Most part of coaches search for reliable performance tools, which can assist the coaching staff to better understand the behavior of the team and players. This way, the coaching staff can properly adapt the game-play strategies, or to define more efficient training methodologies. Due to the high volume of information currently available for coaches, it is necessary to know how to recognize the relevant information and, then, to properly use the selected information. This is a complicated or even impossible task, if the right methodology is not applied, or when no adequate tool is available.

In this work published in the Human Movement Science Journal, we present a methodology for analyzing the interactions between players of a football team, from the point of view of graph theory and complex networks.

We model the complex network of passing interactions between players of a same team in 32 official matches of the First Division of Spanish Professional Football League, using a passing/reception graph. This methodology allows us to understand the team playing structure, by analyzing the offensive phases of game-play. We utilized two different strategies for characterizing the contribution of the players to the team: the clustering coefficient, and centrality metrics (closeness and betweenness).

We show the application of this methodology by analyzing the performance of a professional Spanish team according to these metrics and the distribution of passing/reception in the field. Keeping in mind the dynamic nature of collective sports, in the future we will can incorporate metrics which allows us to analyze the performance of the team also according to the circumstances of game-play and to different contextual variables such as, the utilization of the field space, the time, and the ball, according to specific tactical situations.