This is fascinating. Obama’s re-election next month is guaranteed if he can hold on and win the five contiguous states outlined in orange below, all of which, even after his precipitous decline since the Denver debate, he still leads by between 2.4% and 5.0% in the Real Clear Politics averages:
That map gives Obama exactly 270 electoral votes. It assumes that Romney wins Virginia (currently tied) and Colorado (Mitt by 0.2%), as well as North Carolina and Florida (which I’ve always felt would be in Romney’s column in a close election). The map also posits Romney comebacks in Nevada, where Obama currently leads by 3.0% in the RCP average (and where there’s some reason to believe the polls may be understating Obama’s support) and New Hampshire, where Obama leads by 0.8%. The map even gives Romney the sparsely-polled rural 2nd Congressional District of Maine, which became about 0.8% less Democratic after redistricting, bringing it slightly closer to New Hampshire (it has ranged between 1.5% and 4.5% more Democratic than New Hampshire in recent presidential elections) and thus making it potentially more competitive — if still a bit of a reach for Romney. Yet even though we’re generously giving Romney all of those swing states (and one swing district), Obama would still win, 270-268, because of the Midwestern firewall.
The current poll averages in the “firewall” states? Obama is up by 2.4% in Ohio, by 2.8% in Wisconsin, by 3.3% in Iowa, and by 5.0% in both Pennsylvania and Michigan. Those aren’t exactly landslide margins, but they’re robust enough that you’ve got to take this potential firewall seriously, at least for the moment.
Because of their geographic proximity and rough similarity, these five states are likely to move somewhat in sync with one another in the polling. It would be surprising, for instance, to see Obama stay at +3% in Wisconsin, but suddenly sink to -3% in Iowa. Thus, while it’s entirely possible that Romney’s overall standing will improve enough that he sweeps the Midwestern bloc — or at least its three closest states (Iowa, Wisconsin and Ohio) — it’s harder to see him micro-targeting and “picking off” just one specific Midwestern state without moving the polls in the other. Put another way, he needs to generally get 2-3 points better in the Midwest to have a real shot.
The flip side of this point: if Obama starts showing vulnerability — in the poll average, not just in an individual outlier poll or two — within the Midwestern firewall, that’ll mean it’s time for Democrats to really start to #PANIC. My guess is that the firewall will either hold completely, or crumble completely.
Exit question: Why on earth did Romney write an editorial titled “Let Detroit Go Bankrupt” back in November 2008? I’m not addressing the substance of it, just the politics. He wasn’t a member of Congress; he wasn’t the governor of an affected state; he didn’t have to decisively weigh in, vote “Yes” or “No,” or otherwise take a firm stance. He could have vacillated, waited to see which way the wind blew, and taken a position after-the-fact, as he and so many other politicians have done on so many issues. Instead — to his credit, in a certain respect — he came out and took a stand. And that stand might cost him the presidency. If Obama’s Midwestern firewall holds, the candidates’ respective positions on the auto bailout will surely be a key reason why.
P.S. An earlier version of this post, containing various elaborate scenarios regarding New Hampshire, Maine and Nevada, was erroneous because I made a critical error in composing the map. In flipping states from 2004, I forgot to give New Mexico to Obama. So the post, as originally written, was totally wrong. Oops.
Romney beats Obama at least as badly as Obama beat McCain.
Republicans increase the majority in the House, and take control of the Senate.
Aiken wins.
“Romney beats Obama at least as badly as Obama beat McCain.”
Hmm. Obama won 365-173. If I give Romney every single swing state, including marginal swing states like Michigan and Pennsylvania (sweeping Obama’s entire “firewall”), and Maine’s 2nd District, I can get him to 338-200. Even if I give him New Mexico and Minnesota, which aren’t really on anyone swing-state radar right now, he’s still only at 353-185, 12 fewer votes than Obama. How do you get Romney to 365? Oregon (7) doesn’t quite do it. Washington (12) seems implausible. New Jersey (14)? That’d be 367-171, or 376-164 if you throw in Oregon for good measure. That’s your prediction? Really? I want to go with you to Vegas. 😉
Two words:
Preference cascade
That’s fine, but just so we’re on the same page… if you give Romney all of the states that Obama won by less than 15% in 2008 — basically, all the swing states plus Minnesota (surprisingly, at least to me, Obama only won by 10.24% there) — you’re at 332-206 Romney. Now, to get Romney to 365 or more, you need at least 33 electoral votes from the following group of states:
NM (5) 15.13%
NJ (14) 15.53%
OR (7) 16.35%
MI (16) 16.44%
WA (12) 17.08%
Depending on electoral vote combinations, you need either 3 or 4 of those states, and you can’t do it without flipping either Michigan or Washington; just NM, NJ and OR aren’t enough to get you there.
