THE BLOG
10/19/2014 10:54 pm ET Updated Dec 19, 2014

Early Voting Rounding the Bend

With early voting in full swing, more than 1.9 million people have cast a ballot in the 2014 midterm elections. I track early voting statistics here.

Predictably, dueling spin about early voting has emerged from the Democratic and Republican camps. Republicans point to a better showing among early voters than the 2010 election, and point to this as evidence of greater enthusiasm among their voters. Democrats counter that new Republican voter mobilization efforts have merely shifted the furniture around such that people who would have voted on Election Day are now casting an early vote.

When dueling claims were made it 2012 it was easy for me to adjudicate between the two camps. Republicans claimed that the electorate would be demographically similar to 2010 while Democrats said it would be like 2008. The early vote looked more like 2008 than 2010, so the Republican spin could be unwound.

This time both camps have truth to what they are saying. Republicans are doing much better than 2010 in key states like Iowa. Democrats are doing better, too, but they start from a higher base. While I cannot verify the claim that Democrats are mobilizing their Iowa supporters who may drop off in a midterm election, evidence from Iowa polling, from activity among persons without a party registration, and statistics from other states, support their claim.

Ultimately, these dueling arguments encapsulate a historical midterm dynamic that works against Democrats, and what they are trying to do to change it. Key Democratic constituencies -- young people, minorities, and the poor -- tend to vote at lower rates in midterm than in presidential elections. Without discontent against a Republican president to boost Democrats, as it did in 2006, the party is trying to reshape the electorate through mobilization of peripheral voters usually seen in presidential elections.

Republicans are ironically at a bit of a disadvantage when it comes to voter mobilization since their supporters tend to vote at higher rates; they haven't needed it as much as the Democrats. Nevertheless, Republicans have made a sizable infrastructure investment in this election that is at least capable of mobilizing their supporters to vote early. In the coming weeks we may see clearer signals in the polls and the early vote statistics as to which effort, if either, is more successful at truly bending the electorate to their favor.

Iowa

All eyes are on Iowa, so it is appropriate to start here.

On Thursday, the number of returned ballots was 185,254, which is more than half the 349,216 returned in 2010. In 2010, this midway mark was reached on Oct, 20, 2010, which -- because the election was Nov. 2 in 2010 and is Nov. 4 in 2014 -- is in five days later than when this milestone was passed in 2014. Furthermore, the number of ballot requests on Oct. 20, 2010 was 295,838. The 2014 number as of last Friday was 347,776. In 2010, the pace of ballot returns increased as the election neared and there is no reason to suspect this will be reversed in 2014. So, if past trends hold, Iowa early voting should approach, and may exceed, 400,000 voters. The most consistent signal from Iowa is that both parties' efforts are successful at mobilizing voters. Most likely, turnout will be high.

Republicans are doing better so far in mobilizing their voters to vote early. In 2010, Democrats held a commanding 49% to 36% lead among the 133,977 voters who had cast ballots as of Oct. 15, 2010. As of last Friday, Democrats hold a narrower 43% to 40% lead among the 185,254 who have cast ballots. Republicans are doing slightly better among the 347,766 who requested ballots as of last Friday, trailing Democrats 39% to 41%. (In 2010, Democrats also led ballot requests by a narrower margin, 46% to 37%.)

On Wednesday the DSCC released internal projections claiming a 15,000 Braley vote lead among early voters. At the time, 170,275 early votes had been cast. If accurate, this would give Braley a 9 point lead among early voters. This is a little at odds with a Des Moines Register poll that found a 16 point lead among early voters, and a Quinnipiac poll that found a 14 point advantage, but this may be due to the earlier time that these polls were in the field.

The DSCC also claims 31,000 Iowa early voters have not participated in the 2010 election, and that Braley has a 10,000 vote lead among these crucial voters.

I cannot independently verify these claims. However, the polling backs up Democrat's claims (if anything, the DSCC indicates the polls are over-estimating Braley's support among early voters -- something campaigns rarely do). Furthermore, trends I noted last week persist: Republicans appear to be mobilizing registered Republicans while Democrats are mobilizing both registered Democrats and persons without a party registration. So, these claims have some validity. I can more directly assess these claims in North Carolina and Georgia, which I turn to next.

North Carolina

Democrats started strong among North Carolina mail ballots. This was unusual as Democrats in the in-person early vote and Republicans win the mail ballots. Republicans have not tried to spin North Carolina, but that may change. I noted last week that Republicans finally started showing signs of life, apparently reaping the benefits of a mail ballot application drive targeted at registered Republicans. By the end of last week, Republicans had taken a 43% to 33% lead among the 62,125 mail ballot requests and were closing in on the Democrats narrow 39.2% to 38.5% lead among returned ballots. After their shaky start, Republicans look to be on back on track to equal their 2010 45% to 35% lead among mail ballots.

