For nearly a week now, I have said, repeatedly, that gloating over the absence of a polling "bounce" for the Romney-Ryan ticket from their Republican National Convention was premature, because too much of the polling sample was still culled prior to the major addresses and events of said convention.
Well, as we sit here on Labor Day evening, that is no longer the case. Even the seven-day polling sample conducted by the folks at Gallup has reached the point where roughly three-quarters of the respondents were queried after the convention began. Plus, the kind folks at Gallup have dropped some hints about the day-to-day characteristics of their numbers. Furthermore, they polled people over the weekend on what they thought of the Romney convention, and they gave it uncharacteristically low marks.
And the verdict? This may be among the weakest convention bounces of the modern era, and far from what RNC Chair Reince Priebus was anticipating when he flatly declared before the gathering in Tampa that the bounce would be real and visible.
On to the numbers:
PRESIDENTIAL GENERAL ELECTION TRIAL HEATS:
NATIONAL (Gallup Tracking): Obama d. Romney (47-46)
NATIONAL (Rasmussen Tracking): Romney d. Obama (48-44)
COLORADO (PPP): Obama d. Romney (49-46)
FLORIDA (PPP): Obama d. Romney (48-47)
NORTH CAROLINA (Elon University for the Charlotte Observer): Romney d. Obama (47-43)
NORTH CAROLINA (PPP): Obama tied with Romney (48-48)
NORTH CAROLINA (SurveyUSA/High Point University): Romney d. Obama (46-43)
DOWNBALLOT POLLING:
NC-GOV (Elon University for the Charlotte Observer): Pat McCrory (R) 52, Walter Dalton (D) 37
NC-GOV (PPP): Pat McCrory (R) 45, Walter Dalton (D) 39, Barbara Howe (L) 5
A few thoughts, as always, await you just past the jump...
Perhaps it was irrational exuberance, and perhaps it was simply the mother of all brain farts. But GOP mouthpieces spent much of the week prior to the GOP confab in Tampa goosing expectations about the impact the convention would have on the state of play in the election.
Three weeks before the election, in a briefing with reporters, Team Romney put a fairly audacious number out there for the press to digest:
Displaying a power point slide, the Romney adviser showed two sets of historical polling dating back to 1976, and explained:
"The incumbent averaged a minus-four on the ballot going into the convention and came out plus three. The challenger, because the challenger is less well known and not as well defined, came in at minus-four and came out at plus seven. So they picked up about 11 points. So the challenger picks up more points than the incumbent does, which makes sense."
Why team Romney felt comfortable putting (with caveats, granted) an 11-point number out there, I will never know. Even if the GOP convention had been a quality gathering, recent bounces have been mostly
muted.
Romney's problem, as it happens, is twofold. Not only did his campaign and his party leadership put up some pretty unreasonable expectations about what the post-convention landscape would look like, the convention, as it happens, was apparently not well received.
This bit of data, perhaps more than any other we have seen to date, may well explain why the bounce out of this convention essentially does not exist. Nearly as many people were
less inclined to vote GOP than were
more inclined to do so, when all was said and done.
GOP conventions, of late, have been pretty lousy (the three worst conventions, by this metric, have been the 2012, 2004, and 2008 GOP conventions, respectively). But this was the worst of them all, and it was the only one in which the Republican were the party out of the White House.
Typically, the challenger scores a knockout with their convention. Look at the recent history of convention support for the party out of the White House:
Support for Partisan Conventions, as measured from the difference between those saying the convention made them MORE likely to support the party and those saying the convention made them LESS likely to support the party
1984 Democrats: +16 (45/29)
1988 Democrats: +35 (56/21)
1992 Democrats: +45 (60/15)
1996 Republicans: +15 (44/29)
2000 Republicans: +17 (44/27)
2004 Democrats: +14 (44/30)
2008 Democrats: +14 (43/29)
Clearly, conventions have weakened over time as a way for voters to reassess parties and candidates. Therefore, one could assume a more muted bounce, as a result. Political scientist Thomas Holbrook
predicted very modest bounces in a piece for Pollster a couple of weeks back.
Worth noting, however: the average bounce (by averaging together the Gallup, Ipsos/Reuters, and Rasmussen polls) for the GOP stands at an incredibly modest 2.0 percentage points. That failed to even meet Holbrook's very low bar (3.6 percent) for the Romney convention.
Holbrook was even more bearish on the Democratic convention, saying Obama would get a bounce of just 1.1 percent. Tomorrow, we can start the clock on measuring the Democratic bounce.
In other polling news...
- Looking where we really should (state-by-state polling), is there a better case to be made for a Romney convention bounce? According to five statewide polls today, the answer is...well...no, not really. PPP polled three states. They found that Romney's fortunes improved between 0-3 points depending on the state, when compared to the last PPP poll in the state. Meanwhile, Elon University and SurveyUSA had Romney leading by 3-4, which, at best, is a modest improvement over where the critical mass of polls in the Tar Heel State were prior to the conventions.
- Head-scratcher of the day: the state of play in the North Carolina gubernatorial election. Elon and PPP had slightly different numbers in the presidential race (a four-point spread), which could be almost entirely attributed to the fact that Elon's sample was considerably more white than PPP's sample. However, the two pollsters were nine points apart on the gubernatorial race. Could it simply be the fact that PPP asked about Libertarian Barbara Howe, and Elon apparently did not? That seems to be too simple an explanation, but nothing else jumps off of the page.