In this project, a comprehensive analysis of the UEFA Champions League Final 2019 game is performed. The passing network of both teams is created and studied. Meaningful insights are extracted by applying network science techniques on the passing networks of both teams.
The analysis performed aims to answer the following questionsI regarding the game:
- Who is the most valuable player in terms of ball circulation for each team?
- Which tactical plan was followed by each team, and how was it respectively countered by the opponent?
- Which are the "subgroups" of players of each temam who frequently interact during the game?
- Who was the team-player and the individual during the game?
- What alternative tactics could be followed?
- Ultimately, is ball possession a significant factor in winning a football game?
The full report of the analysis can be found here.
The dataset utilized for this analysis was sourced from StatsBomb's Open Data platform, a reputable and comprehensive resource for football-related statistics. More specifically, StatsBomb, one of the leading companies in the collection and analysis of data from sports events, freely provides statistics for passes, interceptions, and crosses for a total of 4 Champions League finals in its GitHub repository statsbomb/open-data.
Matplotlib, in conjunction with mpsoccer, was employed for visualizing the passing networks. The latter is a comprehensive library specifically designed for visualizing soccer teams on the pitch, offering a holistic approach to football data visualization. The code which generates the passing networks can be found in the generateCharts.py file.
The network analysis part of the project was performed using Gephi, an open-source network analysis software package. Insights were extracted from the passing networks by applying popular network science metrics :
- Weighted Degree
- Weighted In-Degree
- Weighted Out-Degree
- Closeness Centrality
- Betweenness Centrality
- Eigen Centrality
- PageRank
- Node Clustering Coefficient
- Modularity
- Bridging Centrality
- Bridging Coefficient
