As an epistemic issue, I have to use all the evidence available to me. What I've noticed is that human physiology is a dynamic equilibrium of a massive number of variables and the current state of our analytic tools have had immense trouble evaluating it efficiently. Ancestral eating patterns looks like worthwhile evidence under the circumstances.
Incidentally, essentially no one is blindly promoting traditional eating. Folks are using it as a guide and together with science to make decisions in the face of limited understanding. One of the most popular ancestral diet proponents is Stephan Guyenet, a Ph.D. neurobiologist who studies the link between the brain and obesity for a living[1]. Everyone is for science and no side on the issue owns it.
Which means that in theory, any random undergrad in a Biochem major is equally qualified to speak on the subject. This is not to put down Dr. Guyenet's work, it's important as obviously there's a neurological component to many kinds of obesity. But then to extend that work into his advocacy for traditional dietary guidelines is a stretch.
If he was just an undergrad student in Biochemistry with a blog telling you to do all of the things exactly opposite Dr. Guyenet, would you follow that instead? Why not? That person would be just as educationally qualified. I suspect the reason has to do more with what diet you've chosen to follow a priori and not with Dr. Guyenet's advocacy. I'm not going to pick apart specific issues with this blog, but a casual perusal of it shows lots of fad diet advice and logical "appeal to tradition" fallacies. "We should eat this way because people used to eat this way".
The reason people used to eat that way is because they had nothing else available. That's it. There's no secret "the ancients had figured out nutrition from their folk ways" that we should be parroting. There are lots of ingredients in traditional diets that we understand today to be outright poison (like bracken). And lots of ingredients our ancestors thought were poison but are just fine (like potatoes).
Traditionally, people didn't even know what to eat to stop scurvy. A nutritional disease which killed millions and turns out to be trivial to treat.
What we know about good nutrition (using the power of science) happens to be in pretty broad strokes with pockets of ultra-specificity where we've bothered to do serious research. Science has gotten it pretty wrong as well in the past and probably will for some time as the study of nutrition matures. And you're absolutely right, we can't wait 200 years for nutrition science to mature in the way that Physics has, we have to eat today (and multiple times at that).
All I'm saying is that nutrition blogs are a dime a dozen, from people with all kinds of credentials, and lots of them are full of all kinds of advice. "People used to eat this" is just cargo culting dietary advice (and then trying to backfill in justification after the fact) and I would chalk it up to pseudoscience and just make my own decisions based on reasonable advice like "eat in moderation, eat in variety".
Scientific models haven't been all that great either, but science hopefully will eventually self correct as nutritional models become more sophisticated. If there is light at the end of the tunnel it will be because of the eventual understanding that the scientific method provides, not reverting back to traditional models.
Traditional models don't self correct and there are plenty of instances of people consuming outright poisons under such systems or people following traditional nutrition models finding themselves surrounded by an abundance of previously rare foods that the modern food supply system easily provides, and practically gorging themselves on it to negative health results.