I sorted the whole index by evidence and it is bleaker than I expected

Went through all thirty-four compounds on the index and counted the labels.

The strong pile is tiny and is almost entirely approved metabolic and hormone drugs that went through phase 3 like any other medicine. The none pile is large and contains most of the compounds with actual followings behind them.

Nothing in that is a surprise individually. Seeing the whole distribution at once is a different experience.

The correlation is the interesting part, and it runs the opposite way to what a newcomer would guess.

Enthusiasm is inversely related to evidence, fairly reliably. The compounds with completed trials are discussed calmly and mostly by people who have a prescription. The compounds with nothing are discussed with real passion.

The mechanism is not mysterious once you say it: a completed trial puts a ceiling on the claim. No trial, no ceiling.

And there is a filter upstream of that. Compounds that work well enough get developed, approved and prescribed. Compounds that do not get sold in vials.

So the grey market is, to a first approximation, an inventory of things that did not make it — plus a smaller set that did make it and are being sold around the price.

The second group being the incretins, essentially. Which is a genuinely different situation and gets flattened into the same word.

It is a really important distinction and I wish there were a word for it. "Unapproved because it failed" and "unapproved because it has not finished" and "approved but people are buying it outside the system" are three completely different risk profiles wearing one label.

And a fourth: approved, but for something else. Methylene blue and SS-31 both sit there, and both get sold on the strength of an approval that assessed a different question entirely.

Four categories in one word. No wonder the conversation is a mess.