Over the weekend, the question on the minds of the community of political junkies was a simple one: had Mitt Romney's clear victory in the first debate, in the eyes of a majority of voters, actually translated into votes? In other words, could the debate have enough of an impact to recenter the race for the White House?
We only have a preliminary set of data on that point, but the answer appears to be "sort of, but it does not seem likely." Perhaps predictably, the more conservative the pollster, the more likely they were to see Romney make "game-changing" gains. Other outlets, however, were considerably more restrained in their assessments. And one pollster completely recalibrated, changing a years' worth of tilt in precisely the opposite direction in a single poll. That one, as you might expect, is the one sucking up the disproportionate amount of oxygen in the public conversation today.
More on that later. For now, though, on to the numbers:
PRESIDENTIAL POLLING:
NATIONAL (Gallup Tracking): Obama 49, Romney 46
NATIONAL (GWU/Battleground for Politico): Obama 49, Romney 48
NATIONAL (Ipsos/Reuters Tracking): Obama 47, Romney 45 (LV); Obama 46, Romney 41 (RV)
NATIONAL (Pew): Romney 49, Obama 45 (LV); Obama 46, Romney 46 (RV)
NATIONAL (Rasmussen Tracking): Obama 48, Romney 48
COLORADO (Gravis Marketing--R): Romney 49, Obama 46
COLORADO (Rasmussen): Obama 49, Romney 48
COLORADO (Selzer/Univ. of Denver): Obama 47, Romney 43, Others 4
IOWA (Rasmussen): Obama 49, Romney 47
MASSACHUSETTS (Western New England College): Obama 63, Romney 33
MICHIGAN (EPIC-MRA): Obama 48, Romney 45
MICHIGAN (Foster McCollum White/Baydoun): Obama 49, Romney 46
NORTH DAKOTA (Mason Dixon): Romney 54, Obama 40
OHIO (Wenzel Strategies for Citizens United--R): Romney 48, Obama 47
PENNSYLVANIA (Susquehanna Research--R): Obama 47, Romney 45, Johnson 3
VIRGINIA (PPP): Obama 50, Romney 47
WISCONSIN (PPP): Obama 49, Romney 47
DOWNBALLOT POLLING:
FL-SEN (Rasmussen): Sen. Bill Nelson (D) 51, Connie Mack IV (R) 40
FL-SEN (We Ask America--R): Sen. Bill Nelson (D) 44, Connie Mack IV (R) 44
MA-SEN (Harstad Research for the Warren campaign): Elizabeth Warren (D) 50, Sen. Scott Brown (R) 44
MA-SEN (Western New England College): Elizabeth Warren (D) 50, Sen. Scott Brown (R) 45 (LV); Elizabeth Warren (D) 50, Sen. Scott Brown (R) 44 (RV)
MI-SEN (EPIC-MRA): Sen. Debbie Stabenow (D) 55, Pete Hoesktra (R) 35
MI-SEN (Foster McCollum White/Baydoun): Sen. Debbie Stabenow (D) 51, Pete Hoekstra (R) 43
NV-SEN (Precision Opinion): Sen. Dean Heller (R) 45, Shelley Berkley (D) 43
ND-SEN (Mason Dixon): Heidi Heitkamp (D) 47, Rick Berg (R) 47
OH-SEN (Rasmussen): Sen. Sherrod Brown (D) 46, Josh Mandel (R) 46, Others 2
OH-SEN (We Ask America--R): Sen. Sherrod Brown (D) 46, Josh Mandel (R) 44
OH-SEN (Wenzel Strategies for Citizens United): Josh Mandel (R) 46, Sen. Sherrod Brown (D) 42
VA-SEN (PPP): Tim Kaine (D) 51, George Allen (R) 44
WA-SEN (SurveyUSA): Sen. Maria Cantwell (D) 53, Michael Baumgartner (R) 40
WI-SEN (PPP): Tammy Baldwin (D) 49, Tommy Thompson (R) 46
AZ-01 (NRCC IVR--R): Jonathan Paton (R) 50, Ann Kirkpatrick (D) 45
CA-24 (Public Opinion Strategies for the Maldonado campaign): Abel Maldonado (R) 45, Rep. Lois Capps (D) 44
