Someone else mentioned the varation in height over time and among populations, so I won't go into that. But:
> No? I'm saying that (as is documented by the olympic records!) the peak ability has increased. Things like https://en.wikipedia.org/wiki/Mile_run_world_record_progress...
This doesn't tell us anything. Assume a normal distribution of running speed among people who seriously practice running. Assume precision timing instruments were invented in the last couple hundred years.
What we would expect to see, even if physiological conditions haven't changed at all, is a steady decline in record mile times as you sample more and more people on the right side of the distribution, until eventually better times require more and more extreme, rare outliers and average times between records begins increasing.
That does not mean that the fastest mile ever was run in 1999, or that athleticism is increasing, or anything. (It might be. There are certainly more people training for specifically competitive athletic reasons than ever before.) But it would not be surprising to learn that the fastest mile ever was run 2000 years ago.
Consider this footprint analysis of a random Australian aboriginine. [1] They calculate him as running at 23mph, which is comparable to Usain Bolt's record setting average speed of 23mph in a 100m sprint. Did we get lucky finding the footprints of one of the fastest Australians in history? Maybe. Or maybe he just wasn't that unusual in an environment where people ran a lot.
https://www.nationalgeographic.com/history/article/20-000-ye...