Automated Detection of Match Phases in Football from Spatio-Temporal Tracking Data Using Graph Neural Networks
Abstract
Spatio-temporal tracking data has opened new possibilities for detecting complex tactical patterns in football, yet modeling the interactive movements of multiple players remains challenging. This paper proposes a framework combining graph neural networks (GNNs) with a sequential model to classify match phases on a second-by-second basis across a seven-class taxonomy. Match phase classification is tactically meaningful, and the availability of rule-based labels across 203 matches makes it a suitable testbed for a systematic comparison of adjacency constructions and message-passing layers, a question that has received limited attention in existing research. Our selected GNN-LSTM model outperforms all aggregated-feature baselines, including XGBoost and a Long Short-Term Memory (LSTM) network, as the strongest baseline scores 4.6% lower in macro F1. Graph representations using a domain-informed Delaunay triangulation that approximates passing lanes, paired with a custom Spatial Edge-Augmented Convolution (SEAConv) layer that injects edge attributes directly into messages, achieve the best performance by capturing spatial dependencies while limiting uninformative messages from redundant edges. Integrated Gradients attributions indicate the model uses spatial player configurations, particularly horizontal positioning, in combination with possession and ball-status indicators. This work offers an automated solution for fine-grained tactical analysis, reducing the need for manual tagging and providing deeper insight into dynamic team behavior.
Related
- Integrity at Stake: Statistical Detection of Anomalous In-Play Betting in Football
- Distributed Cross-Layer Optimization for Covert Multi-Hop, Multi-Modal Networks: Exponentially Fast Convergence and Robust Tracking
- Feature tracking in physics-informed neural networks via joint optimization of nonlinear deformation manifolds: application to shocks
- Automated correction of training sets for biological image segmentation by convolutional networks
- FPScan: An Automated Constraint-Based Analyzer for Floating-Point Anomaly Detection
BibTeX
@misc{renner2026automated,
title = {{Automated Detection of Match Phases in Football from Spatio-Temporal Tracking Data Using Graph Neural Networks}},
author = {Vincent Renner and Nils Koster and Pascal Bauer and Melanie Schienle},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.11571},
eprint = {2610.11571},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.11571},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Automated Detection of Match Phases in Football from Spatio-Temporal Tracking Data Using Graph Neural Networks
AU - Vincent Renner
AU - Nils Koster
AU - Pascal Bauer
AU - Melanie Schienle
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.11571
UR - https://arxiv.org/abs/2610.11571
AB - Spatio-temporal tracking data has opened new possibilities for detecting complex tactical patterns in football, yet modeling the interactive movements of multiple players remains challenging. This paper proposes a framework combining graph neural networks (GNNs) with a sequential model to classify match phases on a second-by-second basis across a seven-class taxonomy. Match phase classification is tactically meaningful, and the availability of rule-based labels across 203 matches makes it a suitable testbed for a systematic comparison of adjacency constructions and message-passing layers, a question that has received limited attention in existing research. Our selected GNN-LSTM model outperforms all aggregated-feature baselines, including XGBoost and a Long Short-Term Memory (LSTM) network, as the strongest baseline scores 4.6% lower in macro F1. Graph representations using a domain-informed Delaunay triangulation that approximates passing lanes, paired with a custom Spatial Edge-Augmented Convolution (SEAConv) layer that injects edge attributes directly into messages, achieve the best performance by capturing spatial dependencies while limiting uninformative messages from redundant edges. Integrated Gradients attributions indicate the model uses spatial player configurations, particularly horizontal positioning, in combination with possession and ball-status indicators. This work offers an automated solution for fine-grained tactical analysis, reducing the need for manual tagging and providing deeper insight into dynamic team behavior.
ER -