Affirmative-action policies bias tournament rules in order to provide equal opportunities to a group of competitors who have a disadvantage they cannot be held responsible for. Critics argue that they distort incentives, resulting in lower individual performance, and that the selected pool of tournament winners may be inefficient. In this paper, we study the empirical validity of such claims in a real-effort pair-wise tournament between children from two similar schools who systematically differ in how much training they received ex-ante in the task at hand. Our results show that performance was not reduced for either advantaged or disadvantaged subjects and that it was in fact enhanced. Additionally, while affirmative action balanced the proportion of disadvantaged individuals winning their respective tournament, the average performance of the pool of winners only decreased slightly.