As many of you probably know, Early voting in North Carolina has started. Voting by mail has been going on for a few weeks now, but early voting in person just started on Thursday, October 18. North Carolina has "One-Stop early voting," which means that you can register to vote and actually cast your ballot all in one place at the same time. That is what is happening right now.
Because North Carolina makes its voting data freely available to the public, it is possible for us to see exactly how the election is playing out in NC. We can also compare 2012 early voting to 2008 early voting, to make an apples to apples direct comparison. And that is exactly what I intend to do.
You can throw out all the polls right now, because the actual voting is going on now, and we can tell what is happening. Overall, it looks a lot like 2008, except both Democratic and Republican turnout is higher.
Obama is probably ahead by about 8,000 votes right now in North Carolina, after the first day of One Stop Early Voting (on Thursday). By comparison, after the first day of One Stop Early Voting in 2008, Obama was probably ahead by about 17,000 votes.
Click on the picture below for a full sized chart.
Most of that difference between 2012 and 2008 so far is attributable to the fact that Romney managed to build up a slightly larger lead than McCain did in the mail ballots that were cast before Thursday. However, those mail ballots will ultimately be only a miniscule share of the total vote, and after a few days they will be swamped by the One Stop Early Voting votes.
Of the 156,129 votes cast on Thursday October 18th, most of which were cast in One Stop Early Voting, 39,355 were cast by White Registered Republicans. That is an increase of 13,661 votes from the first day of 2008 One Stop Early Voting.
However, minority turnout increased by slightly more, which roughly offsets the increased White Republican turnout. On the first day of OSEV in 2008, 45,634 minority voters voted (94% of whom were African American). On the first day of OSEV in 2012, that increased by 14,049 to 59683 minority voters (94% of whom are African American).
There is obviously still a lot of time on the clock, but anyone who thinks that Obama can't win North Carolina is just not looking at the record turnout that is happening right now, in front of our eyes.
White Republican tracks closely with 2008, but it is up. Those are pretty much votes in the bank for Romney:
But Minority + White Democrat turnout is also up over 2008. Those are pretty much votes in the bank for Obama:
Overall, Obama's total vote margin right now is probably just a tad below where it was in 2008:
But you can see that this has more to do with the mail votes that were cast before OSEV began than with OSEV:
Obama is winning early voting, and the vote percentages track pretty closely so far with 2008. The reason why the percentages is really that turnout overall is higher. More White Republicans are voting, so that drags down Obama's vote percentage. But at the same time, that is not really helping Romney very much, because minority and white Democratic turnout is also up, and is roughly offsetting (in raw numbers) Romney's gains:
You can also see that with the daily percentages:
In order to project how early voting is going, I assigned a certain percentage of votes to both Obama and the GOP, depending on each voter's race and party registration, as in the chart below:
I got these numbers primarily from 2008 exit polls. They will of course not be 100% accurate, but will be accurate enough for a good apples to apples comparison.
So what we are doing is projecting how votes are likely to be cast based off of voter demographics. Importantly, I have to assume that actual voter preferences have not changed between 2008 and 2012. What we are comparing is simply voter turnout, but not necessarily who each voter votes for. To get these estimates, we are comparing the demographics and party registration of early voters from 2008 vs. the demographics and party registration of early voters from 2012. This assumption is not realistic, but it is more realistic for North Carolina than for most other states, because North Carolina has relatively few true "swing voters." Another way to say this is that North Carolina is not a very "elastic" state. So the election in North Carolina is first and foremost a turnout battle. So if voter turnout in 2012 in North Carolina is similar or better to voter turnout in 2008, Obama could win North Carolina even if Romney makes up some ground nationally against Obama.
It's also worth noting that my numbers for 2008 are a bit higher than the actual 2008 numbers. This is mostly but not entirely because I am not removing undervotes and people who voted for third party candidates from the 2008 data. This is necessary for an apples to apples comparison because we can't really tell how many people are undervoting and how many people are voting for third party candidates, and thus the 2012 early vote numbers will also overcount the actual 2012 Obama and Romney votes to some degree. Specifically, when I apply my model to 2008, it predicts that Obama should have won 2008 early voting by about 1,458,227 votes to 1,183,702 votes (a margin of 274524 votes). In reality, we know that Obama won 2008 early voting by 1,382,199 votes to 1,077,145 votes (a margin of 305,054 votes). It is important to keep this in mind, but this does not make a material difference to our analysis.
Final note -
Time allowing, I am going to try and do this pretty regularly over the next few weeks. For DKE readers, I won't keep using the DKE tag to avoid cluttering the DKE side bar, so look for future updates if you are interested.