All that said, actually waiting for FDA approval does seem a lot less risky
https://medical-dictionary.thefreedictionary.com/anecdotal
I think it's fair to say that medical research is not scaled up anecdotes.
While I find the anecdotes compelling. I would assume that to make multiple anecdotes evidence you'd have to at least register these anecdotes systematically, otherwise you only hear the extreme ones that people like to tell.
A temperature sensor that you know isn't desiged to work at subzero is reporting -30 consistently. It may be -40 or -20 but you know it's below zero.
Someone reporting symptom relief from depression, (which is something who's most severe symptoms are by definition subjective,) after taking psilocybin is a good data point. It without a doubt has helped.
Why, how, and discovery of caveats notwithstanding; it's something worthy of understanding, and with anecdotes (plural) consideration for use * . Just because the picture is incomplete, doesn't devalue the existing data points. Which is the thing I was trying to push back on. My data points may be weak, or noisy. But my data points are worth more than your scepticism. Full disclosure, I'm a card carrying sceptic saying that. I always think there's more to be understood, but merely devaluing the data points you currently have is not the way to go about it.
Ask questions, don't raise complaints.
* : some restrictions apply