Working Paper: CEPR ID: DP10081
Authors: Alex Krumer; Reut Megidish; Aner Sela
Abstract: We study round-robin and elimination tournaments with three players where one player is dominant, i.e., he has a higher value of winning than his weaker opponents. In every stage, a pair-wise match is modelled as an all-pay auction. We demonstrate that the expected payoff of the weak players in the round-robin tournament is higher than or equal to their expected payoffs in the elimination tournament. On the other hand, the expected payoff of the dominant player in the elimination tournament could be either higher or lower than in the round-robin tournament. We also show that if a contest designer wishes to maximize the dominant player's probability to win he should organize a round-robin tournament. However, if he wishes to maximize the players' expected total effort, then if the asymmetry between the players is relatively low, he should prefer the elimination tournament, while if the asymmetry is relatively high, he should prefer the round-robin tournament.
Keywords: All-pay auctions; Elimination tournaments; Roundrobin tournaments
JEL Codes: D44
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| tournament structure (L10) | expected payoff of weak players (C72) |
| tournament structure (L10) | expected payoff of dominant player (C72) |
| roundrobin tournament (C72) | probability of winning for dominant player (C72) |
| tournament type (Z29) | expected total effort (C13) |
| asymmetry of player strengths (C72) | tournament type preference (Z29) |