There are easy ways to see how frequency dependence could work. The key is that no one is arguing bad guys get all the reproduction. At most it seems that only 20-30% of babies are born to “non-social” fathers. There are many strategies, each of which does better in certain circumstances. And men (and women) can try playing more than one “strategy” serially…etc etc. And you are correct in pointing out that women who are on BC may act differently than women who are not.

We know that many of these traits are correlated to testosterone, cortisol and estrogen and that the levels of those hormones are heritable. I believe there even are some candidate loci identified. But that’s just one of a number of ways these could be genetically controlled.

What the paper needs is a good null model that gives the expected # of partners if women were not at all choosy—ie if each man had the same probability of finding a new mate with given effort. Then, you could determine if the difference between # of partners for “good” and “bad” guys can be fully explained by differing effort levels (which would be consistent with a “women don’t, in the aggregate, discriminate AGAINST bad guys or good guys”), if it “overexplains” the data (suggesting that women do in fact avoid bad guys to some level) or “underexplains” the data (suggesting that women may in fact overselect bad guys). After that (and actually surveying women and probably some “field studies”), trying to find QTLs for such preferences (or at least hormonal or MRI correlates) would be a nice touch.

Certainly, culture can impact how genes are expressed. The availability of birth control is one example, and religious edicts are probably another. I will admit that it is very difficult to use contemporary sex data to answer evolutionary questions, but sometimes there are enough surveys, even with methodical problems, pointing to the same answer, that it becomes valid to think they may be valid.

Let me give you a hypothetical example of what I mean. Let’s assume that every time a man and woman have intercourse, there’s a 10% chance the woman gets pregnant (it’s actually about 20-25% during a cycle given normal fertility, which also varies). If Bad Boy sleeps with five women one time, he has a lower chance of successfully reproducing than if Good Boy sleeps with the same woman six times (and presumably, if Good Boy is spacing them out, he stands an even better chance of getting her at the right time of her cycle of ovulation). The Jonason method would assume, however, that Bad Boy is much more reproductively ’successful,’ when in fact he’s just ‘getting it’ with more women.

Actually, in the scenario you describe, Bad Boy has a higher chance of reproductive success. The maximum number of children Good Boy can have is 1; assuming 10% fertility and 6 tries, the chance of him having 1 child is about 47% (1-.9^6, with .9^6 being the probability that she never gets pregnant). The chance of Bad Boy having 1 child after 5 tries with 5 different women turns out to be about 33% (.9^4*.1*5, for the chance that he gets one and only one particular woman pregnant, times the number of women), but you have to also factor in the 7.3% probability that he ends up with 2 children, the 0.8% probability that he ends up with 3 children, 0.045% chance that he gets 4 offspring, and .005% chance that he gets 5 offspring (forgive my sloppiness with significant figures). When you put it all together, the expected number of offspring for Bad Boy after having sex 5 times with 5 different women comes out to about 0.5, while the expected number of offspring for Good Boy after 6 times with the same woman was only .47. The reason is that once Good Boy’s woman is pregnant, it’s game over, while Bad Boy can keep trying and have a chance for additional offspring.

When you increase the fertility rate up to 20%, the discrepancy gets even bigger. Bad Boy’s expected number of offspring is now 1, meaning on average his behavior will produce 1 child. Good Boy’s expected number of offspring (after 6 times with one woman) increases as well, but only to 0.74. It’s impossible for Good Boy’s expected number of offspring to reach 1 even if he slept with the same woman 100 times, since that would mean that he had a 100% chance of getting that one woman pregnant. Bad Boy, on the other hand, now has a pretty decent shot at producing more than one offspring here. (The inherent upper limit of his expected number of offspring is 5, not 1.)

Of course, this doesn’t take into account one of the points you mentioned, namely that Good Boy’s behavior makes it more likely that he will hit his partner’s maximum fertility window at least once. I’m not sure what effect this would have; my gut suggests that it might not make that much of a difference because Bad Boy has 5 chances to hit the fertility jackpot, but if we take anything away from your post it’s that gut feelings are no substitute for actual science.

I guess there’s “But… but if women AREN’T attracted to assholes then… why won’t they sleep with MEEEE? “