Robert is correct that X passes unchanged from a father to daughter, but that would likely only reduce the overall number of crossovers by about 25%, as a son's X is from his mother and recombines normally. So that reduction in crossovers shouldn't increase the number of matches by very much. Update Dec 25: I posted some analysis I did that discusses the reduction of the crossovers for X by only 25%.
I expect that the criteria that the company is using for X matching is what is causing your larger number of X matches. For example, at FamilyTreeDNA, the criteria for an autosomal match requires at least one segment matching at least of a certain length. The company doesn't go out of its way to post its criteria, but Blaine Bettinger in his Family Tree Guide to DNA Testing and Genetic Genealogy says it may be between 5 and 7 cM. He says for X-DNA, the criteria is twofold: the individual must be an autosomal match AND must share 1 cM on their X-DNA.
Since the X only need share 1 cM to show up, a higher proportion of X matches may exist than you'd expect. Of course, this could mean more false positives (matches that really aren't) for the X than the autosomal, but still it may account for your observation that matches on X occur more frequently.
This lower matching criteria for X, may also be true for other companies, 23andMe, GEDmatch, etc.