Will Silver Be Correct Again In Election 2014



It’s two weeks until Election Day, and I’m curious whether statistician Nate Silver will be correct again this year? Silver runs the popular blog www.fivethirtyeight.com and bases his election predictions more on the laws of probability than political analysis.

In the 2008 presidential election, Silver correctly predicted 49 of 50 states and followed that with a perfect 50 of 50 states in 2012. Silver became popular on the political left because in both elections he predicted that President Obama would be the winner and his blog appeared to a wider audience in The New York Times. Ironically, I noticed some recent left wing social media criticism of Silver because of his 2014 predictions.

In 2012, I believed the GOP talking heads that claimed the race between President Obama and Mitt Romney to be closer than the polling indicated and actually thought Romney might win. I was wrong, but Silver was right, and he became known as the most accurate forecaster in the country. Silver’s 2014 analysis is getting a lot of attention because he predicts that Republicans have a 65% chance of taking back the US Senate (LINK:  www.fivethirtyeight.com/interactives/senate-forecast). I also expect a GOP Senate win on November 4 based on the current polling, but recognize that a lot can happen in two weeks and believe some races are still too close to call.

Jim Maisano

One thought on “Will Silver Be Correct Again In Election 2014

  1. Silver is a great prognosticator, but looking at the results and saying Silver was 51/51 while true, is not a good assessment of his results. In reality, he correctly picked the 10-12 toss up states perfectly. Any with an ounce of political knowledge could have gone 40/40 as most of those states (NY, California, Texas, Alabama, Mississippi to name a few) were totally noncompetitive.

    I hope Silver is right about the Senate results this year, but he has had some really bad results just two years ago in the Senate projections. He had ND and Montana as less than 50% odds of going Democratic and both did. He had Heitkamp projected to win only 8% of the time. Again he probably was 28 for 30 or something close to that, but most of those states were not competitive. Not to take anything away from him, he is great at what he does and I love the website, but I have to call him out on some of his predictions.


Comments are closed.