Comparative Analysis of Soccer Players in the Dutch Eredivisie Using Machine Learning and Statistical Methods on Tracking and Performance Data

This thesis provides a comprehensive analysis of the impact of substitute players on elite soccer team performance, examining their influence from both individual and collective perspectives. Utilizing various data sources, including diverse tracking data, event data, and physical data provided by SciSports. The study explores the dynamics of player substitutions and their ramifications on team outcomes. Key research questions addressed include the assessment of player and team performance metrics, the analysis of team dynamics and tactical strategies influenced by player substitutions, and the exploration of machine learning techniques to optimize player scouting and substitution strategies. The findings demonstrate that machine learning techniques significantly enhance the identification and integration of newly scouted players or substitutes, ultimately maximizing team performance in elite soccer teams. By identifying gaps in current literature and posing pertinent research questions, this review sets the stage for future research endeavors aimed at enhancing team success in competitive soccer matches through advanced sports analytics approaches.