Ohio rests at the center of Obama's claim to an advantage in the Electoral College, but analysts offer diverging assessments of the race in the country's most critical state. Some say Obama has a big lead, others say it's a true toss-up and that Romney has the momentum. But what do the polls actually say in the Buckeye State? Here are five things to know.
1) Obama's lead is small, but consistent
On average, Obama leads by 1.9 points in surveys conducted entirely after the first debate.
But although Obama’s lead is relatively modest, it’s also consistent. Romney only leads in two polls with a clear Republican-lean, and both were conducted in the immediate aftermath of the first presidential debate. Put differently: Romney hasn’t led a poll of the Buckeye State conducted that wasn’t conducted from October 4-8.
2) Obama is beneath 49 percent.
Most Ohio polls show Obama beneath 49 percent, averaging about 47.8 percent of the vote. That gives Romney a more credible path to victory than he has in Wisconsin, Iowa, or Nevada, where the majority of polls show Obama at or above 49 percent. Obama has the advantage with 18 days to go, but there’s room for Romney to run the table and squeak out a narrow victory. If Ohio is supposed to be a firewall, it doesn't stretch high enough to preclude a Romney victory--at least not yet.
3) There isn’t strong evidence that Romney has made additional gains since his initial post-debate bump
While there’s plenty of talk about Romney’s “momentum” in Ohio, there isn’t much evidence that Romney has made additional gains since the first presidential debate.
Four polling firms have conducted multiple surveys since the first debate, and two show Obama making gains, one shows no change, and one poll shows Romney improving. Perhaps additional polls will show Romney making gains, but so far there isn't much evidence that Romney's post-debate push is paying dividends.
4) Obama holds a larger lead in polls that survey cell phones.
First, there is a clear divide between surveys contacting cell phone voters and those that do not.
At the moment, every automated survey that doesn’t contact cell phones now shows Obama leading by 1 point or less, while Obama leads by at least 3 points in every poll contacting cell phone (with the exception of the dubious ARG poll) voters. This division hasn’t always held as perfectly as it does today. As recently as two days ago, PPP showed Obama leading by 5 and SurveyUSA showed Obama leading by just 1. But the gap between automated and live interview pollsters contacting cell phones isn’t new, especially in Ohio.
Pew Research found that the absence of cell phone voters produces bias toward Republicans. Cell phone-only voters weren’t just younger and non-white, they were more Democratic than other demographically similar voters, so weighting wasn't a sufficient remedy. And when polls generally underestimated the strength of incumbent Democratic Senate candidates in 2010, cell phones were raised as a possible explanation.
Good polls start with a representative sample, so the possibility that automated pollsters can't sample a quarter of the electorate could undermine a prerequisite to their ability to sustain success. If you had to choose between leading in polls that can survey the entirely electorate and those that cannot, you would choose the ones that can. And when you read articles about campaign internals showing Obama up by 3 or 4 points in Ohio, remember that the campaign pollsters are calling voters with cell phones.
5) Several of Obama’s better pre-debate pollsters are still outstanding.
Although just about all of Romney's better surveys have reported post-debate results in Ohio, several of Obama's better pollsters haven't resurveyed the Buckeye State.
The Washington Post, CBS/NYT/Quinnipiac, and the Columbus Dispatch each showed Obama leading by at least 9 points prior to the first presidential debate. While Obama probably won’t lead by anything near as much in their next surveys, they each seem likely to show Obama leading by more than 1.9 points. In contrast, each of Romney’s best surveys has polled since the first presidential debate, and Republican-leaning firms like ARG and WAA have jumped into the fray.
The bottom-line: Ohio is close, but Obama's lead remains clear with 16 days to go.