Many of the polls in Ohio report data on the preferences of those early voters. We also have a pretty extensive archive of polls with similar data from the 2008 race. So what can we conclude, so far?
First off, the early claim by Romney's campaign that they were running neck and neck with Obama in early voting clearly does not hold up to scrutiny. We now have data from more than a dozen polls (all listed in reverse chronological order below) in Ohio, every one of which shows Obama with a significant lead over Romney among early voters, ranging from a high of 76%-24% to a low of 52%-46%.
Moreover, the margin of Obama's lead has not changed significantly as the number of early voters has increased over time. For example, Survey USA has conducted four polls in Ohio during October. In the first, Obama was ahead among early voters by 59%-39%; in the second it was 57%-38%; in the third it was 58%-39%, and in the most recent it was 56%-40%.
The most recent polls suggest that between a quarter and a third of Ohio likely voters have already gone to the polls, so Romney will have his work cut out for him come election day to make up that deficit. In seeing how likely that is, it is instructive to look at the 2008 data. When averaging the polls, it does appear that Obama's 2012 advantage among early voters is about 4 points lower than in 2008. Since Obama won by 4% in 2008, this could be a very close race. However, it appears that a greater percentage of voters are voting early this time around, so Romney has a smaller base to use on election day to climb back in it.
Finally, we can do a little head-to-head comparisons between 2008 and 2012 with polls taken by the same organizations.
Survey USA conducted polls in 2008 and 2012 that were concluded eight days before the election. In 2008, 22% of it's sample had already voted, favoring Obama by 56%-39% over McCain. In 2012, 25% of the sample had already voted, favoring Obama by 56%-40% over Romney--basically no difference.
Likewise, Public Policy Polling conducted polls in 2008 and 2012 that were concluded about a week before the election. In 2008, 30% of it's sample had voted, going for Obama by 65%-34% over McCain. In 2012, 33% had already voted, favoring Obama by a margin of 62%-35% over Romney.
We wish we had some data from 2008 from Rasmussen to compare to it's 2012 polls. However, even in 2012 alone, Rasmussen appears inconsistent. On Oct. 25, it reported that 35% of it's sample had already voted, giving Obama the edge by 52%-46% over Romney--the smallest margin in any Ohio poll so far. But three days later, Rasmussen reported another Ohio poll. This time, 32% had already voted, but Obama's margin was 62%-36%--a big swing from three days earlier. So, which Rasmussen poll to believe?
Obviously, both campaigns are in high gear in Ohio right now. But time is running out, as many Buckeye voters have already cast their lot. It's pretty clear that Romney's campaign will have to get it's voters out on election day to catch up. That's still quite possible, but for now Obama retains an edge.
OHIO EARLY VOTING
Note: As of today, we have 17 polls from Ohio with early voting data. We compared the average
of those polls to the average of all similar polls from Ohio in 2008. While
Obama, on average, has a big lead in the early voting in Ohio this time around
(average of 59%-37%), his lead in 2008 was even bigger, at 60.5%-34% over
McCain.
Oct. 26-29: Survey USA--25% already voted
Obama 56%
Romney 40%
Oct.
29-30: Public Policy Polling--33% already voted
Obama 62%
Romney 35%Oct. 29-30--CBS/Quinnipiac--doesn't say what % of survey already voted
Obama 60%
Romney 34%
Obama 62%
Romney 35%Oct. 29-30--CBS/Quinnipiac--doesn't say what % of survey already voted
Obama 60%
Romney 34%
Oct.
28: Rasmussen--32% already voted
Obama 62%
Romney 36%
Note: Not sure how to square this result with the Rasmussen poll just three days earlier!!
Obama 62%
Romney 36%
Note: Not sure how to square this result with the Rasmussen poll just three days earlier!!
Oct. 25:
Rasmussen--35% already voted
Obama 52%
Romney 46%
Oct.
23-25--ARG--28% already voted
Obama 55%
Romney 44%
Oct. 23-25: Purple
Strategies--26% already voted
Obama 58%
Romney 32%
Oct. 25:
CNN--59% EXPECT to vote early, no breakdown beyond that.
Obama 59%
Romney 39%
Oct. 23: Rasmussen--31% already voted
Obama 53%
Romney 43%
Oct. 22-23: Time--20% already voted
Obama 60%
Romney 30%
Oct. 20-22: Survey USA--26% already voted
Obama 58%
Romney 39%
Oct. 21: Suffolk Univ.--20% already voted
Obama 54%
Romney 41%
Oct. 21: CBS News--20% already voted
Obama 54%
Romney 39%
Oct. 18-20: Public Policy Polling--21% already voted
Obama 66%
Romney 34%
Oct. 16: Survey USA--18% already voted
Obama 57%
Romney 38%
Oct. 12-13. Public Policy Polling--19% already voted.
Obama 76%
Romney 24%
Oct. 5-8. Survey USA--11% of sample had already voted.
Obama 59%
Romney 39%