By exploiting an unusually rich panel data set from the National Database of Student Athletes, this article addresses the issue of peer effects and superstar effects on performance in multi-stage swimming competitions. Four key findings are presented. First, the evidence from relay teams supports a positive inter-team peer effect from competitors in a male relay race, but a negative inter-team peer effect from competitors in a female relay race. Second, the evidence from both female and male relay teams shows that there exists a positive intra-team peer effect from teammates. In our estimations for female relay races, a foolish teammate does more harm (+3.11 seconds) than a brilliant opponent does good (+0.55 seconds) in the estimation of the Heckman Selection Model based on panel data. Third, a male team with average-quality swimmers performs better than a team with dispersed-quality swimmers. Fourth, for the super-team effects in these tournaments, on average, the female relay teams' (/male relay teams') times are approximately 2.85(/2.09) seconds faster/slower when the previous year's winning team participates.3 Brown (2011) provides a simple formal model of the effects of incentives and strategies created by the presence of a superstar in a professional sports setting.