Wednesday, January 21, 2004


Arnold Kling has a great long post about the fog of imperfect information we all struggle through in life--we don't really know what reality is and do the best we can. In a recent Slate column where people who usually vote Democrat discussed why they (once) supported the war in Iraq, one of the contributors noted that a reader argued that the decision to invade was good or bad depending on how things turn out. He thought that this was wise way to settle things, but in fact, it is not.

Suppose you have a bet where 90% of the time you will win big, but 10% of the time you will lose a little. On average though, you should expect to come out ahead. Suppose you take that bet and you end up losing--was it a bad decision to take the bet?

Rob Rubin is someone who seems to understand this perfecta well. He also understands that in order to do this type of thing -- weight probabilities, take bets, learn from mistakes, but also understand you can just get unlucky even with perfect information (which you don't have). Michael Lewis takes the easy shot and paints this as an inhuman failing, and while I agree it is a little inhuman, I'm not sure it's a failing at all. The truth is that it takes a certain detachment to tolerate the vertigo that comes with taking bet after bet after bet, knowing that in aggregate you will come out ahead, but that plain old bad luck could push you deep into the red, as could simply being wrong, and it is not clear how you would seperate one from the other.


Post a Comment

Subscribe to Post Comments [Atom]

<< Home