What I should have said is that it is a heuristic, which is generally believed to hold in all cases where between group variation is less than within-group variation. (Which is why it’s okay that athletics are gender segregated)
Well, depending on the metrics you use, within-group variation for men’s and women’s athletic ability is much greater than between-group variation. After all, the difference between the most and least athletic women, or between the most and least athletic men, is much greater than the average difference between men and women, or the difference between the top men and the top women.
Between-group differences can be small relative to within-group differences (certainly significantly smaller than the difference in average athletic ability between men and women,) while still dominating representation of the groups at the tail ends of an activity.
Well, depending on the metrics you use, within-group variation for men’s and women’s athletic ability is much greater than between-group variation. After all, the difference between the most and least athletic women, or between the most and least athletic men, is much greater than the average difference between men and women, or the difference between the top men and the top women.
Between-group differences can be small relative to within-group differences (certainly significantly smaller than the difference in average athletic ability between men and women,) while still dominating representation of the groups at the tail ends of an activity.