> how many randomly selected people would you need to produce a score aver 200 on an IQ test, produce a proof of Fermat’s Last Theorem or build something like the SpaceX Falcon.

It's been proven time and again that IQ numbers are more or less meaningless. Setting that aside, I think the point of the article is exactly that you might not need a huge sample to find someone with the potential to do these things, key word here being "potential", nor these things require a different order of magnitude in talent or intelligence, just the right combination of drive, luck, inspiration, opportunity, etc...