Ever since the first debate in Denver, Mitt Romney has been on an upward trajectory in the polls. While he has leveled off somewhat over the last week, nationally he has turned a four-point deficit into a one-point lead. The lead actually jumps to two points if you include only the eight most recent national (non-online) polls. In those polls, Romney leads independents by an average of 17.5 points, which is a remarkable increase over the past month, and an amazing reversal of Obama’s 8-point lead with independents in the 2008 election. Romney has now been at or above 50 percent in Gallup for 12 straight days and Rasmussen for 5 straight days.
Those same national polls have a Democratic advantage of 4.4 points on average. In 2008, Democrats outnumbered Republicans by 7 points and President Obama won the election by 7.2 (52.9 to 45.7). The current polls have the Democratic edge dropping by just 2.6 points while Romney is up 9.2 points from McCain’s 2008 results. This shift is due largely to independents; it’s a worrisome picture for Obama that suggests he is losing, given that Democratic turnout in these polls is probably higher than what the electorate will look like after the votes have been counted.
While the numbers have brightened considerably for Romney nationally, he is still having a hard time breaking through in some of the key battleground states. Ohio is of course the most closely watched of the battleground states, and Romney is currently behind in the Real Clear Politics average, by 2.1 points. At the end of September, Obama held a much larger lead of 5.6 points, but Romney has been able to gain only 3.5 points in Ohio, whereas nationally he has gained 5 points. This has many wondering if Ohio is going to remain an impossible victory for Romney even if he has a lead of 1 to 2 points nationally.
There are many reasons why it is very unlikely that Romney could win nationally by 2 points and still lose Ohio. I went through some of them last week, but some of them could use a little more explanation.
In Ohio, Republicans tend to outperform their share of the national vote: In the last six presidential elections, only in 2004 has the Republican candidate performed worse in Ohio than he did nationally, and even that was a difference of only 0.3 percentage points. In the other five elections, the GOP candidate outperformed his margin of the national vote by an average of 3.1 percentage points. While it’s clearly possible for Republicans to perform worse in Ohio than nationally, it is very difficult to imagine a scenario where there is more than a point difference between them based on past elections. History would suggest Romney could not be up 2 points nationally while being down 2.1 in Ohio, which would mean that Republicans would be under-performing in Ohio by more than 4 points.
Democrats’ national turnout advantage is usually bigger than their Ohio turnout: Not since 1996, during Clinton’s reelection campaign, have Democrats had a larger turnout advantage in Ohio than they did nationally. In 2000 and 2008, Democrats were 2 points under their national turnout advantage and were actually 5 points under in 2004. Polls are currently showing an average Democratic turnout advantage of 6.3, which is 1.9 points higher than their current advantage nationally.
Both of these points suggest that it is more likely that either the national polls or the Ohio polls are wrong, rather than the possibility that both can be right. There is a big reason for this discrepancy: the partisan makeup of the Ohio polls.
As mentioned above, in current Ohio polls, Democrats have a party-ID sample advantage of 6.3 points. In 2008, Democrats had a 5-point turnout advantage in Ohio. That means that while national polls have the turnout advantage down 2.6 points, in Ohio it has actually increased 1.3 points. It is almost impossible to conclude that while the nationwide party-ID advantage of Democrats has dropped since the wave election of 2008, Ohio has actually increased over the last four years.
If that’s not enough, the Ohio polls have actually become more Democratic since the post-DNC polls that gave Obama the significant bounce that led many pundits to declare Romney’s chances in Ohio DOA. Of all Ohio polls from September 7 to September 19, Obama held an average lead of 4.2 points, with a Democratic party-ID advantage of 5.7 points. Today Obama leads by 2.1 points, with a party-ID advantage of 6.3 points. In the last month, while Romney has had surges in polls all over the country, the polling in Ohio has actually found more Democrats even while Obama’s lead was cut in half.
The biggest explanation for this discrepancy appears to be the prevalence of early voting. Many Ohio polls state that more voters claim to have voted early than the county records show; this inflates the likely-voter pools, with Obama voters pushing the Democrats’ advantage up as a result. It is a big problem for pollsters — many respondents want to say the socially popular thing, which is that they have already voted. When voters say they have voted (unlike when respondents say they plan on voting), pollsters ask no follow-up questions to determine how likely it is that they are telling the truth. Respondents who say they have voted go straight into the likely-voter pool, even though it’s clear that the number of respondents who claim to have voted is much larger than the actual early-voter turnout, based on the county records. From this, it’s reasonable to conclude that Democrats’ party advantage may prove to be lower, which would ultimately cut Obama’s lead in Ohio dramatically.
In little more than a week, we’ll know which polls are right — the national ones or Ohio’s — but history suggests that come Election Day, we’ll be looking back at the Ohio polls and trying to figure out how they could have been so wrong. With early voting such a key element of the Ohio election this year, it is impossible to know how much (or whether) it is skewing the polls until the election is over. It’s hard to imagining an electorate in Ohio that is more Democratic than it was in 2008, but without it, Obama probably can’t win Ohio. After seeing Romney’s rallies and increasing voter enthusiasm for him, compared with McCain’s rallies in 2008, it’s easy to understand why Team Obama is fighting so hard to promulgate the storyline that only state polls matter — that’s the only domino left to fall.