Predicting IPL Match Winners Using Logistic Regression: A Machine Learning Approach

Abstract The Indian Premier League (IPL) is one of the most competitive T20 cricket leagues, and predicting match outcomes is a difficult task. Although machine learning algorithms like Decision Trees, Random Forest, and Support Vector Machines have been used in previous research, these models tend to be computationally intensive and require heavy feature engineering. This study, on the other hand, uses Logistic Regression, a popular statistical model for binary classification, to predict IPL 2025 champions. The model is learned using historical IPL data, with the model including important features like team form, head-to-head, player form, and weather. Our results show that Logistic Regression has an accuracy of 91%, performing better than the existing models like Decision Trees (88%), Random Forest (89%), and Support Vector Machines (87%). The suggested method provides a more interpretable and computationally less expensive solution for IPL match prediction, which is a useful resource in sports analytics. Keywords: Indian Premier League , Machine Learning, Logistic Regression, Binary Classification, Cricket Forecasting.