IPL Player Valuation: A Data-Driven Approach for Efficient Auctions Using K-Means Clustering

This research presents a data-driven methodology for determining cricket players’ base prices in the Indian Premier League (IPL) auctions, aiming to streamline the auction process. Utilizing a comprehensive ball-by-ball information dataset covering all IPL matches until 2023, we established a preprocessing pipeline incorporating player performance metrics and auction data from 2014 to 2022. The Runs Above Average (RAA) metric is crucial to player valuation. Employing K-means clustering, we group players based on similar performance attributes (RAA), refining base price determination through Euclidean distance comparison across 14 key attributes. Evaluation metrics reveal a trade-off between accuracy and efficiency, with optimal accuracy at 76.34% achieved at k = 8 with an efficiency of 43.59%. Acknowledging limitations tied to cricket’s dynamic nature and cluster homogeneity assumptions, future research directions propose exploring dynamic clustering models, integrating external factors, and incorporating advanced machine learning techniques. This research introduces a systematic, data-driven approach to IPL player valuation, contributing to enhanced auction process efficiency and providing a valuable tool for informed decision-making by IPL franchises and organizers during player acquisitions.