Classification Model to Predict the Outcome of an IPL Match

Indian Premiere League or IPL is a very popular form of cricket in India. With the growing interest in the game of cricket, IPL is becoming increasingly popular among the young and old alike. With so many teams participating in IPL, new and young players are getting an opportunity to showcase their talent, play alongside international players and are duly compensated for their talent. With so many regional teams, various players and a vast array of matches, the franchise owners and the fans are very keen and enthusiastic about the winner of these matches. This paper aims to build a Classification Model to predict the outcome of an IPL match using the history data of IPL matches from 2008–2020 and the initial few matches of IPL 2021. Four models have been built using supervised machine learning algorithms like Logistic Regression, Support Vector Machines (SVC), Decision Trees and Random Forest in Python. These models have been built using various features like Season, Venue, Team1, Team2, Toss_winner, Toss_decision, Result_margin etc. The models predict which of the two teams (Team1 or Team2) wins a particular match. Finally, these models are compared for the accuracy of their prediction and the best model is chosen. The best model could match predict the outcome with an accuracy of 73.60%, better than the other models studied in the Literature Review section. This model can be used as a tool to predict which team will win or lose a match, thus enabling teams to modify their strategies, generating money minting options for betting companies etc.