Monday, November 05, 2012

Official Viking Pundit prediction: Romney 285 - Obama 253

I'm absolutely exhausted today but I figured I owed it to my readers (both of them) to put it on the line before the polls open tomorrow.  In addition to the 191 votes allocated by the RCP electoral map, I believe Romney/Ryan will get Florida, North Carolina, Virginia, Ohio, Wisconsin, and Colorado.  Ohio is key.  I don't see a pathway for Romney without Ohio but I think it's very doable in what has traditionally been a Republican state.  Popular vote will be, let's say, 51%-48%.

I wanted to write so much more about predictive models, "Shy Tory", crowd sizes, early voting and such but I'll focus on this one fact: the party affiliation of this race will look nothing like 2008.  In that election, Obama enjoyed a D+8 advantage at the polls.  Rasmussen, which has had a good track record of predicting the electorate, says that the voting electorate will be R+5.8.  This seems like a huge swing but it's based on a very large sample of 15,000 likely voters.  Gallup found a partisan split of R+1 (R+3 with leaners) which would swing 300+ electoral votes to Romney with a landslide in the popular vote.

As an additional note, the WashPost noted that 13% of Americans who voted for Obama in 2008 said they will vote for Romney.  In 2008, Obama got 53% of the popular vote so: 53%*87% = 46% of the popular vote for Obama in 2012.  This is curiously on line with the estimates based on the party affiliation models.  If Republicans and Democrats cancel each other out, Romney's persistent and significant lead among Independent voters will push him over the top.

We'll see tomorrow.

Extra - More predictions for Romney from Michael Barone, Fred Barnes, and Josh Jordan.

GMTA update - Karl Rove also says 285-253 but he gives Iowa and New Hampshire (6+4) instead of Wisconsin (10).

5 comments:

Dead man's hand said...

Saying that the man holding the 10 of diamonds and the ace of diamonds will fill in a royal flush isn't a prediction, it's a prayer.

Anonymous said...

The best prediction is from the ballsy debunker at UnskewedPolls, who just courageously dropped his Romney EV estimate from 359 to 274.

MikeAOR said...

And you thought you had only two readers. I think your prediction is well-reasoned. I also pray it's correct.

Liberals have a fair point that the party id of today's voters may not reach the level of Rasmussen and Gallup's, but they're also missing something when they just assume 2008 turnout levels are going to be repeated. Neither side knows where that turnout is going to fall. Those who pretend they do are the ones substituting card analogies for critical thought.

Horseshoes said...

Missed it by that

Horseshoes said...

much.