alt Hacker NewsWith the caveat that I'm just a random non-expert on the internet who has nevertheless spent too much time reading scattered studies and scholarly opinion articles:
We don't need "more" evidence exactly, but rather a better model of how the effects of exercise map to a given individual's physiology. Exercise is good overall, but it's also considerably overhyped due to a procession of weak and narrowly-applicable results being misconstrued as adding up to a massive pile of benefits that applies to the average person. In reality, the average person does not get anywhere close to the sum of all the touted benefits; they get some constellation of some of the benefits, while other outcomes are flat or even regress [1].
So yes, "exercise is good" at a sufficient level of abstraction, but it's much harder to make the case that it's "good for [specific outcome] for [specific person]". Which is one reason that it's such an obnoxious trend for specific health complaints to be met with generic recommendations to exercise (or exercise more, or exercise differently).
It seems a bit silly to me that in many circumstances it's just way more acceptable for exercise to be zero part of your life than the reverse, and that an argument needs to be made for geting more than even 30 mins worth of movement in a week. A culture of getting exercise by default seems like the way things should be, and the should be an affordance for those who can't rather than a luxurious escape from not having to.
It's not as true in some places in some circles, but it's hard not to notice a difference when you temporarily visit the others.