The pseudo-science comes when claims are made as fact without evidence.

>of sufficiently different populations is just demonstratably wrong.

The difficult lies in translating 'race' to 'populations.' A population of 'people born in India' is extremely well defined and lends itself to stochastic and scientific analysis. Many ideas of race are not so well defined, and can't be put into clearly laid out population of persons. The slight of hand here is your use of the word 'population' when we were speaking of 'race.'