One of the themes of The Black Swan, an excellent book I'm reading now, is that experts are terrible at prediction and that all of us, but especially experts, overestimate the precision of our estimates. A seemingly great example is what Stephen Biddle at the Council on Foreign Relations told George Packer about Iraq:
The likelihood is that it doesn’t become a regional war, but there’s a roughly thirty-to-forty-per-cent chance that it’ll spread. During the Cold War, we spent trillions worrying about infinitesimally small risks. Thirty-to-forty-per-cent chance of a real, honest-to-goodness catastrophe is something that ought to factor into our policymaking now.
Is that estimate of "roughly" 30-40% based on real data? I'm guessing the answer is no. In any case, I am highly skeptical that we can estimate the risk of regional war with that level of precision.
Comments