Skip to content
Preprint

On minimum-size matroid counterexamples to the multiplicative Merino--Welsh inequality

Oct 2026 · 0 citations · 15 references
Mathematics

Abstract

Let $M$ be a matroid and let $\Tutte_M(x,y)$ be its Tutte polynomial. The multiplicative Merino--Welsh inequality \[ \Tutte_M(2,0)\Tutte_M(0,2)\geq \Tutte_M(1,1)^2 \] was proved to be false for matroids by Beke, Cs\'aji, Csikv\'ari, and Pituk (\emph{Adv. Math.} \textbf{446} (2024), 109674). Let $U_{n,r}$ denote the rank-$r$ uniform matroid on $n$ elements, where $0<r<n$, and let $U_{n,r}^{(q)}$ denote the matroid obtained by replacing each element of $U_{n,r}$ with $q$ parallel copies. They constructed an infinite family of counterexamples of the form $U_{n,r}^{(2)}$, including $U_{33,22}^{(2)}$. In this paper, we prove that every $U_{n,r}^{(q)}$ with $q\geq3$ satisfies the inequality. For $q=2$, we prove that the minimum number of elements in such a counterexample is $62$, attained by $U_{31,21}^{(2)}$. For loopless and bridgeless graphs, Noble and Royle (\emph{Eur. J. Comb.} \textbf{38} (2014), 24--35) showed that every edge of a minimum-size counterexample to the multiplicative Merino--Welsh conjecture lies in a series or parallel class of size two or three. For matroids, we further prove that every element of a minimum-size loopless and coloopless counterexample lies in exactly one series or parallel class of size two.

View source

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.