For those who live and die by the polls at the tail end of the election cycle, the past 24 hours, and the next 24 hours, are going to be nothing short of excruciating.
That is because, for all practical matters, it will be at least that long until we get any hint as to whether or not President Obama's well-received second debate performance has managed to alter the trajectory of the presidential election. Despite the fact that we are greeted with 35 new bits of data today, all of them (except for the House of Ras, of course) are samples that were either partly or wholly conducted prior to the showdown on Tuesday.
Therefore, all we have now is far too much reading into single-day samples in tracking polls, and hints dropped in tweets like this rather pessimistic one from PPP. Of course, the margin of error normally inherent in single-day samples makes overanalysis of them very, very dangerous.
More on that in a little bit. For now, though, on to the numbers:
PRESIDENTIAL POLLING:
NATIONAL (Gallup Tracking): Romney 52, Obama 45 (LV); Romney 48, Obama 47 (RV)
NATIONAL (IBD/TIPP Tracking): Obama 46, Romney 46
NATIONAL (Ipsos/Reuters Tracking): Obama 47, Romney 44 (LV); Obama 47, Romney 39 (RV)
NATIONAL (PPP Tracking): Obama 48, Romney 48
NATIONAL (Rasmussen Tracking): Romney 49, Obama 47
COLORADO (PPP): Obama 50, Romney 47
CONNECTICUT (PPP for League of Conservation Voters): Obama 53, Romney 44
CONNECTICUT (Univ. of Connecticut): Obama 51, Romney 37
IOWA (NBC News/Marist): Obama 51, Romney 43 (LV); Obama 50, Romney 43 (RV)
MICHIGAN (Denno Research): Obama 44, Romney 41
MICHIGAN (EPIC-MRA): Obama 52, Romney 46
MINNESOTA (SurveyUSA): Obama 50, Romney 40
NORTH CAROLINA (Rasmussen): Romney 52, Obama 46
OHIO (Rasmussen): Obama 49, Romney 48
PENNSYLVANIA (Susquehanna Research for the PA Republican Party): Romney 49, Obama 45
VIRGINIA (PPP for the League of Conservation Voters): Obama 49, Romney 48
WASHINGTON (Univ. of Washington): Obama 52, Romney 41
WISCONSIN (NBC News/Marist): Obama 51, Romney 45 (LV); Obama 50, Romney 44 (RV)
DOWNBALLOT POLLING:
CT-SEN (PPP for the League of Conservation Voters): Chris Murphy (D) 48, Linda McMahon (R) 44
CT-SEN (Univ. of Connecticut): Chris Murphy (D) 44, Linda McMahon (R) 38
NE-SEN (Hickman Analytics for the Kerrey campaign): Deb Fischer (R) 50, Bob Kerrey (D) 45
NV-SEN (Mellman Group for the Berkley campaign): Shelley Berkley (D) 41, Sen. Dean Heller (R) 38, Others 5
NV-SEN (Rasmussen): Heller 50, Berkley 43, Others 4
NV-SEN (SurveyUSA): Heller 46, Berkley 40, Others 8
VA-SEN (PPP for the League of Conservation Voters): Tim Kaine (D) 50, George Allen (R) 45
WA-SEN (Univ. of Washington): Sen. Maria Cantwell (D) 58, Michael Baumgartner (R) 35
WI-SEN (NBC News/Marist): Tammy Baldwin (D) 49, Tommy Thompson (R) 45
NH-GOV (Rasmussen): Ovide Lamontagne (R) 48, Maggie Hassan (D) 46
WA-GOV (Univ. of Washington): Jay Inslee (D) 48, Rob McKenna (R) 45
CA-44 (FM3 for the Hahn campaign): Rep. Janice Hahn (D) 43, Rep. Laura Richardson (D) 27
FL-09 (Gravis Marketing for the Grayson campaign): Alan Grayson (D) 56, Todd Long (R) 41
FL-18 (PPP for local media): Rep. Allen West (R) 51, Patrick Murphy (D) 42
FL-22 (Voter Survey Services--R): Lois Frankel (D) 47, Adam Hasner (R) 47
NY-01 (McLaughlin and Associates for the Altschuler campaign): Randy Altschuler (R) 48, Rep. Tim Bishop (D) 43
PA-04 (York College): Scott Perry (R) 39, Harry Perkinson (D) 22, Others 8
A few thoughts, as always, await you just past the jump ...
