> But if there are more Asians in China than Korea, which obviously there are, they should be the more accurate representation of Asian genetics. The tallest p

> But if there are more Asians in China than Korea, which obviously there are, they should be the more accurate representation of Asian genetics. The tallest province in China is still shorter than the average European height. And Asians aren't a monolith, there can be genetic variation between Han Chinese, Koreans and Japanese in a phenotype such as height. The average height in Japan for men is 161cm despite being a rich country (so it is likely driven by genetics, not nutrition). > You say it shouldn't matter yet you also admit your argument is weakened by using the bigger data set. I say no such thing. I say it is less prone to nit picking. The argument is unchanged. > By definition, professional athletes are genetic exceptions. So it would make no sense to compare the average member of any population to the elites in the NBA and make claims about race in general. This is statistically illiterate. If you have two distributions with the same kurtosis, and one distribution has a mean, say, 0.4 standard deviations lower than the other, then the frequency of extreme right-tail performance is going to be much less common in one population than the other. This is just elementary statistics. Yes, you can have an Asian who is extremely tall and therefore not disadvantaged in the game of basketball, but that is simply not relevant to the statistical claim I'm making. If you have two equally sized racial groups, one has 2% of the population that are extremely tall, and the other has 0.3% of the population that are extremely tall, then the former racial group has an advantage due to this genetic difference in this specific domain, and any elite performers in this specific domain are likely to come from the former racial group than the latter (all else equal).