3 ms·

There never has and never will be a good stats-based study on human diet. We are the most complex things we know of in the universe. Confounding factors aboun

There never has and never will be a good stats-based study on human diet. We are the most complex things we know of in the universe. Confounding factors abound. Isolating one is impossible, unless researchers have access to a million identical twins separated at birth and forced to eat a certain diet all their lives, in which case there’s probably a huge human rights crisis going on… and the study would still be invalid because identical twins and authoritarian diet regimes are not the norm. The only way is to map out the entire chemical pathway in the human body and figure out what is good and bad that way. That will take a very long time; in the meantime, I will eat red meat, butter, animal fat, and eggs like my very-long-lived and exceedingly robust ancestors all did. One other approach might be to personally try a diet and get regular blood panels and heart scans done. I believe this is the best way.

copperx 4y ago
I don't think handwaving away an entire field of study is going to benefit you in any way, flawed as that field might be. Living is easy with eyes closed. Dying is too.
- FredPret 4y ago
  Studies with n=10 aren't opening any eyes. They are adding garbage to a crowded information landscape. Even in the larger studies, there are so many issues. Take the first one about fat and heart disease. Riddled with errors you'll find warnings about in most data science for beginners tutorials. Or the one about the Mediterranean diet - zero thought given to confounding effects regarding many possible factors. People who eat the Mediterranean diet also live near the sea, and share genetic, cultural, linguistic, political, and economic factors. Sure, maybe it's the fish and olives, but maybe it's another one of the billion things they have in common versus other Europeans or Americans. This style of studying human metabolism is net-negative. Ignoring it is indeed a benefit. On the other hand, tell me that sugar metabolizes into known toxins, and that's a very different story.

copperx 4y ago

I don't think handwaving away an entire field of study is going to benefit you in any way, flawed as that field might be. Living is easy with eyes closed. Dying is too.

FredPret 4y ago
Studies with n=10 aren't opening any eyes. They are adding garbage to a crowded information landscape. Even in the larger studies, there are so many issues. Take the first one about fat and heart disease. Riddled with errors you'll find warnings about in most data science for beginners tutorials. Or the one about the Mediterranean diet - zero thought given to confounding effects regarding many possible factors. People who eat the Mediterranean diet also live near the sea, and share genetic, cultural, linguistic, political, and economic factors. Sure, maybe it's the fish and olives, but maybe it's another one of the billion things they have in common versus other Europeans or Americans. This style of studying human metabolism is net-negative. Ignoring it is indeed a benefit. On the other hand, tell me that sugar metabolizes into known toxins, and that's a very different story.