abstract
- The identification of talented young footballers is a cornerstone of success in professional football. The ability to recognize future elite players has consistently translated into a significant competitive advantage throughout the history of the sport. This study delves into this domain by comparing the performance of three supervised machine learning models. The models were trained using a comprehensive dataset encompassing data for 1,086 male professional footballers. This data incorporates player statistics, game-related attributes, and transfer market values. The Support Vector Machine (SVM) model emerged as the most effective in identifying elite players based on a 5-fold cross-validation process. The feature importance analysis and other valuable insights gleaned from the results pave the way for further research endeavors. The study aims to encourage the adoption of advanced data analytics and statistical methods within football clubs worldwide. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.