A Two-Stage Spatio-Geometrical Clustering of Football Team Shape for Post-Match Review

Abstract The existence of a significant amount of spatio-temporal data in a football match creates a good potential for Post-Match Review and analysis of team behavior. These analyses can be done by focusing on the whole team or individual players. The purpose of this paper is to analyze the efficiency and general behavior of the team in the form of a single entity, from a spatio-geometrical point of view. This process starts by defining a convex hull as the team shape in each time frame. In the next step, a set of spatial, geometric, zone-based, and event-based parameters are introduced and extracted to describe the shape of the team at each frame. These descriptors are the basis of the two-stage spatio-geometrical clustering of the team during the match. What is obtained from clustering is the identification of similar patterns for the shape of the team in situations of in-possession and out-of-possession of the ball. Examining these clusters in the Post-Match Review process determines the overall performance of the team in different situations, the extent of the team's dominance over different areas of the field, as well as the team's technical strategies. No need to transfer the team shape to the image space, no need for image processing techniques for analysis and thus reducing the computational load, introducing a new geometric descriptor and performing clustering in two stages for a better and more meaningful interpretation of the team shape from the points are the distinguishing points of this article.