Monday, January 17, 2011

Today's best prediction is that things are gonna stay mostly the same...

...and don't you let anyone tell you otherwise...

Amplify’d from

We reserve a special place in society for those who promise genuine insights into the future — who can predict what will happen in business, in sports, in politics, technology, and so on. The media landscape is rich with these experts; Wall Street pays millions of dollars every year to analysts to put a precise dollar figure on next year’s company earnings. Those who manage to get a few big calls right are rewarded handsomely, either in terms of lucrative gigs or the adoration of a species that so needs to believe that the future is in fact predictable.

But are such people really better at predicting the future than anyone else?

To find the answer, Denrell and Fang took predictions from July 2002 to July 2005, and calculated which economists had the best record of correctly predicting “extreme” outcomes, defined for the study as either 20 percent higher or 20 percent lower than the average prediction. They compared those to figures on the economists’ overall accuracy. What they found was striking. Economists who had a better record at calling extreme events had a worse record in general. “The analyst with the largest number as well as the highest proportion of accurate and extreme forecasts,” they wrote, “had, by far, the worst forecasting record.”

Their work is the latest in a long line of research dismantling the notion that predictions are really worth anything. The most notable work in the field is “Expert Political Judgment” by Philip Tetlock of the University of Pennsylvania. Tetlock analyzed more than 80,000 political predictions ventured by supposed experts over two decades to see how well they fared as a group. The answer: badly. The experts did about as well as chance. And the more in-demand the expert, the bolder, and thus the less accurate, the predictions. Research by a handful of others, Denrell included, suggests the same goes for economic forecasters. An accurate prediction — of an extreme event or even a series of nonextreme ones — can beget overconfidence, which can lead to making bolder and bolder bets, and thus, more and more errors.

There’s no great, complex explanation for why people who get one big thing right get most everything else wrong, argues Denrell. It’s simple: Those who correctly predict extreme events tend to have a greater tendency to make extreme predictions; and those who make extreme predictions tend to spend most of the time being wrong — on account of most of their predictions being, well, pretty extreme. There are few occurrences so out of the ordinary that someone, somewhere won’t have seen them coming, even if that person has seldom been right about anything else.


No comments:

Post a Comment