Authors: Mert Bildirici, Sam Borremans, Lauren Liu
At the heart of a soccer team's strategy and performance is its passing network — a web of interactions that reflects how players connect and collaborate on the field. These networks hold valuable information about a team's style of play, major contributors, and overall effectiveness.
To uncover deeper structural patterns within these networks, Topological Data Analysis (TDA) can be used to study the "shape" of data by identifying its inherent patterns and structures. TDA has been successfully applied in sports like basketball and hockey, where passing is crucial, offering valuable insights into team dynamics and performance.
With the growing application of data science in soccer [3], and particularly in analyzing passing networks, we wanted to explore how the topology of soccer teams' networks — especially their homology — correlates with their scoring. By examining these relationships, this study seeks to provide insights into passing strategies that can enhance goal-scoring outcomes in soccer.
This paper will look at the 2015/2016 season across Europe's top five leagues: the Premier League, the Bundesliga, Serie A, La Liga, and Ligue 1. It first introduces the construction and interpretation of passing networks, followed by an overview of the TDA applications employed. Finally, it analyzes the relationship between the homology of passing networks and the number of goals scored by a team, while also providing insights into the differences across these five leagues.
Read the full paper here.