I want to point out that this question doesn't quite test for the right thing. One way an organization like CFAR can cause extreme life improvements is by encouraging participants to do extreme things generally in order to increase the variance of alumni outcomes after the workshops. That leads to potentially many extreme improvements but also potentially many extreme... disimprovements? And the latter are harder to notice because of survivorship bias. (There's also regression to the mean to watch out for: you expect the population of CFAR workshop attendees to be somewhat selected for having life problems and those could just randomly improve afterwards.)

I expect the main benefit of CFAR training should be that it improves median outcomes; that is, that it improves alumni ability to consistently win. But this is hard to test for by asking for anecdotes: it would be better to do statistics.