AI

Learning Human Sports Motion Feasibility via Quantile Regression on Tracking Data

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 28/10/2026, 04:03:25 UTC.

Briefings

A new study using quantile regression on optical tracking data from the 2022 FIFA World Cup analyzed time-to-arrival (TTA) for players. The research found that a lightweight multilayer perceptron (MLP) consistently outperformed six existing physics-based player-motion models and a learned reachable-hull baseline. The model was tested on the 2022 World Cup dataset and evaluated for transferability across basketball and NFL tracking data.

Key points

• The study used optical tracking data from all 64 matches of the 2022 FIFA World Cup to train a multilayer perceptron for time-to-arrival prediction.
• The MLP model consistently outperformed six existing physics-based player-motion models and a learned reachable-hull baseline across evaluation metrics.

Context

This research addresses the challenge of estimating the shortest plausible time a player needs to reach an arbitrary location on the pitch, termed time-to-arrival (TTA). The findings are significant for developing lightweight, data-driven models that can predict player movement feasibility without relying on complex physics-based assumptions.

Limitations

The primary evidence is limited to the 2022 FIFA World Cup dataset, which may not represent all sports or player populations. The study also evaluated the model on basketball and NFL data, but the specific performance metrics for these transfers are not detailed in the provided text.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 28/10/2026, 04:56:07 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeidcbwdnm7jcf6l3v2jl2axtgbqv2evylqfxgc6efd6qvyh7k4qqji