> Well you could try to, step by step, experiment with different diets and regularly have a look at your fitness-level, your blood work, your weight, the compos

> Well you could try to, step by step, experiment with different diets and regularly have a look at your fitness-level, your blood work, your weight, the composition of your body (%of water/fat/and so on). A diet is a complex set of variables. I could look at it holistically, ignore individual variables, and try to make sense of it as whole but a) there's more to health than body composition (such as %BF); and b) I wouldn't like to assume that I can read a blood work panel and interpret it as a physician. It's not that I don't do this: I work-out daily, I weight and measure myself daily and I also track everything I eat. However: a) I won't stop getting checked out by competent professionals, so I still put my main trust in physicians; and b) I can't assume that my "diet experiments" are innocuous. Our bodies, as systems, are way more complex than the systems we engineer. We can't really predict the interactions between all the variables we experiment with (and are subject to). What if I seem OK by now with my diet choices, looking at what you mentioned + my subjective well-being, but end up developing something bad in the medium/long-term? Then I won't be happy with my experimentation. :) > Before that you could read a lot. Some studies, some literature. A lot will be on the strange to the bullshit-level. But the self-experimenting part will teach you, what is good for your body and what not. I also do that, quite a lot. I also took a few MOOCs. But I wouldn't say I fully understand what I read, or assume I can interpret a paper in health sciences. It's research-level material, meant to be read by people from the field. I have a PhD in other scientific area: I wouldn't expect someone from other field to make full sense out of a article in my field, so why would I assume I could? > But I would not recommend doing exactly this to everybody - it is something, that subjectively, anecdotally works for me. My point is, although I can fully relate (because, well, that's what I do), I'm not able to say "it works for me". It seems to be working, but if I'm scientifically unable to say "it works" without some degree of certainty in my field, why would I do so in a field in which I'm ignorant?