Document Type

Article

Publication Title

Journal of Sports Analytics

Department

Mathematics

ISSN

2215-0218

Volume

2

Issue

2

First Page

61

Last Page

71

Publication Date

9-1-2016

Abstract

Defensive fouls play an important role in elite men’s water polo generating over half of all goals. Despite their importance, little is known about the relationship between foul calling patterns and other game-state variables in the sport. Here we apply a sequence of hierarchical mixed logistic regression models on data from major tournaments in 2012–2014 to study such relationships and find a number of significant biases in foul calling rates. Offensive teams who are winning/tied are about 31% less likely to draw a defensive foul and 32% more likely to get called for an offensive foul than teams who are losing. The magnitude of losing team bias tends to increase over the course of a game, but is not significantly affected by the size of the lead. A team’s odds of getting called for a foul also increase by about 10% for each consecutive goal scored or foul called in their favor. These biases persist across different offensive and defensive tactical decisions and tournaments suggesting that they are widespread and that it is referees, rather than teams, who are responsible for a lack of independence in water polo foul calling rates.

Creative Commons License

Creative Commons Attribution-NonCommercial 4.0 International License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License

Download

Included in

Mathematics Commons

Share

COinS