The role of clustering in football: focusing on Brighton & Hove Albion's success

This thesis explores the application of K-means clustering in the context of football scouting, focusing on Brighton \& Hove Albion during the 2022/2023 season and the playing style implemented by Roberto De Zerbi. The objective is to identify, through data analysis, players whose characteristics align with the team’s tactical needs. To achieve this, a dataset containing performance metrics on all players from the top five European leagues (Premier League, La Liga, Serie A, Bundesliga, and Ligue 1) was used. Following a data cleaning phase and the selection of the most representative variables, K-means clustering was implemented to group players based on their performance metrics. The results demonstrate the potential to identify clusters that are consistent with Brighton’s playing style, suggesting profiles that could be of interest from a scouting perspective. The thesis concludes with a discussion on the method’s limitations and potential future developments, including the integration of qualitative data or the use of more complex algorithms.