the fact that skin color can be a proxy for socioeconomic factors does not change the statistics. Do you investigate why a rapist has raped someone and then ignore it if the reason is socioeconomic factors?
If applying your logic on skin color leads to discrimination then maybe it's discrimination even when the discriminated party is males.
It doesn't but it contextualises them. An inability to recognise that is a signal.
Have you seen any correlation between socioeconomic factors and perpetrators of sexual assaults?
Recognising that one group commits the majority of certain crimes isn't the issue, as you said it's just stats. The issue is entirely in ignoring other factors.