Meta: There are many correct answers to this question and I'm sure Less Wrong folk will give at least five or ten. However, I've noticed that Less Wrong folk don't seem to be much better than other reasonably smart people at noticing which causal factors contribute most to an outcome when there are many plausible causal factors. Insofar as my perception is evidence this might be a good thing to keep in mind. (Representative datum: The ridiculous number of hypothesis-like-things that attempted to explain large swaths of "akrasia". One should be pretty damn suspicious when ones explanation posits that a single causal factor largely explains the lack of a complex thing.)