5:19 pm Oct. 10, 2012
"Gonna need you to talk me down from this post-debate cliff," read a text message from a normally level-headed Democratic friend.
"Talk me off the ledge," said an email from another one. "Did Obama have a stroke?"
The truth is: everything is not fine for Obama, and his re-election is not anywhere near in the bag. But, with the benefit of a week having passed since last Wednesday’s one-man debate, there are a few things we now know.
While pollsters account for demographic accuracy in their samples—when it comes to measures like race, region, gender, or age—they do not build in standardization for party identification. So when their samples include more people self-identifying with a given party, they can infer that the party and its candidate have achieved traction.
In Gallup’s poll, 34 percent of voters identified as Republicans in the days after the debate, up two points up from polls taken just prior to the debate. At the same time, 33 percent identified as Democrats in the post-debate poll, down by four points from the week before.
But, it’s worth noting, a similar phenomenon occurred, in the other direction, after the Democratic convention and Romney’s “47 percent” comments. At those times, the portion of the electorate self-identifying as Democrats increased markedly.
So where does that leave us? Not too far from where we were this summer. The head-to-head national contest between the two candidates is tight, with both sides having strong bases but also uncommitted “leaners,” liable to stick with them in good times (Romney’s debate, Obama’s convention) and wobble during tough ones (Romney’s 47 percent comments, Obama’s debate).
These vicissitudes, like the one we saw this week, don’t signal a change from the basic layout of the race, but rather a confirmation of it.
If Romney were to, say, score another decisive debate win, he could firm up those leaners. But a bad turn could mean losing them. In no case is it likely that either candidate will build a lead that dramatically exceeds their base plus leaners—meaning the popular vote will probably be close no matter what. (This was also true before the debate, when it was the Democrats who were benefiting from an event-triggered uptick in self-identification, post-convention.)
That means factors like which candidate’s voters are more energized (at the moment, Romney’s) and which candidate has the better operation for getting voters to the polls (traditionally, Democrats) could be decisive. And while the latter is a factor of hardened, institutional advantage and preparation, the former is far more mercurial.
Which raises a related point. While Gallup shows Romney leading by 2 points among “likely” voters, its raw data shows an Obama lead of three points among “registered” voters. Gallup has worked hard to tighten its projection of who likely voters might be, but this is an inexact science, to the extent that it's a science at all. The polling institute asks respondents seven questions believed to correlate with turnout (e.g., past voting habits, etc), but there have long been questions about how plausible it ever is to predict who will vote.
It’s a question that has bedeviled campaigns themselves since time immemorial. Ask anyone working inside the Hillary Clinton campaign four years ago whether they anticipated more than 220,000 Iowans coming out to caucus.
Another thing we know about the current state of the race is that what matters are the tallies in the states, not the national horse race. Focusing on the national numbers, rather than the state data, is like focusing on how many points tennis players win in a match, rather than games and sets. It's interesting, but a particularly crude indicator of the ultimate outcome.
And right now, the electoral map still favors Obama. He has held steady leads in the swing states—especially in places like Ohio—and unless Romney can overcome strong margins there, he cannot win.
A new poll out yesterday of Ohio, by CNN-ORC, shows Obama leading by 4 points among “likely” voters, and surpassing the critical 50 percent mark. In addition, the poll found Obama leading by 53-43 among “registered” voters, suggesting that an advantage in pulling his voters to the polls—an edge the labor-backed Obama campaign is counting on—could clinch the race.
Since this is the first time CNN has surveyed Ohio, there are not consistent trend-lines to analyze. But look inside the numbers and it’s clear that unless Romney can improve his margins among women (he trails by 22 points in the Ohio poll), he’ll be in trouble there. And without Ohio, which gave George W. Bush his victory in 2004, Romney doesn't have much hope of winning. This is why the campaigns will be spending so much time and money there the rest of the way.
But Romney also needs to improve his performance in other swing states like Florida. Whether he’ll be able to devote the sufficient resources and visits needed to significantly move the needle in both—without writing off other states, like John McCain was forced to do in Michigan in ’08—will be the big question the rest of the way.
Finally, it’s worth looking at history. While the Gallup poll has concerned many Democrats, it brought some news that, if the past is prologue, bodes well for the president.
Specifically, it shows Obama with a 52 percent approval rating as of the past week. The historical average for elected presidents in their 15th quarter (i.e., this stage of their fourth year in office) is 50 percent. George W. Bush was at 49 percent at this point in 2004. (Clinton and Reagan were both at 56 percent, but no one has ever expected Obama’s race to resemble theirs). And this is the highest Obama has rated in this metric since his first year in office.
This suggests that while the president has had his worst week of the general election, there's still no reason for my friend out there on the ledge to jump.
Blake Zeff is a former presidential campaign aide to Barack Obama and Hillary Clinton and a former aide to Chuck Schumer and Eric Schneiderman.