Perhaps I, along with the universe of political junkies, should have seen this one coming. Less than 36 hours after the first presidential debate, the two most well-known right-leaning "robopollsters" in the game are out with swing state polls, culled from single-day samples. And their results, perhaps not surprisingly, have been exceptionally kind to the Republican nominee.
What is interesting is that they take pretty different approaches to their polls, even though the toplines are somewhat similar.
What is even more interesting is how badly one of those pollsters might have pissed away their credibility in their polling memo. You'll have to read it for yourself, it really is that bad.
For now, though, on to the numbers:
NATIONAL (Gallup Tracking): Obama 50, Romney 45DOWNBALLOT POLLING:
NATIONAL (Ipsos/Reuters Tracking): Obama 46, Romney 44 (LV)
NATIONAL (Rasmussen Tracking): Obama 49, Romney 47
FLORIDA (Rasmussen): Romney 49, Obama 47
FLORIDA (We Ask America--R): Romney 49, Obama 46, Others 1
MONTANA (The Mellman Group for JET PAC [DGA]--D): Romney 48, Obama 44
NEVADA (Gravis--R): Obama 49, Romney 48
NEW MEXICO (PPP): Obama 52, Romney 43
OHIO (Rasmussen): Obama 50, Romney 49
OHIO (We Ask America--R): Romney 47, Obama 46, Others 1
VIRGINIA (Rasmussen): Romney 49, Obama 48
VIRGINIA (We Ask America--R): Romney 48, Obama 45, Others 2
NM-SEN (PPP): Martin Heinrich (D) 51, Heather Wilson (R) 41A few thoughts, as always, await you just past the jump ...
NV-SEN (Gravis--R): Sen. Dean Heller (R) 52, Shelley Berkley (D) 36
OH-SEN (Public Opinion Strategies for the Mandel campaign): Sen. Sherrod Brown (D) 47, Josh Mandel (R) 44
VA-SEN (Rasmussen): Tim Kaine (D) 52, George Allen (R) 45
FL-02 (Lester and Associates for the DCCC): Rep. Steve Southerland (R) 43, Al Lawson (D) 43
HI-01 (Merriman River Group for Civil Beat): Rep. Colleen Hanabusa (D) 49, Charles Djou (R) 44
HI-02 (Merriman River Group for Civil Beat): Tulsi Gabbard (D) 70, Kawika Crowley (R) 18
IL-12 (We Ask America--R): Jason Plummer (R) 44, Bill Enyart (D) 40
MT-AL (The Mellman Group for JET PAC [DGA]--D): Steve Daines (R) 36, Kim Gillan (D) 34
I have long had some serious internal conflicts about the polling outfit known as "We Ask America." They are a research arm of the Illinois Manufacturing Association, and as such, one has to assume a thumb (and perhaps a forefinger) are going to be on the scales at all times. However, their track record in their home state of Illinois in 2010 was actually pretty decent. Of course, 2010 was a good year to be a pollster that favors Republican outcomes, so any benefit of the doubt they might get from being on the fairway in that particular election cycle is mitigated a bit.
This year, on balance, their numbers have been within the bounds of reason. Some of their House polling has been very favorable to Republicans (including today's relevant example in IL-12), but they have also thrown some numbers bound to make Republicans howl in pain (the Wisconsin numbers from last week come immediately to mind).
Today's polling, however, may jeopardize the ability to seriously consider them as we head towards Election Day. And, hard as it may be to believe, my concern is not with their actual topline numbers. Sure, they seem a little bit too favorable for Mitt Romney, but a single day sample after a universally lauded debate performance should yield the best numbers he has seen in a long time. Some could easily quibble about the sudden pluralities of Republicans in the Florida and Virginia samples, but, hey, let's not all become Mr. Unskewed here.
My problem with the firm was the polling memo itself. It was ... well, honestly, there was no other word to describe it ...giddy. The firm is, presumably, attempting to be professional, but their memo reads like a middle school cheerleading corps chanting "Go, fight, win!" as their team heads toward the goal line.
Pollsters, especially ones that do private client work, have preferences. PPP, our polling partners, make absolutely no bones about the fact that they are a Democratic polling firm. Check their archives, though, and see if you can find a write-up of their data that reads, quite frankly, as silly as the one that came out of the W.A.A. offices this morning.
Ironically, it was every Democrat's least favorite pollster, the House of Rasmussen, which looked measured and responsible by comparison. They also caught a Romney bounce of sorts in Virginia and Florida, but took pains to point out that these were fairly modest differences from previous polling. Now, the House of Ras is still the House of Ras: They introduced a new metric in today's numbers (those who are "certain to vote") which predictably padded Mitt Romney's numbers by a few points. But the odd celebratory tone that flew off the page in W.A.A's write-up was absent.
In other polling news ...
- What are the numbers telling us post-debate? The honest answer is:Iit is pretty damned hard to tell. The daily trackers moved all over the place. Gallup crept forward a point in Obama's favor, Rasmussen stood pat at Obama +2, and Ipsos/Reuters slid another three points in Mitt Romney's favor. What does this tell us? Well, it would seem to tell us that Romney's Thursday was good, but not a sea change in public support. After all, for Romney, what came onto the tracking polls Thursday night was roughly equivalent (or perhaps, in Gallup's case, incrementally worse) than what slid off. On this question, I think I agree with The New Republic's Nate Cohn: We will know what we know by early next week.
- There is one rather scant piece of evidence of a legitimate Romney "surge," and it comes from Clarus Research, the group run by former Campaigns and Elections publisher Ron Faucheux. They claimed in a write-up today to have done two single-day samples, one on Tuesday and one on Thursday. The Tuesday sample was Obama +4. The Thursday sample is Romney +1. Of course, there is some issues inherent in single-day samples. But this is just another data point to add to the pile.
- Downballot, I'd point out three "holy crap" data points (and no, that fundamentally absurd Gravis poll is not among them). Two House polls immediately leap to mind, as we finally understand why the DCCC took a sudden interest in gunning for freshman Republican Rep. Steve Southerland. The assumption had been that the more conservative segments of northern Florida would be tough sledding for the Democrats. But Al Lawson appears to be making this a real race. Which explains why not only the DCCC got involved, but also why the NRCC suddenly jumped off the bench and began funneling resources into the region. The other one is in Hawaii. Now, I suspect that Colleen Hanabusa will be reelected, but seeing Charles Djou incrementally better than he did in 2010 (when he lost this seat to Hanabusa 53-47) is pretty shocking. But, for me, the "holy crap" is in Virginia, where even the House of Ras no longer sees this Senate race as a coin flip. If a sample that yielded a Romney +1 outcome there also sees a seven-point lead for Tim Kaine, George Allen's team has to be very, very nervous.