I understood that is a distribution, not correlation. So, in an abundance context, you have the blue distribution, in scarcity you have the black.
Not that I personally agree with this ideia.
Pretty sure 'people' is not a sensible axis. Unless it's "number of people" but that doesn't seem right.
I think that’s number of people per fitness cohort, the distribution matches the text. Is there empirical data to support it? I kind of doubt there is a second peak instead of the long tail.