For all intents and purposes, the post-convention period is now in the books. Even the slow-moving (by design, courtesy of that seven-day window) Gallup poll now has more than two-thirds of its data emanating from days after the end of the convention. Every other pollster's data is wholly after the convention. Ergo, we now have a conclusive picture of the impact of the convention.
But what remains in front of us is arguably the biggest question: for how long can Obama sustain the unmistakeable bump in support he received from the DNC? If he can stretch out that advantage until the debates, and perform respectably in those, he will be in an excellent position for November. If this bump proves ephemeral, however, that could point to underlying structural problems that could make November very uncomfortable for the Democrats.
On to the numbers:
PRESIDENTIAL GENERAL ELECTION TRIAL HEATS:
NATIONAL (Fox News): Obama d. Romney (48-43 LV; 46-42 RV)
NATIONAL (Gallup Tracking): Obama d. Romney (50-43)
NATIONAL (Ipsos-Reuters): Obama d. Romney (48-45 LV; 47-43 RV)
NATIONAL (Rasmussen Tracking): Obama d. Romney (46-45)
NATIONAL (YouGov): Obama d. Romney (49-45)
CALIFORNIA (SurveyUSA): Obama d. Romney (57-35)
MICHIGAN (EPIC-MRA): Obama d. Romney (47-37)
MONTANA (PPP): Romney d. Obama (46-43-7 w/Gary Johnson; 50-45 w/o Johnson)
NEW MEXICO (PPP): Obama d. Romney (53-42)
TEXAS (WPA Opinion Research--R): Romney d. Obama (55-40)
WASHINGTON (PPP for the League of Conservation Voters): Obama d. Romney (53-42)
DOWNBALLOT POLLING:
CA-SEN (SurveyUSA): Sen. Dianne Feinstein (D) 55, Elizabeth Emken (R) 37
CA-31 (Internal poll for the Dutton campaign): Bob Dutton (R) 22, Rep. Gary Miller (R) 17
(2014) FL-GOV--D (SEA Polling): Alex Sink 31, Charlie Crist 29
MA-06 (MassINC): Rep. John Tierney (D) 46, Richard Tisei (R) 34, Daniel Fishman (L) 7
MI-SEN (EPIC-MRA): Sen. Debbie Stabenow (D) 49, Pete Hoekstra (R) 38
(2014) MN-SEN (PPP): Sen. Al Franken (D) 50, Tim Pawlenty (R) 43; Franken 50, Norm Coleman (R) 43; Franken 52, Michele Bachmann (R) 40
MN-SEN (SurveyUSA): Sen. Amy Klobuchar (D) 55, Kurt Bills (R) 34
MN-SEN (Wenzel Strategies for the Bills campaign): Sen. Amy Klobuchar (D) 52, Kurt Bills (R) 38
MT-AL (PPP): Steve Daines (R) 40, Kim Gillan (D) 37, Dave Kaiser (L) 9
MT-SEN (PPP): Sen. Jon Tester (D) 45, Denny Rehberg (R) 43, Dan Cox (L) 8
NM-SEN (PPP): Martin Heinrich (D) 50, Heather Wilson (R) 41
RI-01 (DCCC IVR--D): Rep. David Cicilline (D) 49, Brendan Doherty (R) 43
WA-GOV (PPP for the League of Conservation Voters): Jay Inslee (D) 48, Rob McKenna (R) 42
WA-GOV (SurveyUSA): Jay Inslee (D) 49, Rob McKenna (R) 44
WA-SEN (SurveyUSA): Sen. Maria Cantwell (D) 54, Michael Baumgartner (R) 38
WI-01 (FM3 for the Zerban Campaign--Informed trial heat): Rep. Paul Ryan (R) 47, Rob Zerban (D) 39, Keith Deschler (L) 4
WI-01 (Public Opinion Strategies for the Ryan campaign): Rep. Paul Ryan (R) 58, Rob Zerban (D) 33
A few thoughts, as always, await you just past the jump...
As both conventions are now securely in the rearview, and even discussion of them has faded from the public conversation, now would be an excellent time to lay out the "final" assessment of the respective bounces.
The simplest way to do it is to just pick an arbitrary one-week window of polling, where the start point was in the days prior to the debates conventions, and the endpoint is in the days prior following to the debates conventions.
Doing this, oddly, does artificially reduce the "average bounce" for the president, because the House of Ras has inexplicably shown the president's support sag by four points in two days (even as Romney's numbers have not moved). But, still, in the name of consistency, here are the numbers.
For CNN, the weekly spread is Monday-to-Monday (the day their polls are typically released). For all the other polls, we will go from Wednesday-to-Wednesday. I debated going from Tuesday, but since few of the respondents on Tuesday evenings would have seen night one of the convention, I made the judgment call of going with Wednesday poll releases. Your mileage may vary.
