Interesting review (that first graph in particular is definitely non-obvious to me), but under the diet section I think it understates the current body of research regarding low-carb/keto diets - some studies have begun to show energy expenditure[1] and insulin[2] benefits that uniquely exist with low-carb. I think we'll find out soon enough that a lower-carb approach, given the current American diet, does definitive…
>"During the test phase, high, moderate, and low carbohydrate diets varied in carbohydrate (60%, 40%, and 20% of total energy, respectively[...]We randomly assigned participants who achieved the target weight loss to high, moderate, or low carbohydrate test diets for a 20 week test phase.)"
Second study:
>"14 days of a low-carbohydrate diet (21 g of carbohydrates per day)"
So the first study went for 5 months but didn't include an actual low-carb diet:
100 g -> ~400 cal -> ~20% of 2k cal/day
The second study only went for 2 weeks but did include a low carb diet: 21g -> ~84 cal -> ~4% of 2k cal/day
Now someone just needs to combine these methods into one valid study.