Law and Probability, Part II


“You don’t win by predicting the future; you win by getting the odds right.”

-William Bonner

After the most recent presidential election, Nate Silver acquired folk hero status for the accuracy of his models in predicting the election. His model correctly predicted how every state would vote in the presidential race.

Before the election, and probably since the first time humans held public elections, “experts” have made predictions about who would win the election and why.

When I was growing up, my parents would watch a show called the McLoughlin Group, which featured a cadre of pundits, some left, some right, who had spent their lives studying politics. These pundits would talk loudly about their opinions on the show and it made for decent entertainment. At the end of each show, each cast member would be required to make a prediction about some event.

Nate Silver studied these McLoughlin Group predictions in his recent book, “the Signal and the Noise,” and showed that the accuracy of these predictions was no greater than random. Which is to say, that there was no value to these predictions at all.

George Will (long-time panelist on the McLoughlin Group) predicted that Romney would win the election with 321 electoral votes out of 538. Dick Morris of Fox News predicted Romney would get 325 electoral votes, including a win in California.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.