Painting soccer pass surfaces improves prediction of player choices
Graph-to-Grid (G2G): Continuous-Coordinate Feature Painting for Soccer Pass Surfaces
Machine LearningComputer Vision and Pattern Recognition
Summary
The paper studies how to better predict where soccer players will pass the ball on the pitch. Instead of just counting players in each grid cell, the authors spread each player’s position smoothly onto the grid, keeping precise location info. This 'painting' method improves how well models predict pass choices, tested on World Cup data and German league games. The biggest gains come from simply painting raw player features, with extra model tweaks adding smaller improvements.
What this means in practice
- •For sports analytics teams: Enhance soccer pass prediction models by encoding player features directly on continuous pitch coordinates, improving decision-making simulations.
- •For sports broadcast technology developers: Integrate fine-grained player location data to generate more accurate visualizations of passing options during live games.$Commercial implications: Enables improved broadcast graphics that can better engage viewers and provide richer tactical insights.
Authors
Kaan Günay, Orhun Gun
Abstract
Dense pass surfaces give, for every pitch cell, whether a pass played there would arrive, whether the carrier would choose it, and what the possession would then be worth. The networks that draw them read the state as a raster of per-cell counts, losing where inside a cell each player stands. LiDAR detectors, bird's-eye-view perception and graph weather models move entity features onto a grid, binning each entity to a cell or learning the transfer. We evaluate the interpolated form: each player's features are scattered bilinearly onto the grid at the player's measured coordinates, so the surface loss trains the per-player encoder end to end. Those systems adopt an interface; this paper measures one. On 53,628 passes from the 2022 World Cup, painting improves selection likelihood over the same core fed rasters alone by about a quarter of a nat: in every match of an eight-fold cross-validation, with every arm tuned over five seeds, and after retraining on seven Bundesliga and 2. Bundesliga matches from another provider. Thirteen pre-specified studies locate the gain: painting the nine raw player features with no encoder carries three quarters of it, and the learned encoder and message passing add a smaller, resolved increment. Painting also helps the original SoccerMap and a canonical U-Net, whereas offset channels, a finer raster, an attention painter and a raster-free decoder do not. Frozen across the provider boundary the likelihood advantage is lost; injected tracking error compresses it. These results concern observed-endpoint prediction, not calibrated evaluation of hypothetical passes.