A spatio-temporal graph-based model for team sports analysis

2026-03-18

AI Summary

This study addresses the challenges of high spatiotemporal complexity and the absence of a general modeling framework in team sports tactical analysis, particularly the difficulty in quantifying how external constraints—such as coaching instructions and defensive formations—influence offensive behavior. The authors propose the first general spatiotemporal graph model tailored for team sports, representing offensive sequences as directed graph paths that integrate ego-centric spatial features of the ball carrier, temporal dynamics, and semantic labels. This approach enables cross-sport transferability and facilitates the empirical validation of diverse coordination hypotheses. Experiments on six-a-side rugby data demonstrate that both coaching pedagogy and initial defensive alignment significantly shape offensive strategies. The model is designed for extension to other team sports, including basketball.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Humans and AI: Teamwork, Team formation

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Team sports represent complex phenomena characterized by both spatial and temporal dimensions, making their analysis inherently challenging. In this study, we examine team sports as complex systems, specifically focusing on the tactical aspects influenced by external constraints. To this end, we introduce a new generic graph-based model to analyze these phenomena. Specifically, we model a team sport's attacking play as a directed path containing absolute and relative ball carrier-centered spatial information, temporal information, and semantic information. We apply our model to union rugby, aiming to validate two hypotheses regarding the impact of the pedagogy provided by the coach on the one hand, and the effect of the initial positioning of the defensive team on the other hand. Preliminary results from data collected on six-player rugby from several French clubs indicate notable effects of these constraints. The model is intended to be applied to other team sports and to validate additional hypotheses related to team coordination patterns, including upcoming applications in basketball.

Problem

• team sports
• spatio-temporal analysis
• tactical behavior
• external constraints
• graph-based modeling

Innovation

• spatio-temporal graph
• team sports analysis
• tactical modeling
• graph-based representation
• complex systems