Nonsense. The subset of "all males" which is labelled "those males who identify as women" is clearly radically different from, say, a subset labelled "those males who support the Chicago Bears", when it comes to an analysis of the propensity to launch sexual (especially) or physical assaults upon females.
Based on what? You've just made a declarative assertion, with no evidence to support it. Not just that it's different, but that it's
clearly radically different.
My starting position is that males are males, that the aggressive correlation of testosterone is fairly well established, and that the social aspect of having been raised and conditioned as male is not insignificant. I start from the null hypothesis. What information I've found suggests that transwomen are no less violent in general than cismen are, and thus, I find it reasonable to assume that such violence does not demonstrate a material difference in distribution than it would by any other male.
You, on the other hand, are making the assumption that a specific subset of males have a dramatically different rate of violence, in particular sexual violence, than other males do.
I suggest that you go find some evidence to support your assumption, given that your assumption is the one that is significantly different from the null hypothesis.