Remember, Obama’s national margin in 2008 was 7.27%. So Michigan and Washington were, respectively, 9.17% and 9.81% more pro-Obama than the nation as a whole. Thus, assuming the states stay in generally the same positioning relative to one another (and there’s no reason to believe they won’t), that means you probably need something on the order of a 9-10% Romney national popular vote landslide in order to get him to 365 electoral votes or higher.
You think a “preference cascade” gets him from a dead heat (and no, a single outlier from Gallup doesn’t change the fact that the consensus points to a dead heat right now) to a 9-10% landslide in the next 2 1/2 weeks? I’m not saying it’s impossible, quite, but again…if that’s your prediction…I’d really love to go with you to Vegas. 🙂
second set of two words:
Bradley effect
Malarkey. It was already pretty clear before the 2008 election that the Bradley Effect no longer existed, if indeed it ever had. And, to the extent there was doubt, Obama’s election in 2008 further debunked this notion. There was no Bradley Effect in 2008, and there is absolutely no reason to believe the Bradley Effect will magically re-appear in 2012.
I suspect you’ll cite the “incumbent rule” next. That doesn’t really exist, either. (More here.)
Next? “One word: oversampling”? Heh.
Stick with “preference cascade.” It’s an intriguing idea, and one that I can’t conclusively debunk. It could happen. I just don’t see how it happens to the tune of a 9-10% Romney landslide.
OK..since you are actually engaging again, I’ll detail my thoughts, that I summed up as The Bradley effect in a sort of shorthand.
We’ll begin with the 2010 elections. Totally dominated by the Tea Party. Unlike all the left’s astroturf knockoffs, the Tea Party has been successful and is still there. Those people are just as pissed off, and just as ready to vote.
Then we’ll go to the Chick-Fil-A incident. Tens of thousands of people were willing to stand out in the sun for several hours to buy a fast food chicken sandwich. There was no rioting, no looting, no violence at all. In fact, except for counter protesters, there was no anger. Just quiet, widespread support for someone being pushed around by Democratic political machines because of their private beliefs.
Then D’Souza’s movie. Forget what you think of the movie, just absorb the fact that this movie that was, has been and continues to be, ignored if not buried by the MSM is the second highest grossing political documentary of all time. People searched it out and made a concerted effort to see it.
Throw in the fact that 91% of the people that pollsters call hang up on them, and I believe that there is a vast ground swell of people feed up with this administration, quietly determined to do something about it and extremely motivated to vote.
Please add to the above the whole Walker recall fiasco in Wisconsin.
gahrie, those are all very interesting and may perhaps be indicators of higher voter enthusiasm against Obama, but I’m not sure what they have to do with the “Bradley effect.”
Brendan, why do you dismiss oversampling? Do you think Democratic turnout will be as high as it was in 2008?
I was using Bradley effect as shorthand for “people are lying to the pollsters”. Except in this case most of them are simply ignoring the pollsters.
i’m intrigued by the legit possibility of Romney taking NV, CO, IA, VA, FL from the list of the RCP toss-ups, making it 269-269 (Romney-Biden). However, I think Romney taking IA & NV would most likely mean he also got OH… and Romney has binders full of Detroit bankruptcy TRUTH to break through at least the OH portion of your firewall.
gahrie, where’d you get that 91% hangup figure? that’s the first stat i’ve ever come across on polling hangups.
Is this really happening? Mitt Romney is solidly getting the vote of everyone who says “we don’t know anything about the current president”? That’s ironic. Has Romney released his school records? No. Has Romney released more than 2 years worth of tax returns? No. Has Romney released the records about his time running the Olympics? No, they were destroyed immediately after the event ended. Bain Capital’s paperwork filed with the SEC lists him as an officer of the company years after he says he retired from the compnay and had nothing to do with the decision making there.
Seems like the Tea Party and the people that have a problem with transparency, don’t have a problem with those things if it’s their candidate hiding information, because now would be the time to demand transparency and information.
Sandy – I don’t know any easy way to break this to you … but …
Mitt Romney is getting more and more of his votes from people who *do* know about Pres’ent Obama …
I hadn’t realised that Sandy was so well-known … and yet Brendan is choosing to ignore him …
“It’s unclear now how threatening Sandy will be come Friday, because its path depends on several weather factors, but there is potential for a “tropical nightmare,” AccuWeather.com reports.
From here …