IPL Match Win Predictor

The Indian Premier League (IPL) is among the largest cricket leagues with extensive statistical data, where estimating match results is difficult because game conditions change continuously. In this study, a predictive learning framework is designed to estimate the winning chances of a team using ball-by-ball data. The proposed system uses important match features such as current run rate, required run rate, wickets remaining, balls remaining, and match phase to capture the real-time match situation. The dataset used for this study consists of historical IPL matches obtained from open cricket datasets collected across several IPL seasons. Multiple predictive learning algorithms, including Logistic Regression, Random Forest, Support Vector Machine (SVM), and XGBoost, are implemented and compared. Among these models, XGBoost provides the highest accuracy of approximately 81%, as it efficiently captures complex feature dependencies and scoring patterns. The results show that the prediction performance improves during the final stages of matches when the outcome becomes more predictable. The proposed system can be used for real-time match prediction and provides useful insights for sports analytics applications.