Vanishing success probability for monotone hat strategies
Conjecture 2.2 · arXiv:2208.06858
Status open high confidence
Conjecture 2.2 from arXiv:2208.06858 states that $p_{\mathrm{monotone}}(t) \to 0$ as $t \to \infty$, where $p_{\mathrm{monotone}}(t)$ is the best success probability in Levine's hat problem using balanced monotone winning families. The source paper itself only establishes the weaker result that the success probability is strictly decreasing in the number of players. No follow-up paper resolving this conjecture was found in a targeted web search across arXiv and published literature as of May 2026.
Reviewer notes. No follow-up found. The source paper proves the weaker statement that the success probability is strictly decreasing (Conjecture 1.1 partial progress), but Conjecture 2.2 (which is strictly stronger than Conjecture 2.1 and Conjecture 1.1) — that $p_{\mathrm{monotone}}(t) \to 0$ — remains open. A Fourier-analytic approach paper by Heilman–Tamuz on Levine's hat puzzle was found but its PDF was unreadable; it appears to predate the 2022 conjecture formulation and is unlikely to address it.
Context
This is progressively stronger than Conjecture 2.1, as $p_{\mathrm{monotone}}(t) \geq p_{\mathrm{intersecting}}(t) \geq p_{\mathrm{dict}}(t)$, where winning sets are all balanced monotone families on $\{0,1\}^n$. The authors describe these two conjectures as 'progressively stronger than Conjecture 1.1'.
Source paper
The success probability in Levine's hat problem, and independent sets in graphs
Noga Alon, Ehud Friedgut, Gil Kalai, Guy Kindler · 2023-08-21
https://arxiv.org/abs/2208.06858
PDF source