Do coaches in the National Basketball Association actually display racial bias? Replication and extension
Abstract
We replicate and extend empirical evidence that has been interpreted as an indication of coaches’ racial bias in the National Basketball Association (NBA) by Schroffel and Magee (2012) and Zhang (2017; 2019). After replicating the published results, we extend them by modeling coaches’ decisions of “resting the starters,” a common tactical decision based on players’ quality and not on their race, thus exploring whether this tactic may provide an alternative explanation for associations that might thus far have been taken to indicate racial bias. Our results show that, after empirically accounting for “resting the starters,” the magnitude of associations that have been taken to indicate coaches’ racial bias becomes small and not statistically distinguishable from zero. We discuss the implications of these findings on literatures related to racial integration in organizations, and on the sociology of sports and race.