Most of our popular models of diet are basically unscientific quackery. Maybe they are bad models? You are right of course in simply having a set of rules to follow seems to help.
> Most of our popular models of diet are basically unscientific quackery.
The most conventional holds up: the one that consuming too many calories causes one to gain weight, and consuming not enough calories causes one to lose weight.
If I'd go with what's popular on my gf's FB, I'd end up believing vegan diet is the panacea. We're living in our bubbles. Your mistake in thinking is the assumption that diets like keto and atkins are popular. They're not popular. They're just hypes in certain circles like silicon valley.
Another thing you'll find is that there are government (based on science) recommended daily intake on things like vitamin D, iron, vitamin B12, etc etc and with a keto diet or high carb diet or whatever diet this may be difficult to reach. Although that's not directly related to weight, it is related to health.
Which is probably one of the most widely spread culture on the internet ( I think ), by virtue of being highly tuned into tech.
In this day and age, minority opinions greatly magnify themselves to seem like they are a) common and b) widely accepted.
Diets are no exception, some more than others.
My personal rough rule of thumb is once you reach the low 20s percentage body fat, you either keep increasing cardio (preferably HIIT), or dropping counted calories, until your moving average weekly weight keeps trending down. Fit keto around that or even drop it, and it won't matter (unless you're the same subtype diabetic like me, and then keto is an essential ingredient to avoiding medication).
If you go in once a quarter for a VAP lipid panel (the most extensive lipid panel I could find), then as long as those numbers trend in the right direction, keep on doing whatever works for you. After being diagnosed with Type 2 and having to figure out myself how to beat the fucker into remission, I've concluded that medical science can give rough guidelines for describing metabolic processes, but it is up to each of us to hack our own bodies. Try one change, test the change with measurements, and repeat; commit the changes that work, fork away from the changes that don't, accept failures and setbacks as signals of what not to do, and move on.
The end goal I would like to see is healthy off the shelf long lasting processed food that tastes great. It runs contrarian to "eat local and natural, big food is evil" but it solves a huge problem.
Of course on the way imperfect models will yield bad results and heuristics are all we have. When doctors refused to wash their hands because they didn't know germ theory despite the massive heuristic evidence, people needlessly died.