I think that some of the motivation for discussing whether short skirts increase stranger rape risk is not-true-rejection behavior to avoid discussing acquaintance rape - a far more frequent kind of rape that requires very different responses.

I'm not certain how different the responses actually are in this particular context. We might expect effective risk-minimization behavior to look different overall, but if there's anything to the theory that victims' social presentation styles are a risk factor for rape in general, I'd expect them to be a risk factor for acquaintance rapes unless we have some particularly good reason to think that rapists of strangers have unique psychology in this respect.

Indeed, the only empirical data I remember being linked in this thread found its strongest links for date and spousal rape, though the association looks to be on the weak side either way.