The shape of the curves are such that men have a very small advantage on the Math SAT at the median, which gets larger up the scale. So using the disparity between the sexes at the median makes the author's argument even worse.
Even being 2:1 why would you expect those results being reflected anywhere else? Why would an employer hire from one of the two pools of people where one haves 50% less chance of being good enough for the position?
If an employer just hired people with perfect SAT math scores, you'd expect a 2:1 distribution. If an employer just hired men, because they were twice as likely to have a high SAT math score, then the employer would be retarded because for both genders there is over a 99% chance that any given person will not have a high SAT math score.
