This is the website of Abulsme Noibatno Itramne (also known as Sam Minter). Comments here or emails to me at are encouraged... or follow me on Twitter as @abulsme.



Electoral College: Romney rebound continues, but not yet enough to win

Three states change categories today. Two states move toward Obama, one state moves toward Romney. But Romney’s state is bigger than the other two combined, so he wins the day. Lets look at them each, from smallest to largest:

The five poll average in New Hampshire had been generally sticking at an Obama lead less than 5%, but with two related polls (both from WMUR/UNH with and without pushing leaners to express a preference) showing Obama with a 15% lead in the state, the average spikes upward These data points may well prove to be outliers, they certainly seem like it initially, but for the moment they move Obama’s lead well out of the “Leaning” category, and indeed almost all the way to the “Strong” category. For the moment though, Obama’s lead stands at 9.8% and we put New Hampshire into the “Weak Obama” category. (Note that even if we’d only counted one of these two polls, the state would still end up as Weak Obama.) This has the effect of moving New Hampshire out of Romney’s reach for the moment.

In Nevada things look less like they are being influenced by an outlier. Instead three of the last five polls in the state now show leads over 5% and the average now jumps to a 5.7% Obama lead. There does seem to be some real movement toward Obama in those last few polls. Will it last? Who knows. But for the moment Nevada also moves out of Romney’s reach as well.

This would all look pretty bad for Romney today, except for the third state:

Florida, where the five poll average had just hit 5.0% exactly last Thursday causing me to move Florida to “Weak Obama”, now drops back below 5% again, putting Florida back into the much more familiar “Weak Obama” category where it has been for most of the last year. The spike upward appears to be over, and Florida has reverted to form as being a close state. Well, sorta close. The five poll average still has Obama at a 4.1% lead in Florida, which is not insubstantial. And he has been “slightly ahead” in the state for most of the year. But this is still a small enough lead that given the right sorts of events it could disappear quickly. So once again, I list Florida as being in play.

And that makes the current status:

Romney Obama
Romney Best Case 263 275
Current Status 191 347
Obama Best Case 191 347

So Romney’s best case if he wins all the close states… now once again including Florida… is still to loose, but now by a pretty small margin. Obama 275 to Romney 263. If he can get Nevada back in play and win that, we could have a 269-269 tie, which would probably end up going for Romney in the House. A tie would be a lot of fun. Still an unlikely scenario though.

In general, to win, Romney has to win Florida (29), North Carolina (15), Virginia (13), Colorado (9) and Iowa (6) for sure. All of those are “must win”. And Romney is currently behind in all of them to various degrees. They are all pretty close though and flipping them is not unrealistic. On top of that he then needs to get at least one of the big states in the “Weak Obama” category, or two of the smaller ones. Those are all states where Romney is behind by more than 5%.

Although the map looks a bit better than it did a few days ago, it is still a very bleak picture for Romney at the moment.

Note: Chart and map from the 2012 Electoral College Prediction page. Both assume Obama vs Romney with no strong third party candidate and show polling as it currently exists. Things will change before election day. On the map red is Romney, blue is Obama, gold states are too close to call. Lines on the chart represent how many more electoral votes a candidate would have than is needed to tie under several different scenarios. Up is good for Obama, Down is good for Romney.

Edit 2012 Oct 3 09:44 UTC to add final note.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.