The trouble with math is that the exceptional mathematicians (including theoretical physics/computer science) are the stereotype, and the regular mathematicians

The trouble with math is that the exceptional mathematicians (including theoretical physics/computer science) are the stereotype, and the regular mathematicians are actually quite different from that. So in fact it's the stereotype that's wrong, not the claim that exceptional mathematicians are not very mathematician-like.