Abstract—The selection of players in cricket, particularly in formats like the Indian Premier League (IPL), is crucial for a team’s success. This process involves analyzing various player attributes, including their batting and bowling performances, to create a balanced team. In cricket, each player’s role—batsman, bowler, or all-rounder—must be clearly defined. This clarity helps in optimizing team dynamics and ensuring that players can perform at their best. For instance, a well-rounded team might include specialists who excel in specific areas, during matches. Recent advancements in Artificial Intelligence (AI) have opened new avenues for player analysis. Unassisted AI models can evaluate bowlers based on their historical performances in the IPL. By leveraging data from past matches, these models can predict future performances, aiding selectors in making informed decisions about player selection. This study employs predictive modelling to analyze bowler performance using unsupervised learning. The developed AI model can extend its applicability beyond the IPL, offering insights into bowlers from other leagues. There are some discriminative features of players’ performance as well as team strategy, which can be obtained by applying machine learning in Bowler Performance Analysis. This data-driven approach provides a decision-making and fast response possibility to managers and analysts in playing T20 matches 2018. By combining various techniques, e.g., K-means and mean-shift, teams are also equipped to select players more effectively, establish strategic and success rate, etc.
 
Keywords—sports analytics, bowler performance, unsupervised Machine Learning (ML), k-means clustering, cross-league prediction, process innovation, silhouette score, clustering algorithms, machine learning in sports


Copyright © 2025 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).