In other words, maybe it's not that individuals are creepy so much as men "naturally" act more rapey if there are only a few women around.

This is unlikely. The idea that male-on-female rape, in humans, is reflective of forced mating as a reproductive strategy makes some big mistakes because it doesn't factor in how human reproduction actually works.

It's true in a general way that if the cost of your gametes is low, and you can get out of the parental investment, then increasing the number of coital acts is an effective way to buy genetic fitness at reduced cost (part of why mammals tend to be much more promiscuous, in a very broad sense, than birds: birds get their embryo out of Mom and into the world early and let it develop there, which means Daddy has a higher incentive to invest parentally -- though this is only a very broad pattern).

Trigger warning for those who'd rather not hear it described in frank, mechanical terms!

But with humans in specific, rape is not a great reproductive strategy. The odds of insemination are lower, because things like self-lubrication and uterine peristalsis (which make a big difference) aren't typically going to occur. Even post-coital cuddling increases the odds of fertilization. Getting into comparative primatology, humans have conspicuously large penises compared to our relatives who do tend to use force as a basic approach to getting sex (gorillas, who have a harem-style arrangement as their basic stable social model).

Rape has been prevalent throughout human history, but forced copulation doesn't seem to be a leading or even closely-tailing human reproductive strategy. It's probably not an adaptation (though if you insist that pretty much every salient feature of behavior is, or is the proximal outcome of some evolutionary adaptation, you can spin a theoretical picture to justify it easily).