The Role of Performance Metrics in Estimating Market Values of Footballers in Europe's Top Five Leagues

Abstract

The transfer economy in football is a multi-billion-dollar industry, where accurate valuation of players is crucial for clubs' financial sustainability and competitive success. This study investigates the role of performance metrics in estimating the market values of football players in Europe's top five leagues (Spain's La Liga, France's Ligue 1, England's Premier League, Italy's Serie A, and Germany's Bundesliga). The study collected 28 performance metrics (e.g., goals, shots per game, assists, and pass success percentage) for 1508 players from the Whoscored platform. Additionally, the players' positions and the leagues they play in were also included as features. These data were combined with market values from the Transfermarkt platform, resulting in a comprehensive dataset. Two main analytical methods were employed: regression and classification. In the regression analysis, seven models (Adaboost, Decision Tree, Gradient Boosting, K Nearest Neighbors, Random Forest, Ridge Regression, and Support Vector Machine) predicted players' market values. The highest accuracy was achieved with the Random Forest algorithm (R-squared: 0.90). In the classification analysis, players' market values were categorized into four classes (low, lower-mid, upper-mid, and high), and their class memberships were predicted based on performance metrics. The CNN algorithm achieved the highest accuracy, with a success rate of 97%. The results indicate that performance metrics significantly contribute to estimating football players' market values, and models based on these metrics can assist clubs in making more informed, data-driven decisions during transfers.

Keywords

• football player valuation
• performance metrics
• transfer market analysis
• machine learning in sports