SWINGING BIDS: PREDICTING IPL AUCTION PRICES WITH MACHINE LEARNING

Abstract

The Indian Premier League (IPL) is a globally recognized, analytics-driven cricket league where player auctions are highly competitive and influenced by a blend of on-field performance and strategic considerations. Traditionally, franchises have relied on both player statistics and subjective judgments to determine auction values. This study introduces a machine learning (ML) approach to predict IPL auction prices, aiming to improve transparency and consistency in player valuation. Using a dataset of over 214 players—including metrics such as runs, strike rate, batting average, wickets, economy rate, and player roles—comprehensive data preprocessing and feature engineering were performed to extract relevant variables. Various models were explored, with gradient boosting emerging as the most accurate in predicting auction outcomes. . Key findings indicate that base price, player role (especially all-rounders), years of experience, and recent performance are major determinants of auction value. Feature importance was assessed using mutual information and model-based techniques, while visualizations and correlation analysis validated the relationships within the data. The results demonstrate that ML can effectively support strategic decision-making for IPL franchises, with future improvements possible by integrating qualitative factors like player popularity and injury history. Overall, this research highlights the practical value of machine learning in sports analytics, offering actionable insights for team owners, analysts, and the broader cricketing community.