Saturday, October 29, 2016

NC Absentee Ballots Reach 1.5M as of 10-29-16

With the second full day of expanded early voting locations and hours in North Carolina, voters are obviously very interested in doing two things: voting and thus being done with this election. North Carolina's overall absentee ballots continues to run ahead of where the state was four years ago, with interesting differences within the voting method and party registration. 

All Absentee Ballots:

Through Friday, October 28th, which is the eight full day of absentee in-person voting (along with absentee mail-in voting), the cumulative total absentee ballots for North Carolina stands at 1,561,390; this includes all requested mail-in ballots and in-person ballots (see here for a brief explanation of my analysis process). 

This 1.5 million of requested absentee ballots represents an increase of 5.4 percent over the same day cumulative totals of all absentee ballots in 2012:

Of the 1.5 million absentee ballots, registered Democrats are 43 percent, registered Republicans are 31 percent, registered unaffiliated voters are 25 percent, and registered Libertarians are less than one percent. By voting method, the party registration breakdowns show advantages to the two major parties in different voting methods:

In looking at this 1.5 million by party registration and comparing it to 2012's same day totals by party:

In comparison to each party's total ballots in 2012 to this date, registered Democrats are 4.8 percent behind their same day totals, registered Republicans are 2.2 percent ahead, and registered unaffiliated voters are 35.5 percent ahead of their same-day totals from four years ago.

In looking at all absentee ballots by gender, women continue to be 56 percent of the ballots, with 48 percent of female voters registered Democrats, 29 percent registered Republicans, and 23 percent registered unaffiliated.

Among male voters, 38 percent are registered Democrats, 34 percent registered Republicans, and 27 percent registered unaffiliated.

In looking at all absentee voters so far by their racial classification and the voting method (mail-in versus in-person):

Of both methods of voting, whites are 72 percent, while blacks are 22 percent. In comparison to 2012's daily cumulative percentages by race, whites are over-performing (the green solid line over the green dotted line) while black voters are under-performing from four years ago (the orange solid line versus the orange dotted line).

The over-performance by white voters lead to me to do an interesting analysis last weekend to see the trends of white voters by their party registration. The next chart is the updated version of the past week of voting:

Compared to four years ago, white Republicans are performing at a lower percentage than they did four years ago, while white Democrats, having a slight over-performance for the past week, has leveled off to their 2012 pattern and white unaffiliated voters are continuing a strong pattern of over-performance. 

Accepted Ballots:

Of these 1.5 million ballots, the first pass is to determine which of these mail-in and in-person absentee ballots have been reviewed and "accepted" by the boards of elections as votes for the November 8 general election. Among the two voting methods (mail-in and in-person):

Among this 1.5 million ballots, the returned and accepted absentee ballots total 1,454,186 of both mail-in and in-person voting methods:

With the party registration among the two methods of absentee voting again showing partisan differences:

Accepted Absentee In-Person Ballots:

In comparison to the same day cumulative totals for all in-person accepted absentee ballots, this year's in-person absentee ballots are running eight percent ahead of where they were in 2012.

In terms of party registration of accepted in-person absentee ballots compared to four years ago:

Registered Democrats are five percent behind their 2012 same day totals, registered Republicans are 12 percent ahead of their same day totals, and registered unaffiliated voters are beating both partisans in performance, at 39 percent ahead of their 2012 same day totals.

Isolating this last week plus of in-person absentee ballots:

And then the party percentages of each day (days 9-17) and comparing 2012's to 2016's:

Next, in looking at race within party registration of the absentee accepted in-person ballots:

and comparing the 2012 daily percentages by race:

If black voters are going to turn out, then historically this week will be the key weekend to test whether they do in comparison to 2012's daily percentages. 

Next, if this year's accepted in-person absentee voters were registered in 2012, what method did they vote by?

Next, a common set of analyses I do based on region (urban, suburban, and rural county voters), native vs. those born outside of North Carolina, and generation cohorts, all by party registration:

Absentee by Mail-In Ballots:

The second method of voting absentee in North Carolina is by mail. The cumulative total accepted absentee mail-in ballots for this year is running behind 2012's same day totals; this year's total is only 83 percent of where the totals stood four years ago:

As noted above, the party registration break-down is 41 percent Republican, 33 percent Democratic, and 26 percent unaffiliated.

Registered Republicans, however, are significantly behind their 2012 same-day totals in this voting method:

Registered Republicans are only 64 percent of where they were four years ago on this same day (down nearly 23,000 accepted absentee mail-in ballots), while registered Democrats are 98 percent of where they were four years ago and registered unaffiliated voters are 111 percent of where they were in 2012 on the same day. 

The outstanding ballots broken down by party registration is:

Some New Analyses:

Finally, I was interested in two areas of analysis: first, the 'propensity' of this year's absentee ballot voters and if they participated in past elections, and second, the rise of the unaffiliated voter in this early voting. 

First, I took the voting records and matched the current absentee voter pool against whether they had voted in the past four general elections (2008, 2010, 2012, and 2014), and then broke it down by party registration:

Of this year's absentee ballot voters (all at this point and both mail-in & in-person), 46 percent of the voters had participated in the past four elections by casting ballots. In isolating just those voters who have voted in-person so far:

Nearly half (49 percent) of all voters who have cast accepted absentee in-person ballots so far participated in the past four general elections. I then ran the same analysis but for only the past two presidential elections (2008 and 2012):

Finally, I was interested in the rise of the unaffiliated voter and if they had participated in the state's March 2016 presidential primary election in one of the two party primaries (in North Carolina, unaffiliated voters can select which party they wish to vote in during a primary election).

Out of 391K unaffiliated voters who have cast any type of absentee ballot so far, 50 percent of them cast ballots in the March 2016 primary election: of those 55 percent cast ballots in the Republican primary, while 45 percent cast in the Democratic primary. 

The following charts present both the raw numbers and in percentages by generational cohort:

If one considers this a 'crude' proxy for how unaffiliated voters may be breaking, one could read into that unaffiliated voters might be breaking slightly to the Republican side. But there are a LOT of factors that could make this a very weak form of a proxy to see how these unaffiliated voters are breaking, let along there's another nearly 200K voters who didn't vote in the March 2016 primary and thus no indication.