As a keto practitioner, the obvious recommendation I'd add for further research is to distinguish between "low-carb" and "<20 grams of carb per day". For achiev

As a keto practitioner, the obvious recommendation I'd add for further research is to distinguish between "low-carb" and "<20 grams of carb per day". For achieving ketosis, "low carb" is a completely orthogonal diet. I'm a little sad that he didn't discover this distinction early on in his findings, since the rest of this is very interesting and well-researched (just irrelevant to the keto question in particular). I'd also add that exercise is another important high-level distinction to make that confounds the rest of the discussion, particularly the analysis of paleo / ancestral diets, which presumably had very high levels of exercise compared to modern sedentarians.