Analysis and Predictions of Winning Indian Premier League match using Machine Learning Algorithm

Cricket is the one of the most popular sports. A well-known domestic T20 league in the world is Indian Premier League (IPL). IPL is a very lucrative sport, hence players are in high demand, earning more than${\$}$ 2 million a season. Focusing on each player’s performance and rating for each season is the need of the hour. A batter or bowler’s performance can be evaluated more precisely using IPL data analysis. There is a huge amount of information and statistics involved in cricket and machine learning algorithms use this data to forecast the result of the match. This paper aims to predict the probability of winning IPL cricket matches in each ball of a over while running the second inning and rank the bowlers and batters based upon their performances. A number of vital factors, including team strength, players’ form, possibility of draw and venue, were combined to provide the result of the game. In this paper, the match predictions are made using logistic regression techniques and its result accuracy is shown.