>As expected, physical activity level, PAL, was greater among Hadza foragers than among Westerners.

Within a standard deviation. Higher on average, but not much, and not significantly.

No where near enough to exert selection pressure.

Within a standard deviation (just) of the hunter gatherers, but not for Western males.

PAL is 2.26 +- 0.48 for Hadza male and 1.81 +- 0.21 for Western Male. That's an average 25% difference in activity. I'm not a biologist but it seems like that could surely be significant.

Especially if the protein being talked about is generated by strenuous activity above some level. 25% total difference in activity could be a very large difference in activity above a threshold, since both groups are going to spend a lot of the day sitting around, sleeping or walking slowly.

Selection pressure doesn't care about standard deviations, believe it or not. Even a small difference can exert evolutionary pressure over a long period of time.

Minute changes are actually a huge part of evolution. Evolution is much more about minute changes than it is about huge (standard-deviation-level) changes.