Predicting bowling performance in cricket from publicly available data

Abstract

Cricket is one of the most popular games worldwide. The aim of this paper is to predict bowlers' performance from publicly available data. Team management follows different strategies to win in the tournaments. They anticipate bowlers' performance of the opposite team in diverse conditions and prepare their batsmen accordingly. Similarly, they also foresee the strength and weakness of opponent teams' batsmen and suggest their bowlers to perform different tricks in various environments. In this paper, we build a machine learning based approach to predict bowler's performance in varying conditions by using 6,031 bowling instances of One day International (ODI) matches. Our classifier shows substantial prediction potential to predict bowler's performance.