Rating Player Actions in Soccer.

graph networks soccer sports analytics trajectory data trajectory prediction

Journal

Frontiers in sports and active living
ISSN: 2624-9367
Titre abrégé: Front Sports Act Living
Pays: Switzerland
ID NLM: 101765780

Informations de publication

Date de publication:
2021
Historique:
received: 19 03 2021
accepted: 26 05 2021
entrez: 2 8 2021
pubmed: 3 8 2021
medline: 3 8 2021
Statut: epublish

Résumé

We present a data-driven model that rates actions of the player in soccer with respect to their contribution to ball possession phases. This study approach consists of two interconnected parts: (i) a trajectory prediction model that is learned from real tracking data and predicts movements of players and (ii) a prediction model for the outcome of a ball possession phase. Interactions between players and a ball are captured by a graph recurrent neural network (GRNN) and we show empirically that the network reliably predicts both, player trajectories as well as outcomes of ball possession phases. We derive a set of aggregated performance indicators to compare players with respect to. to their contribution to the success of their team.

Identifiants

pubmed: 34337404
doi: 10.3389/fspor.2021.682986
pmc: PMC8319236
doi:

Types de publication

Journal Article

Langues

eng

Pagination

682986

Informations de copyright

Copyright © 2021 Dick, Tavakol and Brefeld.

Déclaration de conflit d'intérêts

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Références

PLoS One. 2016 Dec 30;11(12):e0168768
pubmed: 28036407
Big Data. 2019 Mar;7(1):57-70
pubmed: 30321059
Big Data. 2019 Mar;7(1):71-82
pubmed: 30672712
PLoS One. 2020 Mar 10;15(3):e0230179
pubmed: 32155220

Auteurs

Uwe Dick (U)

Machine Learning Group, Leuphana University of Lüneburg, Lüneburg, Germany.

Maryam Tavakol (M)

UAI Group, Eindhoven University of Technology, Eindhoven, Netherlands.

Ulf Brefeld (U)

Machine Learning Group, Leuphana University of Lüneburg, Lüneburg, Germany.

Classifications MeSH