February 23, 2015

BRING BACK ADMIRAL POINDEXTER:

Oscars 2015: An Excellent Night for Prediction Markets (David Leonhardt, 2/23/15, NY Times)

The prediction markets had another good year in predicting the Academy Awards. Perhaps more important, the categories that the markets seemed to get wrong underscored the ways in which such markets are useful without being clairvoyant.

Of the 24 categories, the favorite won in 19, or 79 percent. If anything, that basic percentage understates the markets' predictive accuracy, because they did better in the major categories, which receive more attention and, in many Oscar pools, are worth more points. The favorites won in all six of the major categories, including best director, in which Alejandro González Iñárritu ("Birdman") was a slight favorite when the night began.

The pool in which I've played most often over the years awards five points each for the six major categories, three points for seven other categories (like the two screenplay categories and best documentary) and one point for all remaining categories. With these rules, someone who picked all of the favorites last night would have received 55 out of 64 points, or 86 percent of the total.

Shutter the Intelligence services and replace them with open source prediction markets.

Posted by at February 23, 2015 6:21 PM
  

blog comments powered by Disqus
« TOO TIMID TO BE CALLED REFORM: | Main | WHY NOT JUST CLAIM TO BE A NEWS ANCHOR?: »