by Bob Grundfest
With three weeks to go before the election, and with all political eyes on the debate at Hofstra this evening, it’s time to take a look at the latest polling and to establish some baselines for the poll-a-rama that will ensue in the coming days.
The narrative since the first debate has been the Romney Rebound; the lift he’s received in the polls since his performance at the first debate in Denver, and the negative reaction President Obama received after his less spirited effort. Some news outlets are already proclaiming that Romney has momentum and that it’s only a matter of time before he overtakes the president in the swing state polls.
My take is different. Let’s look.
The Real Clear Politics Index shows Romney with a 0.4 point lead on Obama in national head-to-head polls. Huffington Post/Pollster has Romney’s lead at 0.2.Nate Silver at fivethirtyeight has Obama ahead by 0.6. None of these three pollsters, or any that I’ve seen, show Romney ahead in the Electoral College. The Politico Swing State poll showed Obama with a one point lead while other swing state polls show Romney with a lead. New Hampshire, Florida, Colorado and Ohio are said to be trending towards Mitt, which could set up a scenario where he takes the electoral lead with a good debate.
And yet, every trend line on these sites shows Romney’s lead or bounce fading as of this past weekend. A deeper look at the RCP polls tells a different story than what the general media is reporting. For example, the Gallup poll with Romney +4 is looking like an outlier. If we take that poll out, then Obama leads. If we take out Obama’s ABC/WaPo lead, the race is essentially tied.
In Ohio, where Romney is supposedly trending, I see no evidence in the RCP graph.
Florida is closer than we’re hearing because, again, we have a clear outlier, the TBT/Herald +7 poll giving Mitt a healthy bounce. Florida is trending red, but I’ll need to see another poll with healthy Romney numbers before I call it.
Colorado’s status is currently being influenced by an ARG poll taken right after the Denver debate, when Mitt had a great weekend of polling. Take that outlier out and Obama has a very slight lead.
There is a trend towards Romney in Virginia, and that has to concern the president because Virginia allows Romney a path to 270 without Ohio. Likewise, North Carolina seems to be in Romney’s camp, though a Gravis Marketing +9 poll seems to be an outlier there.
I am not suggesting that we ignore polls that don’t conform to the narrative, but that any poll that uses data from right after the first debate is going to inflate Romney’s numbers. He’s certainly closed the gap from where he was before Denver and he hasn’t made any statements like the 47% comment that got him into trouble in September.
But the evidence strongly suggests that Romney’s bounce is over. The good news for Mitt’s campaign is that he is now poised to overtake Obama if he runs an effective campaign from now on. The good news for Obama is that despite a terrible debate performance, he’s not only leading the electoral college tally, he never lost his lead.