., Wells, J., Ehlert, A., Turner, A., Coughlan, D., Sachs, N., & Murray, A. (2023). Trackman 4: Within- and betweensession reliability and inter-relationships of launch monitor metrics during indoor testing in high-level golfers. Journal of Sports Sciences, 41(23), 2138–2143. Bliss, A., Livingstone, H., & Tallent, J. (2021). Field-based and overspeed potentiated warm-ups increase clubhead speed and drive carry distance in skilled collegiate golfers. The Journal of Sport and Exercise Science, 5(2), 107–113. Coughlan, D., Taylor, M. J. D., Wayland, W., Brooks, D., & Jackson, J. (2019). The effect of a 12-week strength and JSES | https://doi.org/10.36905/jses.2026.02.07

conditioning programme on youth golf performance. International Journal of Golf Science, 8(1), 1–14. Delgado-García, G., Vanrenterghem, J., Ruiz-Malagón, E. J., Molina-García, P., Courel-Ibáñez, J., & Soto-Hermoso, V. M. (2020). IMU gyroscopes are a valid alternative to 3D optical motion capture system for angular kinematics analysis in tennis. Proceedings of the Institution of Mechanical Engineers, Part P: Journal of Sports Engineering and Technology, 235(3), 3–12. Dos'Santos, T., Thomas, C., Comfort, P., McMahon, J. J., Jones, P. A., Oakley, N. P., & Young, A. L. (2018). Between-session reliability of isometric midthigh pull kinetics and maximal power clean performance in male youth soccer players. Journal of Strength and Conditioning Research, 32(12), 3364–3372. Freeston, J., Hardy, S. G. J., Ho, E., Sinclair, P., Chalmers, S., Hollings, M., & Andersen, J. T. (2026). Accuracy of a batmounted sensor for the measurement of bat speed among elite female cricket players. Sports Biomechanics, 25(4), 564–575. Fristrup, B., Krustrup, P., Kristensen, K. H., Rasmussen, S., & Aagaard, P. (2024). Test–retest reliability of lower limb muscle strength, jump and sprint performance tests in elite female team handball players. European Journal of Applied Physiology, 124(9), 2577–2589. Harding, J., Tallent, J., Sheriff, K., McCann, C., Cortes, N., Olsson, L., Shaw, J., & Howe, L. (2026). The reliability of physical performance testing within elite adolescent preprofessional ballet dancers. Journal of Dance Medicine & Science, 30(2), 109–118. Hardy, S. G. J., Edwards, K. M., & Freeston, J. (2025). The physical determinants of bat speed in elite female cricketers. International Journal of Sports Physiology and Performance, 20(10), 1426–1433. Hardy, S. G. J., Stelzer-Hiller, O. W., Edwards, K. M., & Freeston, J. (2025). Criterion validity and reliability of a new medicine ball rotational power test. Journal of Strength and Conditioning Research, 39(3), e429–e435. Honda, T., Hirashima, M., & Nozaki, D. (2012). Adaptation to visual feedback delay influences visuomotor learning. PLoS ONE, 7(5), e37900. Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and reporting intraclass correlation coefficients for reliability research. Journal of Chiropractic Medicine, 15(2), 155–163. Leach, R. J., Forrester, S. E., Mears, A. C., & Roberts, J. R. (2017). How valid and accurate are measurements of golf impact parameters obtained using commercially available radar and stereoscopic optical launch monitors? Measurement, 112, 125–136. Lyu, B., & Smith, L. V. (2018). Evaluation of wireless bat swing speed sensors. Sports Engineering, 21(3), 229–234. Measson, M. V., Ithurburn, M. P., & Rambaud, A. J. M. (2022). Intra-rater reliability of a qualitative landing scale for the single-hop test: A pilot study. International Journal of Sports Physical Therapy, 17(3), 493–500. Meghji, M., Balloch, A., Habibi, D., Ahmad, I., Hart, N., Newton, R., Weber, J., & Waqar, A. (2019). An algorithm for the automatic detection and quantification of athletes' change of direction incidents using IMU sensor data. IEEE Sensors Journal, 19(12), 4518–4527. 68

Brazier et al. / The Journal of Sport and Exercise Science, Journal Vol. 10, Issue 2, 61-69 (2026) Mizels, J., Erickson, B., & Chalmers, P. (2022). Current state of data and analytics research in baseball. Current Reviews in Musculoskeletal Medicine, 15(4), 283–290. Morishita, Y., & Jinji, T. (2022). Accuracy and error trends of commercially available bat swing sensors in baseball. Sports, 10(2), 21. Mukaka, M. M. (2012). A guide to appropriate use of correlation coefficient in medical research. Malawi Medical Journal, 24(3), 69–71. Nuzzo, J. L., Anning, J. H., & Scharfenberg, J. M. (2011). The reliability of three devices used for measuring vertical jump height. Journal of Strength and Conditioning Research, 25(9), 2580–2590. Peploe, C., McErlain-Naylor, S. A., Harland, A. R., & King, M. A. (2019). Relationships between technique and bat speed, post-impact ball speed, and carry distance during a range hitting task in cricket. Human Movement Science, 63, 34–44. Petersen, C., Pyne, D. B., Portus, M. J., & Dawson, B. (2008). Analysis of Twenty/20 cricket performance during the 2008 Indian Premier League. International Journal of Performance Analysis in Sport, 8(3), 63–69. Pinder, R. A., Davids, K., Renshaw, I., & Araújo, D. (2011). Representative learning design and functionality of research

JSES | https://doi.org/10.36905/jses.2026.02.07

and practice in sport. Journal of Sport and Exercise Psychology, 33(1), 146–155. Raymond, F., Lussier, B., Dugas, F., Charbonneau, M., Croteau, F., Kennedy, C., & Berryman, N. (2018). Using portable force plates to assess vertical jump performance: A metrological appraisal. Sports, 6(4), 149. Read, P. J., Lloyd, R. S., De Ste Croix, M., & Oliver, J. L. (2013). Relationships between field-based measures of strength and power and golf club head speed. Journal of Strength and Conditioning Research, 27(10), 2708–2713. Riemann, B. L., & Lininger, M. R. (2018). Principles of statistics: What the sports medicine professional needs to know. Clinics in Sports Medicine, 37(3), 375–386. Stewart, E., Stewart, M., Kodithuwakku Arachchige, S. N. K., Turner, A., Burch, R. F., V., Knight, A., Johnson, J., Younger, T., & Chander, H. (2021). Validation of a bat handle sensor for measuring bat velocity, attack angle, and vertical angle. International Journal of Kinesiology and Sports Science, 9(2), 28–32. Szymanski, D. J., Szymanski, J. M., Bradford, T. J., Schade, R. L., & Pascoe, D. D. (2007). Effect of twelve weeks of medicine ball training on high school baseball players. Journal of Strength and Conditioning Research, 21(3), 894– 901.