Prediction of Football Players’ Value in the Transfer Market of Well-known European Leagues based on FIFA 19 and Real-world Data

Abstract

Abstract: The study delves into FIFA’s role as the global regulatory authority for football, managing the sport’s development and major events like the FIFA World Cup. FIFA’s influence extends to economic goals, impacting football clubs globally as they invest in skilled players. The market valuation of players is crucial, guiding budget allocation for transfers. Using data from the FIFA 19 video game and real-world statistics, the study employs Decision Tree Regression (DTR) and Random Forest Regression (RFR) models, addressing multicollinearity with Variance Inflation Factor (VIF). The Rhizostoma Optimization Algorithm (ROA) and Dwarf Mongoose Optimizer (DMO) optimize models. Results show RFR-based models, particularly RFRO, outperform DTR-based ones, achieving over 99% R2 value and 12% error relative to mean market values. Ensemble models RFRD and DTRD provide a reliable prediction capability of around 98%, aiding real-world decision-making in the football transfer market for club managers, coaches, and stakeholders across different leagues.