[vonmoltke]

"[A] large distribution of software engineers" is a group that has already been, to some degree, filtered by aptitude and interest. "[A] person you'd randomly find at your community pool" is essentially the same as randomly drawing a person from the population at large.

Also, "more talented" is a much milder statement than "much more productive", which in turn is a much milder statement than "10x". (The first two, as well, are not saying the same thing. Talent != productivity.) Nobody disputes that there is a distribution of talent or productivity among software engineers. People dispute the magnitude of the standard deviation of that distribution.

Going to your Olympic swimmer example, the number of people who qualify for the Olympics in swimming is a small portion of those who try out, which in turn is a small portion of those who are involved in competitive swimming at all levels. Thus, the number of Olympic-level swimmers relative to the overall swimming talent pool is tiny. I don't think anyone disputes that there are a few dozen, maybe even several dozen, people who are several standard deviations better than the average software engineer. What they dispute is that there are companies full of them, or that trying to hire them is a viable strategy.