International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 03 | Mar 2025
p-ISSN: 2395-0072
www.irjet.net
AI-Driven Analysis of Cricket Match Trends under Varying Environmental Conditions Omkar Singh1, Amit Kumar Pandey2, Ferhan Ansari3, Virendra Panchal4, 1HOD of MSc Data Science, 2Assistant Professor, 3,4PG Student (MSc Data Science), Thakur College of Science and
Commerce, Thakur Village, Kandivali (East), Mumbai-400101, Maharashtra, India -----------------------------------------------------------------------***-------------------------------------------------------------------------Abstract: Traditional cricket analytics prioritize player statistics and game strategies, often overlooking key environmental factors such as temperature, humidity, wind speed, and air pressure, which significantly impact match dynamics. In order to examine historical T20 match data under various meteorological circumstances, this study integrates Deep Learning and AI-driven methods. Real meteorological data, such as height, rainfall, and dew point, was combined with match statistics to produce an extensive dataset. Advanced machine learning algorithms evaluated how these parameters affected match results, team performance, and individual efficiency. Important things like how wind speed affects swing bowling, how dew affects batting in the second innings, and how humidity affects player endurance were revealed. Our findings demonstrate that incorporating environmental data significantly enhances the accuracy of match outcome predictions and tactical decision-making. This study lays the groundwork for AI-powered, real-time weather-based cricket decision-making by offering insightful information for pitch reports, team tactics, and in-game projections. These discoveries have the potential to transform sports analytics, which would be advantageous to cricket governing bodies, teams, and analysts.
Index Terms: Cricket Match Trends, Environmental Conditions, Machine Learning, Artificial Intelligence, Sports Analytics, Predictive Modeling.
Introduction: One of the most exciting outdoor sports is cricket, where player performance and match results are greatly influenced by the playing circumstances. Environmental parameters including as temperature, humidity, wind speed, and air pressure have not received enough attention, despite the extensive analysis of pitch conditions, team strengths, and strategy. These environmental factors influence key match components such as ball movement, batting effectiveness, player stamina, and the accuracy of rain-affected match calculations like DLS (Duckworth-Lewis-Stern). However, there is a significant research gap in cricket analytics as it currently mostly uses conventional statistical techniques rather than AI-driven environmental modeling. Limited research has systematically incorporated environmental variables into machine learning-based cricket performance analysis, leaving a significant gap in AI-driven sports modeling. The main research needs are as follows: Inadequate use of AI and deep learning to environmental effect studies in cricket. In spite of differing match structures, generalized models do not distinguish across formats (T20, ODI, and Test). There is a lack of integration between meteorological datasets and cricket performance measures. Predictive algorithms that evaluate how in-game weather decisions (such as bowling first versus second in humid circumstances) are affected in real-time are lacking. In order to examine the influence of past weather conditions on cricket match results, this study suggests a deep learningbased methodology. Developing an AI-powered model that forecasts how weather conditions will affect team and player performance is one of the main accomplishments. Investigating format-specific patterns, especially in T20 cricket, where game changes quickly, and figuring out environmental elements that have a big impact on match-winning odds. Giving analysts, coaches, and captains data-driven insights to help them adjust their plans in light of anticipated weather impacts.
Literature Review: Numerous research has investigated machine learning-based cricket analytics, mostly concentrating on score prediction, team selection, and individual performance analysis. Ridge Regression, XGBoost, and Naïve Bayes are examples of traditional
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 1154