Nate Silver has pumped vast quantities of polls, surveys, and other data into his election model. Dick Morris has one theory and a gut feel for the race. The two election prognosticators could not be farther apart in their methodology or their projected outcomes. Silver predicts a close, but very likely victory for President Obama. Morris projects a win for Romney that exceeds Republicans’ wildest expectations.
Silver’s model has been very successful in its short lifespan. It includes poll data, economic data such as the stock market and unemployment numbers, and demographic information.
Recently, conservatives have been angered by the model’s confidence in Obama’s re-election chances. While Silver’s Electoral College projection nearly mirrors the RealClearPolitcs no toss-up map, his level of confidence indicates Mr. Romney is a significant underdog. His reasoning is based a pretty firm foundation of polling. According to RealClearPolitics, Romney leads by less that two points in Florida and is in virtual ties in Colorado and Virginia. Romney’s best path to victory is holding Florida, have the ties break his way in Colorado and Virginia, while overcoming a two-point deficit in Ohio. Obama only needs to win states where polling places him ahead. He could even lose both Iowa and New Hampshire where he has small leads and still win.
While Silver’s model correctly credits Obama’s poll advantage, I believe he errs in making relatively small margins appear insurmountable. As I pointed out in the Daily Caller, six of thirteen battleground polling averages were off by at least three points in 2008. Three of the six (Iowa, Nevada, and New Mexico) were off by six points or more. This seems to indicate that a two-point lead may not always be worthy of the certainty Silver allots it.
Fast and Furious: Family of Slain Border Patrol Agent "Baffled" Obama's Executive Privilege Used to Protect Holder's Wife | Katie Pavlich