Given the clear (and, for Democrats, troubling) tightening of the presidential race, Democrats have been fervently hoping that President Obama's considerably better debate performance at Hofstra would create a post-debate momentum somewhat akin to the one Mitt Romney enjoyed post-Denver.
Can that happen? It can. Will it happen? As maddening as this is for regular readers of the Wrap (polling junkies in good standing, without a doubt), it is simply too early to know.
What little early signs we have are, at best, mixed.
There were some positive signs. In the Ipsos-Reuters daily tracking poll, Barack Obama moved out to his biggest lead among registered voters (O+8) since before Denver, and their likely voter screen held steady at a three-point presidential lead. The RAND political tracker (which I do not include in the Wrap, because it is not a traditional poll asking for how voters will actually cast a ballot) showed sharp movement in the president's favor, moving him out to a 50-44 edge over all. Meanwhile, PPP launched a new tracking poll (sponsored by Americans United for Change), to be conducted daily now through the election. The initial incarnation of that tracker had the presidential race tied, which was marginally good news for Obama fans on two different scores. For one thing, in an apples-to-apples comparison, a tie is better than the four-point edge Mitt Romney enjoyed in the PPP poll conducted on behalf of Daily Kos and SEIU last weekend. For another, PPP's Tom Jensen did tweet that Wednesday was the best day of the three for the president.
Other small signals were far less charitable. The House of Ras issued two post-debate battleground state polls. In one (Ohio), the margin was exactly what it had been pre-debate: Obama +1. In the other (North Carolina), the House actually had Mitt Romney doing better than he had done pre-debate. Rasmussen also had a very sudden reversal of fortune in their "swing state" sample, as a three-point Obama lead on Wednesday disappeared completely. Another tracking poll (IBD-TIPP) also moved a couple of points in Mitt Romney's favor, though that is a seven-day tracking poll and any post-debate movement would comprise only a sliver of the overall number of respondents.
Then there were the truly mixed signals. NBC/Marist's battleground polls in Iowa and Wisconsin were simply fantastic data points for Obama, but the write-up of the polls made clear that the numbers barely budged pre-and-post-debate. Gallup's numbers were even more perplexing. Among registered voters, Obama did one point better than the previous day (not abnormal, given that they, too, are a seven-day tracking poll). But among likely voters, Obama actually was a point worse than he had been on Wednesday, giving Mitt Romney a pretty comical seven-point edge.
That poll was the most talked-about data point of the day, and not necessarily in a good way. Nate Silver, for one, is a skeptic:
...its results are deeply inconsistent with the results that other polling firms are showing in the presidential race, and the Gallup poll has a history of performing very poorly when that is the case.
The most recent iteration of Gallup's LV screen may be the best evidence for its errancy when it deviates from the herd. In 2010, they claimed a generic ballot edge of 15 points for the Republicans, a finding that was a total of nine points off of the final result. Of course, in Gallup's defense, a generic ballot, with 435 different variables, is a little tougher to hit on the head than a presidential election. That said, Singiser's first rule of poll analysis applies here: if you have something different from every other pollster, the odds are pretty good that everyone else is not the one that is wrong.
All that having been said, these things do take time, and thus it is a mistake to assume that today's conflicting data means that the current state of play has been cemented until November 6th. We tend to forget, for example, that Mitt Romney also lost a point in the Gallup tracker in the day after the Denver debate. Democrats took that as a sign to be relieved, but their relief, of course, was very short-lived.
By the time the third debate convenes Monday night, we will absolutely have a clearer picture as to whether or not President Obama has reset the race in any convincing fashion, or whether Romney successfully dodged electoral damage after his debate defeat.
Then, of course, we will be left with another waiting game, as we assess what impact the third and final debate, if any, has on the race for the White House.
The good news? We are now less than three weeks away from having to go crazy reading conflicting and contradictory tea leaves.