In this thesis, the aim was to identify the tactical patterns of football teams using an unsupervised machine learning approach. We considered the passes that every team made as input to the k-means and Hierarchical clustering models to classify the teams into clusters. In the five top European leagues. The top 6 teams had most of their matches clustered in one cluster that was characterized by the high overall number of passes, the domination of the middle area of the football pitch, and the dependence on short passes to attack, Our results agree with. For the Premier Legue, Ligue 1 and the Bundesliga the league winner had different tactical style than other teams. While in Serie A Napoli, the runner-up had this unique play style and for La Liga Barcelona (winner) and Real Madrid (third) had a similar play style.
AhmedTremo/FootballTactics
Unsupervised machine learning techniques to analyze tactical behaviors of football teams.
Jupyter Notebook