A causality-inspired plus-minus model for player evaluation in team sports

Caterina De Bacco, Yixin Wang, David Blei
Proceedings of the Third Conference on Causal Learning and Reasoning, PMLR 236:769-792, 2024.

Abstract

We present a causality-inspired adjusted plus-minus model for evaluating individual players from their performance on a team. We take an explicitly causal approach to this problem, defining the value of a player to be the expected change in the score had we substituted the player for one who has zero value. (This quantity is “causal” in the sense that it is an inference about a hypothetical intervention.) We adapt recent ideas of factor modeling to handle the indirectly measured confounding in estimating player values, considering each player to be a “treatment” who contributes to the outcome of the game. We demonstrate the behavior of the model on data about soccer and basketball.

Cite this Paper

BibTeX
@InProceedings{pmlr-v236-bacco24a, title = {A causality-inspired plus-minus model for player evaluation in team sports}, author = {Bacco, Caterina De and Wang, Yixin and Blei, David}, booktitle = {Proceedings of the Third Conference on Causal Learning and Reasoning}, pages = {769--792}, year = {2024}, editor = {Locatello, Francesco and Didelez, Vanessa}, volume = {236}, series = {Proceedings of Machine Learning Research}, month = {01--03 Apr}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v236/bacco24a/bacco24a.pdf}, url = {https://proceedings.mlr.press/v236/bacco24a.html}, abstract = {We present a causality-inspired adjusted plus-minus model for evaluating individual players from their performance on a team. We take an explicitly causal approach to this problem, defining the value of a player to be the expected change in the score had we substituted the player for one who has zero value. (This quantity is “causal” in the sense that it is an inference about a hypothetical intervention.) We adapt recent ideas of factor modeling to handle the indirectly measured confounding in estimating player values, considering each player to be a “treatment” who contributes to the outcome of the game. We demonstrate the behavior of the model on data about soccer and basketball.} }
Endnote
%0 Conference Paper %T A causality-inspired plus-minus model for player evaluation in team sports %A Caterina De Bacco %A Yixin Wang %A David Blei %B Proceedings of the Third Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2024 %E Francesco Locatello %E Vanessa Didelez %F pmlr-v236-bacco24a %I PMLR %P 769--792 %U https://proceedings.mlr.press/v236/bacco24a.html %V 236 %X We present a causality-inspired adjusted plus-minus model for evaluating individual players from their performance on a team. We take an explicitly causal approach to this problem, defining the value of a player to be the expected change in the score had we substituted the player for one who has zero value. (This quantity is “causal” in the sense that it is an inference about a hypothetical intervention.) We adapt recent ideas of factor modeling to handle the indirectly measured confounding in estimating player values, considering each player to be a “treatment” who contributes to the outcome of the game. We demonstrate the behavior of the model on data about soccer and basketball.
APA
Bacco, C.D., Wang, Y. & Blei, D.. (2024). A causality-inspired plus-minus model for player evaluation in team sports. Proceedings of the Third Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 236:769-792 Available from https://proceedings.mlr.press/v236/bacco24a.html.

Related Material