An inequality for means with applications

Abstract

We show that an almost trivial inequality for the first and second mean of a random variable can be used to give non-trivial improvements on deep results. As applications we improve on results on lower bounds for the Riemann zeta-function on the critical line, the determinant of a skew-symmetric matrix with entries $\pm 1$, and on the maximal order of an irreducible character of the symmetric group.