Monday, November 5, 2012

Goodnight Ohio?

As far as I can remember throughout this cycle -- and to be fair, I was mostly following my own advise surprisingly well and ignoring quite a bit of the state polling until after Labor Day -- Ohio has been the tipping point state. And in fact, Nate Silver still has Ohio as the overwhelming favorite to be the tipping point state: 50% chance, with Virginia next at 12%.

But in the most recent Pollster averages, that's not what it looks like any more. There may be a more likely path for Romney. Granted, at this point the more important truth is that Obama leads solidly, and if the polls are for some reason wrong then it's not particularly likely they'll be uniformly wrong across all states, but at any rate the numbers aren't quite saying what they have been saying.

Pollster currently has Barack Obama leading in states with 277 electoral votes, with 70 votes in tossup states. So start by giving each tossup to Mitt Romney -- that's North Carolina and Florida, where he's leading in the Pollster averages, along with New Hampshire, Colorado, and Virginia, each of which has small Obama leads. That makes the total count 277 Obama, 261 Romney.

There are currently four light blue states -- that's "leans" Obama, which if I recall correctly means an Obama lead of 2.5 through 5 points. Here's the list, with electoral votes and the Obama lead

Wisconsin (10)  4.4
Ohio (18)          3.4
Nevada (6)       3.2
Iowa (6)           2.7

Just Iowa or just Nevada wouldn't be enough...but both of them together would be, for Romney.

Now, obviously, there's no real difference between a 3.2% lead and a 3.4% lead, and if Romney gets Ohio then he doesn't need either Nevada or Iowa. But it does appear not only that there's another way to get there for Romney, but that it's a technically very slightly shorter, at least if the polling average is precisely correct and we assume uniform swing. Of course, if swing isn't uniform, then it matters that Romney would need two states (plus the tossups) instead of just one.

(I should also mention here somewhere, if only because sharp-eyed readers will wonder about it, that most of Nevada has already voted. That's true. If you think that makes the polls in Nevada more reliable, then there's that. However, it might not).

Meanwhile, and again just using the Pollster estimates, the electoral college advantage for the Democrats is a bit smaller than it was, but still fairly large. The national vote estimate right now is Obama 47.9, Romney 46.8, or a 1.1 point advantage -- compared with the 3.2 point uniform swing that would be needed to move NH, CO, VA, IA, and NV from Obama to Romney. So that's a 2 point electoral college advantage -- down from about 3 at one point last week.

The larger point, I suppose, is that there's no point worrying about these tiny differences, especially for those of us who are not making choices about limited resources. It is worth knowing that there appears to be an electoral college tilt. On the other hand, if it turns out that the polling is generally correct but wrong in only, say, Ohio, then that tilt could disappear instantly.


  1. if swing isn't uniform, then it matters that Romney would need two states (plus the tossups) instead of just one.

    Right, and IA and NV are about as different as you can get, demographically speaking. So the chances of both swinging in the same direction is less than, for example, IA and WI swinging together, or PA and OH swinging together. Bottom line: OH is still pretty much a must-win for Mitt.

  2. Just released NYU study of President Obama and Mitt Romney's body language shows on a word by word basis what each candidate emphasizes.

    be sure to click on the more info tab and read the information just so you have seen it.

  3. There has been some very good reporting indicating that the polling in Nevada has underestimated the Spanish-speaking vote there, and that Obama's lead is probably higher than 3.2 points. The Romney miracle is more likely a result of putting Wisconsin in the R column.

    1. You're talking about Ralston, of course, the guy who pegged the Reid win. Bear in mind that there's also a significant Spanish vote in Colorado, as well as Virginia (of all places!).

      I think that if there's a significant bias, it's in systematically excluding Latinos, either through English-only, lack of cell phone polling, or what time of day they call.


Note: Only a member of this blog may post a comment.

Who links to my website?