Batsmen Performance Prediction Using Regression Models

Cricket is among the most widely practiced and loved team sports worldwide. Machine learning algorithms are having a significant impact on sports analytics and cricket is one of them. By using machine learning techniques, we can create new methodologies, predict player performance and select a team for the next matches or series by predicting player performance. The proposed model is made out of statistical information from reputable sports sources and websites on the best ODI players for the Indian national cricket team. The performance of the players depends on various factors such as the opposition team, the venue, and the number of balls faced by the batsman. This paper makes an attempt to forecast the performance of batsmen based on several features. The main motive behind this study is to predict batsmen performance based on available data using different Regression models like Linear Regression (LR), Decision Tree Regression (DT), Random Forest (RF) Regression, Support Vector Regression (SVR), and evaluate which model is best for the prediction of different batsmen performance. Over the past few years, Fantasy sports have experienced tremendous growth, with a burgeoning fan base. Participants in fantasy sports can assemble fictitious teams from actual players. This project was motivated by the desire to assist selectors in predicting and identifying players who would do well in upcoming ODI series as well as provide sports fans with a pleasant way to interact with statistics and assist them in building their ideal teams on fantasy sports using our model.