[geofft]

I can't watch the video right now, but assuming the transcript accurately transcribes what was said then:

JS: I did it by assembling a wonderful group of people. When I started doing trading, I had gotten a little tired of mathematics. I was in my late 30s, I had a little money. I started trading and it went very well. I made quite a lot of money with pure luck. I mean, I think it was pure luck. It certainly wasn't mathematical modeling. But in looking at the data, after a while I realized: it looks like there's some structure here. And I hired a few mathematicians, and we started making some models -- just the kind of thing we did back at IDA [Institute for Defense Analyses]. You design an algorithm, you test it out on a computer. Does it work? Doesn't it work? And so on.

That's basically what I said. I didn't claim that he had a brilliant and correct answer before he started. (Nor am I claiming that you need to have a brilliant and correct plan for a cryptosystem before you start.) I'm claiming that he had the resources to make experimentation not ruinous (the first part of that story is how he acquired those resources, namely, "pure luck" which "certainly wasn't mathematical modeling"), and he engaged in that experimentation and took the scientific method seriously (the second part of that story).