Edge-disjoint Hamilton cycles in highly strongly connected tournaments.

Status solved high confidence

Thomassen's conjecture was proved by Kühn, Lapinskas, Osthus, and Patel (2014), who showed that every strongly $f(k)$-connected tournament with $f(k)=O(k^2\log^2 k)$ contains $k$ edge-disjoint Hamilton cycles. Pokrovskiy (2017) subsequently improved the bound to $f(k)\le Ck^2$, which is optimal up to the constant.

Cited literature (2)

Reviewer notes. The conjecture as stated (existence of $f(k)$) was proved by Kühn-Lapinskas-Osthus-Patel (PLMS 2014) with $f(k)=O(k^2\log^2 k)$, and the optimal-order bound $f(k)=O(k^2)$ was established by Pokrovskiy (IMRN 2017, arXiv:1406.7556 from 2014). Both papers postdate the OPG posting of 2013-03-08.

Auto-reviewed 2026-05-08 with claude-sonnet (subagent) (web search enabled).

Conjecture. For every $ k\geq 2 $ , there is an integer $ f(k) $ so that every strongly $ f(k) $ -connected tournament has $ k $ edge-disjoint Hamilton cycles.

Discussion

Kelly made the following conjecture which replaces the assumption of high connectivity by regularity. Conjecture Every regular tournament of order $ n $ can be decomposed into $ (n-1)/2 $ Hamilton directed cycles. Kelly's conjecture has been proved for tournaments of sufficiently large order by Kühn and Osthus [KO].

Bibliography

  • [KO] Daniela Kühn and Deryk Osthus, Hamilton decompositions of regular expanders: a proof of Kelly's conjecture for large tournaments, Advances in Mathematics 237 (2013), 62-146.
  • [T] C. Thomassen, Edge-disjoint Hamiltonian paths and cycles in tournaments, Proc. London Math. Soc. 45 (1982), 151-168.