Book

Description

Simply, cricket can be called as a game between bat and ball played by two teams. There are three main rolls among the players and they are batsmen bowlers and allrounders. Predicting a top rank batsman and selecting the best pool of 15 players is very important in the game. It is the main purpose of this research. Data collected from www.cricinfo.com website related to Virat Kholi, Rohit Sharma, Babar Azam (top three batsmen at 16.08.2021) and t20 internation cricket matches played by Sri Lanka (2019 – February 2022). In this paper, the success of a batsman has been defined as scoring greater than or equal to 50 runs. Decision tree, probabilistic regression, support vector machine standardization variables were used as techniques. Among the developed models, the decision tree models showed good accuracies (72%) when predicting a top rank batsmen. To get this 72% accuracy, details of first five balls were needed for Virat Kholi and Rohit Sharma. For Babar Azam, details of first seven balls were needed to get that accuracy.