I recently stumbled on this realization, when I was talking with friends about the myriad of problems, inconsistencies, and annoyances of the Star Wars universe. After the umteenth conflict was brought up, agreed upon, then given a plausible-sounding explanation, I was a little bit shocked at how amazingly easy Lucas' Advocacy came to me. It wasn't a big extrapolation to see that Devil's Advocacy was nearly as easy.

I think I've come to enjoy role playing games less since stumbling upon Less Wrong. Aside from being a communal activity, I find diminished value in obsessing over a map that is only as detailed as the players and DM construct. Maybe if the other players were as interested in models of reality as me it would be different. But in a game where you just Use The Force to fix all problems, clever tricks are largely unrewarded and thus unsatisfying.

Far more interesting to try and figure out this Bayes stuff. For instance, I saw the results of a study in a newscientist article. The article claimed that of 108 women, 75 claimed to have intuitive knowledge of their baby's sex, and of them, 60% were correct. I was trying to figure out the probability of a woman being intuitive, given that she was correct. It was a failure, since they did not say how correct the women who did not have intuitions were. But it was fun, and deeply satisfying to apply knowledge to a problem. I just realized the article linked to some press release, which seems like a bunch of woowoo. Now I really want to know how accurate the women who didn't profess magical intuition dream powers were.