Abstract

Football is one of the world´s most popular sports and constitutes a global industry where vast amount of money circulate, especially through player transfers. The ability to accurately estimate a player´s market value is therefore central for football clubs all around the world. Such decisions have historically heavily relied on intuition and subjective assessments by coaches and scouts. However, the increasing availability of data and the rise of football analytics have made it possible to approach these decisions in a more systematic and objective way.

This study developed a data-driven machine learning model to support football clubs in the valuation of football players and assist in making informed decisions during transfer negotiations. By combining data on player performance, contract details, and football club financial information, the model aimed to identify the key factors influencing transfer fees and translate these into reliable value estimations.

The model successfully identified several key determinants of player market values. The analysis revealed that the most significant predictors were a player’s remaining contract duration, age, and on-field performance relative to peers in the same age group and division. Furthermore, the study found that market valuation dynamics differed notably across football divisions. For example, remaining contract length proved to be a substantially more critical factor for player valuation in top-tier divisions compared to lower-tier divisions. These findings highlight the nuanced nature of player valuation and suggest that context-specific models may be beneficial, underscoring the need for further research to fully understand how influencing factors vary across different footballing contexts.