http://www.newyorker.com/online/blogs/jamessurowiecki/2009/
02/obama-and-proba.html James Surowiecki: “Decisionmakers ... should try to come up with their best estimates of how likely a policy is to succeed, as well as its potential payoff, and then pursue the policy with the greatest expected value.” This is not a good idea! If this were followed, no one would ever buy insurance, since the expected value of any sort of insurance purchase is less than zero. (By and large, insurance companies make money, meaning that policy holders lose money.) Pursuing a “highest expected value” policy would mean bets on, amongst other things, extremely unlikely events with huge payoffs. 00:11