In this paper, we study the relation between a novel personal entropy measure for online signatures, based on local density estimation by a Hidden Markov Model, and the performance of several state-of-the-art classifiers for online signature verification. We show that there is a clear relation between such entropy measure of a person’s signature and behavior of the classifier. We carry out this study on a Dynamic Time Warping classifier, a Gaussian Mixture Model and a Hidden Markov Model as well. Signatures were split by the K-Means algorithm in three categories which are coherent across four different databases of around 100 persons each: BIOMET, MCYT-100, BioSecure data subsets DS2 and DS3. We studied the impact of such categories on classifier’s performance with a larger signature data subset of DS3, of 430 persons