[ElJefe]

(11-05-2012 01:03 AM)Capitán Peligroso Wrote:  

The reason confidence intervals are a lot harder to predict in midterm elections is because it's hard to tell who is going to turn out to vote. In the case of 2010, a really large portion of the electorate that supported Obama in 2008 didn't vote in 2010. Voter behavior is much easier to predict during a presidential election, and Silver is being pretty conservative in his estimates. Sam Wang over at Princeton Election Consortium is working with very similar data and has much higher odds of Obama winning. If Romney is going to win, they will have to disenfranchise a lot of people in Florida and Ohio, because he sure as hell isn't going to run the table of swing states. We're already seeing this with people spending hours in line at early voting sites in Ohio and Florida, in heavily democratic areas, of course. You never see lines at the polling places in the predominantly republican areas that I live in. Never. And you never will.

Dear God, I hope you're wrong. But I have a hard time imagining RealClear is so off the mark, so I'm expecting Obama to pull it off tomorrow. I find it mind-boggling that there will be a +5D affiliation though, after such a shitty economic performance. If Obama wins, then GWB was probably the worst President ever, seeing how he enabled 8 years of Barry.