Here’s the way I see tonight’s election. It’s not quite Obama’s to lose. I think he’ll win, and win decisively (winning both the electoral college and the popular vote), and it’s certainly what I want to see happen, but it’s not a done deal, and there is a road to the White House for Romney. This is what I think is Romney’s best chance, and then I’ll take a guess as to what will happen. (It’s a guess, not a prediction. I’m not Nate Silver.)
Of the polls I’ve been watching (TPM is following a bunch of them), there are only three states that could be considered a tossup, with the margin of victory under 2 percent for either candidate: Florida, Virginia, North Carolina. There aren’t any states that could be considered “lean Romney” right now. Those states he has in his column he has with a polling share of more than 5 percent. What he has adds up to 191 electoral votes (which I think will be unlikely to swing to Obama). If we assume Romney snags all three tossups, that gives him 248. So to win, Romney will need to peel away at least two of the “lean Obama” states, where the president is polling favorably by 2-5 percent. Those states, in decreasing order of electoral clout: Pennsylvania (20), Ohio (18), Colorado (9), Iowa (6), Nevada (6), and New Hampshire (4). Even with Pennsylvania or Ohio in his column, he needs another one to push him up to 270.
Romney could do it, conceivably. I think the most realistic road to the White House would be Florida, Virginia, North Carolina, Ohio, Iowa. But, as they say, it’s an uphill road in both directions. For one, peeling away Ohio and Iowa is going to be tough. For another, Virginia isn’t a done deal by any means. The state that once ran Oliver North for the Senate also elected Tim Kaine governor. The more recent trend has been a shift from the Bible Belt/Old Confederacy Richmond-to-Lynchburg axis to a consolidation of power in the greater D.C. area. Arlington, Alexandria and Fairfax weigh very heavily in the state, and they lean blue. Off the three tossups, Virginia is going to give Romney the most trouble.
If it does? Assuming he takes N.C. and Florida (I think he will), he’ll need Ohio (or Pennsylvania—but I don’t think he can take Pennsylvania: its urban centers are too big and too East Coast), Iowa, Colorado and one other lean Obama state, either Nevada or New Hampshire. A long shot either way if he loses Virginia. Same situation if he somehow loses North Carolina (N.C. went blue in 2008, so it could happen).
If Florida swings for Obama, Romney needs both Pennsylvania and Ohio to win, and at least two more in the “lean Obama” column.
Where does that leave us? If all the “lean Obama” states turn out for Obama, Obama wins. Romney’s needs to do some serious chipping away in those states to pull out a victory.
Here’s my guess for the night: “lean Obama” states—Pennsylvania, Ohio (because of the auto bailout/strong union support), Colorado, Iowa, Nevada, New Hampshire—go for Obama. So does Virginia. Romney takes North Carolina and Florida. There may be some monkey business up in Maine, which could split off one of its four electoral votes to the Romney camp or an independent candidate, but it doesn’t change the math. Nebraska won’t split its vote (Bob Kerry’s getting creamed in the senate race, and he’s Obama’s biggest bellwether in the state).
Final: Obama 303, Romney 235.