Where Democrats have some hope that they are indeed reshaping the electorate in their favor is among those with a 2010 vote history. Unlike Iowa, this data is readily available for analysis. Registered Democrats hold a 41% to 33% lead among the 32% of voters who have returned ballots and do not have a record of voting in the 2010 election, while Republicans hold a 41% to 38% lead among those who do have a record. These patterns are consistent with the DSCC's Iowa claims.

While these numbers are interesting, in the big scheme they are small change. Over 90% of North Carolina voters cast their early vote in-person in 2010. We will have to wait until the end of the last week before the election -- when early voting is now scheduled to occur -- to see if the far reaching changes the Republican state government made to voting will depress (or increase) Democratic turnout.

Georgia

In-person early voting started this week in Georgia. In just one week, 72.5% of Georgia voters have cast an in-person vote, while 27.5% have cast a mail ballot (or electronic ballot for overseas and military voters).

The demographic composition of the early vote changed dramatically. Georgia does not have party registration, so I analyze race, which is recorded on the state's voter registration file. Whereas Whites were 78% of voters who cast mail ballots, they were 66% of those who voted in-person early. Whites are now 69% of early voters by both voting methods. (The voter file I have is post-2012, so I am missing race for new registrants since 2012, I suspect that these new registrants are less likely to be White, so these statistics most likely under-estimate the non-White participation.)

This is a typical pattern for states that have mixed early voting methods. Voters tend to prefer to vote an in-person vote, particularly, Democrats or groups that constitute their base.

We can also see some evidence in support of the DSCC's Iowa claim. Among the 20% of Georgia early voters without a record of voting in the 2010 election, Whites are 64%, compared to 70% with a 2010 vote history.


Florida

Florida is leading the nation in raw votes, with 891,869 returned to date. With 1.6 million mail ballots outstanding, it seems likely Florida will easily surpass the 1.3 million mail ballot cast in 2010. The change in behavior is due to a change in the law. Starting in 2012, mail ballot voters can automatically request a mail ballot for the next election. With the Obama campaign encouraging their supporters to vote a mail ballot -- out of concern for a reduction in early voting days in 2012 -- Democrats closed a 14 point Republican advantage among mail ballots in 2008 to 4 points in 2012.

Gov. Scott's campaign released a memo touting their performance among mail ballots. I cannot directly assess their claim that they are doing better percentage-wise than they did at a similar point in time as 2010. In total in 2010, registered Republicans led 52% to 34%, so far in 2014 they lead by a narrower 48% to 35%. But it seems almost assured with the large volume of mail balloting that Republicans have surpassed their 2010 raw vote tallies.

Among the 1.6 million outstanding mail ballots to be returned, registered Democrats lead 41% to 38%. It seems most likely that Democrats will start catching up in returned mail ballots; there are some developing signs late in the week that could be early indicators. If not, perhaps they will surrender their ballots and vote in-person (indeed, many of the 2012 Democratic votes were "counter votes" -- mail ballots cast in-person at an election office). Perhaps Democrats will throw their ballots away, in which case the Democrats will have to assess the failure of their mail ballot mobilization effort.

Colorado

The Colorado Secretary of State released the first day of mail ballot return statistics last week. To give some perspective, Republicans led the 1.7 million mail ballots cast in 2012 by 37% to 35%, but numerous surveys found Obama leading Colorado's early vote and Obama carried the state 51.5% to 46.1%.

This year, Colorado is holding all-mail ballot elections. While all voters have a mail ballot, voters may still vote an in-person ballot at special polling locations starting Monday, Oct. 20. There is no real precedent for what to expect from in-person voting when a state is using all mail ballot elections, but we know in-person voting is Democrats' preferred voting method elsewhere.

With one day of reporting Republicans lead the 27,640 ballots have returned 46% to 32%. Republicans are leading as expected. Beyond that, it is a little early to make projections, especially since in-person early voting has not yet started.

South Dakota

South Dakota continues to putter along, with only 13,517 ballot returned, representing 4.2% of their 2010 total votes. This is a much slower early voting pace than other states with hotly contested Senate races. Given the high uncertainty of the three-way race, I expect voters will continue to hold their ballots so they can become more informed about their choices.

Montana

Montana leads the nation with its 77,724 returned ballots representing 21% of their 2010 total vote. A MSU-Billings poll this last week suggested the Senate race is not close, and neither is the House race. What may be happening here is the opposite of South Dakota. With uncertainty removed, people are feeling confident to cast their vote.

More

For more early voting statistics, see my early vote tracker.

This entry was edited to update South Dakota ballot statistics.