FL-02 (St. Pete Polls--R): Al Lawson (D) 47, Rep. Steve Southerland (R) 46
FL-09 (St. Pete Polls--R): Alan Grayson (D) 45, Todd Long (R) 42
FL-10 (St. Pete Polls--R): Rep. Dan Webster (R) 51, Val Demings (D) 40
FL-13 (St. Pete Polls--R): Rep. Bill Young (R) 49, Jessica Ehrlich (D) 40
FL-16 (St. Pete Polls--R): Rep. Vern Buchanan (R) 55, Keith Fitzgerald (D) 38
FL-18 (St. Pete Polls--R): Rep. Allen West (R) 53, Patrick Murphy (D) 41
FL-22 (St. Pete Polls--R): Adam Hasner (R) 45, Lois Frankel (D) 44
FL-26 (St. Pete Polls--R): Rep. David Rivera (R) 43, Joe Garcia (D) 41, Others 4
IL-08 (DCCC IVR--D): Tammy Duckworth (D) 52, Rep. Joe Walsh (R) 42
IL-17 (Anzalone-Liszt for the DCCC): Cheri Bustos (D) 45, Rep. Bobby Schilling (R) 44
NV-03 (Precision Opinion): Rep. Joe Heck (R) 49, John Oceguera (D) 38
NV-04 (Precision Opinion): Danny Tarkanian (R) 44, Steven Horsford (D) 36
NY-27 (Siena College): Rep. Kathy Hochul (D) 47, Chris Collins (R) 47
ND-AL (Mason Dixon): Kevin Cramer (R) 49, Pam Gulleson (D) 37
WI-07 (NMB Research for the Duffy campaign): Rep. Sean Duffy (R) 51, Pat Kreitlow (D) 40
A few thoughts, as always, await you just past the jump...
Anyone looking for absolute clarity in the state of play in the presidential election, based on the available polling data to-date since the first presidential debate, ought to just go ahead and take a week's vacation.
Either Mitt Romney has become the betting favorite for election (fixed!), or his victorious debate performance did absolutely jack shit to move the needle in terms of presidential support, depending on the pollster in whom you decide to place your trust.
The daily tracking polls have been particularly instructive, given that we can track the movement on a day-to-day basis. Their data, as well as some tweets and commentary from PPP, begin to paint a reasonably coherent narrative. Here it is: in the presidential head-to-heads, Obama simply had a dreadful Thursday and Friday. But, and this is critical, he recovered pretty nicely on Saturday and, in particular, on Sunday.
Take Gallup as the touchstone here. Much was made this morning of a "new poll" showing Gallup putting both Obama and Romney tied at 47. The problem? It was not a separate poll, and too many media outlets crapped the bed as reporting it as such. All the Gallup article (which you can read here) stated was that in the three days of their seven-day tracking sample that followed the debate, Obama and Romney were running even. The overall tracking poll remained at Obama +3, however, because Obama averaged a five-point advantage pre-debate.
So, what happened today? The Obama lead jumped out to a five-point (50-45) lead. That tells us, quite simply, that Obama must've had a monster of a day on Sunday's single-day sample for Gallup. Given the fact that Rasmussen's three-day sample also crept two points in Obama's direction, it appeared as if Sunday was a very good day for the president.
Then, along comes Pew with the best numbers for Mitt Romney in, literally, months. Their four-point Romney lead has been pilloried by many Democrats for having a sample composition (not just on the oft-cited party ID question, but on age and ethnicity, as well) that was dramatically different from every poll they've conducted in recent months.
