The efficacy of tournament designs

Tournaments are a widely used mechanism to rank alternatives in noisy environment. This paper investigates fundamental issue of economics tournament design: what is the best usage limited resources, that is, how should be compared pairwise approximate their true but latent ranking. We consider various formats including knockout tournaments, multi-stage championships consisting round-robin groups followed by single elimination, and Swiss-system. They evaluated via Monte-Carlo simulations under six different assumptions on winning probabilities. Comparing same pair multiple times turns out an inefficacious policy. While seeding can increase efficacy group-based designs, its influence remains marginal unless one has unrealistically good estimation ranking players. The Swiss-system found most accurate among all these formats, especially ability participants. A possible explanation it does not eliminate player after loss, while takes history comparisons into account. results interesting for emerging esports, where designs yet solidified.