The effect of timeouts on team performance : evidence from e-sports

Abstract

This study aims to determine the causal effect of timeouts on team performance. To do so, I constructed a data set on 103 high-tier Counter-Strike: Global Offensive matches. I use exogenous variation stemming from technical difficulties, for example, latency issues. When defining a timeout as one having lasted at least 30 seconds - the usual tactical timeouts’ duration- having had a technical timeout increases the probability of reversing the previous round’s winner by 14.76 percentage points. This study has important implications for sports. First, it shows that the belief that timeouts can help break the other team’s momentum, and are, therefore, a valuable tool to increase the probability of winning, has some empirical support. Second, given the first point, knowing when to call a timeout could be critical to the team’s success. That is, coaches should be selected on this capability as well. Third, increasing the number of timeouts, or even making timeouts available on sports that do not currently allow them, could make the sport more competitive by making it easier to stop a team’s momentum.