But, to acquit Pew a little bit (and to save them from a bizarro round of Democratic "unskewing"), that may not be a poor read on where the electorate was immediately after the debate. Republicans, finally seeing a path to election, were over the moon with their enthusiasm, while Democrats had to be largely coaxed out from under their beds. It was, if the daily tracking polls are an indicator, an incredibly transient sentiment. But that does not mean, in the final analysis, that a poll conducted in the 3-4 days immediately after the debate wouldn't show exactly what Pew showed. It also does not mean that Obama might not be back up by a solid margin with the same pollster in a week or two. Bounces are just that.
Further evidence of that, of course, will be forthcoming. As Markos tweeted earlier today (and Slate's Dave Weigel immediately converted into a very funny tweet of his own), this week's edition of the Daily Kos/SEIU State of the Nation Poll (conducted by PPP) will show a Romney lead. Given that the overwhelming bulk of that poll was conducted in the three days after the debate, this really should come as a shock to no one. As PPP's Tom Jensen noted over the weekend, Obama's Saturday numbers looked similar to pre-debate numbers. This would imply that the debate boost for Romney was an ephemeral one.
But, given the conflicting data, there is plenty in the numbers for partisans of both sides to alternately exult and weep, as they flit from individual data point to individual data point.
And, as a fitting coda to a complete perplexing day, Gallup is apparently going to debut their likely voter screen tomorrow, and that shift from registered voters to LVs will move the race to a dead heat. Of course, Gallup's likely voter screen has a pretty chequered past. It crapped the bed a little bit in 2008, when it overestimated Barack Obama's margin of victory by four points. But then the likely voter screen for Gallup really took a dump in 2010, when their RV/LV gap on the generic ballot was a cartoonish eleven points. Among registered voters, Gallup saw an R+4 electorate. Among "likely voters", Gallup saw an R+15 electorate.
The final outcome? R+6. So...yeah, let's say that the assumptions made in likely voter screens can be, shall we say, a little off.
In other polling news...
- Florida gets a ton of polling love today. Aside from the Senate polls (more on those in the next bullet point), the odd firm known as St. Pete Polls is back in the saddle, polling every competitive House seat in the Sunshine State. And whether their numbers favor Republicans or Democrats (with the former slightly more common than the latter), their numbers just seem a little off. For example, if FL-22 was really deadlocked, then why have neither the DCCC nor the NRCC chose to play very heavily in that district. Conversely, why would there be so much attention in FL-18 if West was really putting Patrick Murphy away? Think both of those numbers are way off. Still, we post everyone, so grab the polls you see here, and add salt to your preferred taste. I'd put minimal stock into this firm for a couple of reasons. One, I am not an adherent to marketing/strategy firms that have polling wings. Second, when your numbers run counter to everyone else, as I have often said, it is pretty unlikely that everyone else is wrong, and you are right.
- At the Senate level, the two GOP-leaning robopollsters generally agree on one battleground Senate poll, but come far apart on the other. Both agree on a close Ohio Senate race between incumbent Democrat Sherrod Brown and Republican Josh Mandel (and a right-wing poll by Wenzel Strategies goes a step further, and actually claims a lead for the challenger), but We Ask America has the Senate race tied in Florida. No one else has had that, and even the House of Rasmussen left W.A.A. on an island, giving incumbent Democrat Bill Nelson an 11-point edge (52-41). Frankly, I see that as far more likely, as it tracks well with other polls. I also must admit to some newfound trepidation about W.A.A. The reason? I am becoming steadily more convinced that they sit on data that is not amenable to right-wing narratives. Last week, they informed us that a pre-debate Ohio poll would not be released, because the entity funding the poll did not give them permission to release it. Presumably, that is because it was bad news for the GOP. Today, they released the Ohio and Florida Senate polls, but quite notably did not release Senate polling from Virginia. This is despite the fact that they polled all three states last week. What gives? Could it be that W.A.A., like Rasmussen and PPP, found Republican George Allen now trailing Democrat Tim Kaine by a solid margin? Their reticence on the state of play in that Virginia Senate race, while eagerly releasing polling in the two other states that were more positive for the GOP narrative of a plausible Senate majority, is worth noting.