PathCRF: Ball-Free Soccer Event Detection via Possession Path Inference from Player Trajectories

TL;DR

PathCRF addresses scalable on-ball event analytics in soccer by eliminating the need for expensive ball tracking. It frames event detection as inferring a possession path over a fully connected dynamic graph using a neural backbone to produce edge embeddings and a dynamic masked CRF to enforce feasible transitions, with Viterbi decoding for inference. The approach achieves strong event-level performance (F1 around 75%) and produces accurate downstream analytics such as heatmaps, possession statistics, and pass networks, while enabling semi-automated annotation workflows. The work demonstrates that high-quality event-based soccer analytics can be obtained from player tracking data alone, broadening access beyond top-tier competitions.

Abstract

Despite recent advances in AI, event data collection in soccer still relies heavily on labor-intensive manual annotation. Although prior work has explored automatic event detection using player and ball trajectories, ball tracking also remains difficult to scale due to high infrastructural and operational costs. As a result, comprehensive data collection in soccer is largely confined to top-tier competitions, limiting the broader adoption of data-driven analysis in this domain. To address this challenge, this paper proposes PathCRF, a framework for detecting on-ball soccer events using only player tracking data. We model player trajectories as a fully connected dynamic graph and formulate event detection as the problem of selecting exactly one edge corresponding to the current possession state at each time step. To ensure logical consistency of the resulting edge sequence, we employ a Conditional Random Field (CRF) that forbids impossible transitions between consecutive edges. Both emission and transition scores dynamically computed from edge embeddings produced by a Set Attention-based backbone architecture. During inference, the most probable edge sequence is obtained via Viterbi decoding, and events such as ball controls or passes are detected whenever the selected edge changes between adjacent time steps. Experiments show that PathCRF produces accurate, logically consistent possession paths, enabling reliable downstream analyses while substantially reducing the need for manual event annotation. The source code is available at https://github.com/hyunsungkim-ds/pathcrf.git.

Figures (6)

• Figure 1: Overall architecture of PathCRF.
• Figure 2: Comparison of the ground-truth possession path and events with model predictions on an in-play segment of the test dataset. Predicted events with incorrect sender or receiver attributes are highlighted in yellow, and physically impossible transitions are highlighted in red.
• Figure 3: Comparison of kernel density estimation (KDE) heatmaps computed from true and detected events, for both teams and the most event-involved player from each team.
• Figure 4: 5-minute timeline of home team's possession shares computed from true and detected events, respectively.
• Figure 5: Visual comparison of pass networks constructed from ground-truth and detected events.