Mitt Romney's Republican Convention bounce, from 8/29-9/5
CNN RV: From Obama +9 to Obama +7 (Romney +2)
CNN LV: From Obama +2 to Obama +0 (Romney +2)
Gallup Tracking: From Obama +1 to Obama +1 (Romney +0)
Ipsos/Reuters RV: From Obama +2 to Obama +0 (Romney +2)
Ipsos/Reuters LV: From Romney +0 to Romney +2 (Romney +2)
Rasmussen: From Obama +1 to Romney +3 (Romney +4)
YouGov: From Romney +1 to Obama +1 (Obama +2)
NET BOUNCE (7 polls): Romney +1.4 percent
NET MARGIN POST-CONVENTION (7 polls): Obama +0.6 percent
Barack Obama's Democratic Convention bounce, from 9/5-9/12
CNN RV: From Obama +7 to Obama +8 (Obama +1)
CNN LV: From Obama +0 to Obama +6 (Obama +6)
Gallup Tracking: From Obama +1 to Obama +7 (Obama +6)
Ipsos/Reuters RV: From Obama +0 to Obama +6 (Obama +6)+
Ipsos/Reuters LV: From Romney +2 to Obama +3 (Obama +5)+
Rasmussen: From Romney +3 to Obama +1 (Obama +4)
YouGov: From Obama +1 to Obama +4 (Obama +3)
NET BOUNCE (7 polls): Obama +4.4 percent
NET MARGIN POST-CONVENTION (7 polls): Obama +5.0 percent
(+)--Tuesday's data, because election tracking poll was supplanted today by a traditional, telephone-based survey
So, in the final analysis, Obama's bounce more than tripled that of Mitt Romney and would have come reasonably close to quadrupling it, had Rasmussen not scaled back the Obama lead over the past couple of days.
In an additional set of evidence, several recent polls hint that Obama had a much more substantial convention bounce than did Romney. The most recent evidence for that came from Fox News, of all places, where today's poll showed that the net effect of the conventions (FNC's last poll was before the GOP convention gathered in Tampa) was a movement of several points in the direction of Barack Obama. This included a fairly substantial shift of six points among likely voters, and a more incremental movement (two points) among RVs. Similar moves were seen in both the ABC/WaPo poll, and our own Daily Kos/SEIU State of the Nation poll, where the shift to likely voters from registered voters did not move the needle at all, despite the tendency of LV samples to favor the GOP.
The question, of course, is: now what? Today's events, and Mitt Romney's painfully amateurish response (rarely do I say a picture is worth a thousand words, but this one ought to be worth a few percentage points), will be a fascinating case study in public opinion. Logic would dictate that if the incumbent convention bounce was ephemeral (and they aren't always a temporary blip), we'd know this week. But Romney's unmitigated disaster of a Wednesday may change that calculus. The question is whether he merely suffered a "flesh wound" or if today, as some Twitter follows on the left and right have suggested (including, in my favorite bit of political commentary, pro poker star Daniel Negreanu), may have been the day that the GOP lost any hope of reclaiming the White House.
In other polling news...
- I have a minor correction to make on one of the subjects of last night's Wrap commentary. As it happens (and thanks to technopundit for the heads-up via Twitter), the crew at ABC/WaPo did have a previous likely voter sample. And that sample reflected a smaller bounce, from Romney +2 to Obama +1. However, there are two issues here. For one thing, the new ABC/Washington Post polling memo only listed the RV results in all of their previous polls, which gave the illusion that Balz and Cohen had made an apples-to-oranges comparison. For another, the writers do share a small amount of the blame. While describing the race as little changed, they did completely neglect to point out the far more substantial shift among registered voters (where the movement was seven, rather than three, points).
- There is a ton of downballot polling today, far too much to comment on every tidbit (note to readers: get used to it). I have to point out, though, in the name of accuracy, that those two polls out of Paul Ryan's district in southeastern Wisconsin are not created equal. The Democratic poll is a so-called "informed ballot test." In short, respondents are read a short description of both candidates, and then the results are taken. I have three problems here. One, if they took a trial heat before they did the informed trial heat, they didn't say so. That's a problem: it wouldn't take much to impeach the poll on the grounds that they buried the trial heat numbers because they were lousy. Second, the inherent problem with informed trial heat numbers is ensuring that the descriptions are truly on the level. The article citing the poll data did not give the descriptions, for what it is worth. Third, and this is a matter of personal taste, I think the time window has essentially lapsed for "informed trial heat" surveys. We are weeks away from Election Day. In Zerban's defense, he probably has the resources to run a campaign to define himself. But, of course, Paul Ryan has a ton of resources to ill-define Zerban. At the end of the day, it is probably a little late in the game to say "when we're better known, these numbers will improve". Feel free to debate in the comments.
- Speaking of a little late in the game, I just gotta say to the pollster for challenging Republican Bob Dutton (who is in a GOP-on-GOP battle in CA-31): dude, we are less than 60 days out. Now is a fine time to push your leaners a little bit. 22-to-17?! Really?!