Various folks are linking a New York Times piece on how the frequently-reported disparity in the average number of male and female sex partners is mathematically impossible. Maybe I’m missing something here, but while it’s clearly impossible for the (heterosexual) mean to be different, there’s no strictly mathematical problem with the median being different—which is the claim made by at least some of the studies under discussion—because the median doesn’t tell you anything about what’s going on at the far ends of the distribution. If the most sexually active quarter of the female population suddenly doubled its average number of lifetime sex partners, you wouldn’t necessarily see the median budge.

The Times analysis purports to debunk the old promiscuous-males, chaste-females stereotype. But if you think about what such a difference in the shapes of the partner distributions would require, it would actually fit pretty well. It would entail that the male outliers—men who have large numbers of partners—are mostly sleeping with people at the most active end of the female distribution, whereas the most active women are also finding willing partners in the middle of the male distribution.

Now, this is all so obvious that if the academic researchers and mathematicians the Times quotes are maintaining the position they do, it must be that this picture is also not borne out by the data. In other words, it must be false that there’s some subset of the respective population for which the usual relationship flips, with women in (say) the most active quintile having substantially more partners than their male counterparts. And, indeed, the idea that men would frequently fudge up and women fudge down in self-reports is highly plausible. But it seems awfully sloppy of the Times reporter not to raise this issue and clarify it.