Every HN thread on diet mentions Gary Taubes' book Good Calories, Bad Calories, so I'll be the first to do it here. The book makes a strong case that the key y

Every HN thread on diet mentions Gary Taubes' book Good Calories, Bad Calories, so I'll be the first to do it here. The book makes a strong case that the key you mention is in fact a fallacy. There is a correlation between calorie surplus and weight gain, but saying the first caused the second is like saying (to use an example from the book) that alcoholism is caused by drinking too much — tautological, and not useful in fixing the condition. Taubes goes on to argue that because of an underlying metabolic condition (induced by a carb-heavy diet, as it turns out), your body decides to gain weight and so makes you eat more calories than you burn. So it's not that you gain weight because you eat too much, it's that you eat too much because you're gaining weight. Edit: sorry to rsheridan6 for using the "true, and not useful" turn of phrase without having realized someone else beat me to it. Guess that means it's a good one here.