PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball using Tracking Data

O. Müller, M. Caron, M. Döring, T. Heuwinkel, J. Baumeister, in: 8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021), n.d.

Download
No fulltext has been uploaded.
Conference Paper | In Press | English
Abstract
Over the last years, several approaches for the data-driven estimation of expected possession value (EPV) in basketball and association football (soccer) have been proposed. In this paper, we develop and evaluate PIVOT: the first such framework for team handball. Accounting for the fast-paced, dynamic nature and relative data scarcity of hand- ball, we propose a parsimonious end-to-end deep learning architecture that relies solely on tracking data. This efficient approach is capable of predicting the probability that a team will score within the near future given the fine-grained spatio-temporal distribution of all players and the ball over the last seconds of the game. Our experiments indicate that PIVOT is able to produce accurate and calibrated probability estimates, even when trained on a relatively small dataset. We also showcase two interactive applications of PIVOT for valuing actual and counterfactual player decisions and actions in real-time.
Publishing Year
Proceedings Title
8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021)
Conference
European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML PKDD 2021)
Conference Location
Online
Conference Date
2021-09-13 – 2021-09-17
LibreCat-ID

Cite this

Müller O, Caron M, Döring M, Heuwinkel T, Baumeister J. PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball using Tracking Data. In: 8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021).
Müller, O., Caron, M., Döring, M., Heuwinkel, T., & Baumeister, J. (n.d.). PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball using Tracking Data. 8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021). European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML PKDD 2021), Online.
@inproceedings{Müller_Caron_Döring_Heuwinkel_Baumeister, title={PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball using Tracking Data}, booktitle={8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021)}, author={Müller, Oliver and Caron, Matthew and Döring, Michael and Heuwinkel, Tim and Baumeister, Jochen} }
Müller, Oliver, Matthew Caron, Michael Döring, Tim Heuwinkel, and Jochen Baumeister. “PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball Using Tracking Data.” In 8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021), n.d.
O. Müller, M. Caron, M. Döring, T. Heuwinkel, and J. Baumeister, “PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball using Tracking Data,” presented at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML PKDD 2021), Online.
Müller, Oliver, et al. “PIVOT: A Parsimonious End-to-End Learning Framework for Valuing Player Actions in Handball Using Tracking Data.” 8th Workshop on Machine Learning and Data Mining for Sports Analytics (ECML PKDD 2021).

Export

Marked Publications

Open Data LibreCat

Search this title in

Google Scholar