PREDICTION IS HARD.

PREDICTION IS HARD. About a year ago, I posted here about FiveThirtyEight, a political website which extends sabermetric statistical analysis (that is, statistical analysis of baseball) to politics. The moving spirit of FiveThirtyEight is Nate Silver, whose PECOTA system is at the heart of Baseball Prospectus. This post from VegasWatch discusses how PECOTA had a bad year in 2009. Out of five different statistical systems for predicting the final standings of major league baseball teams, PECOTA had been the leader for the years 2005 to 2008. For 2009, it had the least success out of a larger group of 16 systems.

This entry was posted in Baseball, Politics, Sports. Bookmark the permalink.

2 Responses to PREDICTION IS HARD.

  1. Nick says:

    Steve Phillips being fired was a blow to that annual piece Vegas Watch runs, as his pre-season predictions were always a source of amusement.

  2. Phil says:

    One test is how well the predictions do compared with simply predicting that the standings will be the same as the previous year. Nick, from 1961 until the leagues went to three divisions, I collected predictions for the year from friends. Each year there was a theoretical monkey who simply repeated the previous year. Two or three predictions each year beat the monkey, but it was hard to do.

Leave a Reply

Your email address will not be published. Required fields are marked *