We study the relationship between tournaments and random walks. This connection was observed by Moser, who conjectured that the number of score sequences Sₙ~ C4ⁿ/n5/2. Winston and Kleitman came close to finding the correct order of Sₙ. Building on this, and subsequent works by Tak\'acs, Kim and Pittel confirmed that Sₙ=Θ(4ⁿ/n5/2). In this work, we verify Moser's conjecture, using limit theory for integrated random walk bridges. Moreover, we show that C can be described in terms of random walks. Combining this with a recent proof and alternative, number-theoretic description of C by the second author, we obtain an analogue of Louchard's formula, for the Laplace transform of the squared Brownian excursion/Airy area measure. Finally, we describe the scaling limit of random score sequences, in terms of the Kolmogorov excursions, studied recently by B\"{a}r, Duraj and Wachtel. Our results can also be interpreted as answering questions related to a class of random polymers, which began with influential work of Sina{i}. From this point of view, our methods yield the precise asymptotics of a persistence probability, related to the pinning/wetting models from statistical physics, that was estimated up to constants by Aurzada, Dereich and Lifshits, as conjectured by Caravenna and Deuschel.
No takes yet. Share an insight, caveat, or question.
Donderwinkel et al. (2024) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: