McKay–Wanless conjecture on subsquares in random Latin squares

Let LL be a random Latin square of order nn, and let cmathbbEa(n)cmathbb{E}_a(n) denote the expected number of subsquares of order aa in LL. For 2an/22\leqslant a\leqslant n/2, McKay and Wanless's conjecture.

Ea(n)={n24(1+o(1))if a=2,118(1+o(1))if a=3,o(1)if 4an/2.\mathbb{E}_a(n)= \begin{cases} \dfrac{n^2}{4}(1+o(1)) & \text{if } a=2,\\ \dfrac{1}{18}(1+o(1)) & \text{if } a=3,\\ o(1) & \text{if } 4\leqslant a\leqslant n/2. \end{cases}

The conjecture was a major goal in the study of subsquares of random Latin squares. Its a=2a=2 asymptotic was established up to bounds by Kwan and Sudakov in 2018, and the remaining part referred to in the source was completely resolved by Kwan, Sah and Sawhney in 2022.

Sources & referencesView supporting material

Primary source

Jack Allsop and Patrick Morris, “Universal probability bounds for partial Latin squares”, arXiv:2606.18174 (2026).

Progress summary

Never refreshed

Nothing recorded yet. Refresh searches the literature and the public web for attempts on this problem, and writes the first summary here.

Solutions 0

No solutions have been posted yet.