Early voting is underway in the states that are hosting competitive statewide races this November and where casting a ballot before election day is permitted. The early vote turnout is therefore starting to get media attention, with many of the stories talking about how Republicans seem to have an edge. Because I don’t have access to data from TargetSmart or analyses from the “NBC News Data Analytics Lab” I decided to calm myself and find out what I could from the Secretaries of State that supervise those elections. What I found is that these states vary in the type of data they provide (big surprise), which then requires in some cases assumptions and modelling to know what the partisan, gender, and age breakdowns of these voters are likely to be (thus the lab techs at NBC and TargetSmart being deployed to look at individual voter registrations and rolls and previous election voter make-up).
Following is what I was able to glean, through October 23. Please correct me if I’m wrong about something because I don’t know most of these states as well as a lot of you. I’ll try to keep a stripped down version updated, maybe in a spreadsheet version, until Election Day.
Nevada
Nevada has both in-person early voting and mail-in absentee and provides a partisan breakdown across both methods in most counties.
Total Vote |
In-Person |
|
|
ABSENT/MAIL-IN |
|
|
73,390 |
Dem |
66,525 |
43.2% |
Dem |
6,865 |
34.0% |
69,137 |
Rep |
59,248 |
38.4% |
Rep |
9,889 |
49.1% |
31,784 |
Other |
28,388 |
18.4% |
Other |
3,396 |
16.9% |
176,272*
|
Total |
154,161 |
|
Total |
20,150 |
|
* There are 1,961 additional mail-in votes not broken down by party designation
Democrats are ahead in the in-person vote, which is consistent with past practice. Republicans are ahead in the (much smaller) number of absentee ballots that have been mailed in, which is also consistent with voting patterns between the parties.
Florida
Florida has optional in-person voting in certain counties until October 27th when it is required to be offered statewide. Nevertheless the more populous counties are offering the option and reporting in-person numbers.
TOTAL VOTE |
In-Person |
Vote Cast
|
% Total Cast
|
Absent/ Mail-in
|
Returned |
% Total RETURNED |
OUTSTANDING |
570,732 |
Dem |
108,366 |
42.9% |
Dem |
462,366 |
38.7% |
881,866 |
623,582 |
Rep |
102,899 |
40.8% |
Rep |
520,683 |
43.5% |
765,801 |
245,641 |
No Party Affiliation |
39,155
|
15.5% |
No Party Affiliation
|
206,486 |
17.3% |
465,406 |
8,296 |
Other |
1,928 |
0.8% |
Other |
6,368 |
0.5% |
13,365 |
1,448,251 |
Total |
252,348 |
|
Total |
1,195,903 |
|
2,126,438 |
The proportion of the Nevada and Florida in-person vs. mail-in ballots is almost exactly reversed. Democrats clearly lag in the mail-in ballots being returned in Florida (although more Democrats requested mail-in ballots overall). However, all counties have started reporting mail-in returns; by contrast we are only two days into the optional one-week in-person voting period that is limited to certain counties. Although this lack of access may ultimately have its intended suppressive effect on the in-person early vote, it may be a bit of an overstatement to say that Republicans are more enthused in the early vote in Florida before more in-person numbers start coming in, especially when “early voting” in the media is usually portrayed as people lining up somewhere to cast a ballot.
Arizona
The Secretary of State has an interesting “dashboard” that draws from different data sets to present their early voting information and some voter demographic characteristics. Although Arizona offers early in-person voting at certain county offices along with mail-in voting, the site aggregates everything into a daily updated total of “Early Ballot Returns.”
|
EARLy Ballot Returns |
|
|
Demographics |
|
|
Dem |
164,738 |
32.4% |
Female |
48.9% |
|
Rep |
227,653 |
44.8% |
Median Age |
57 |
|
Unaffiliated & Minor |
116,258 |
22.8% |
Mean Age |
56.7 |
|
Total |
508,649 |
|
|
|
Republicans are indeed running ahead here. Something to keep an eye on, however, is the Unaffiliated & Minor number (the vast majority of which is Unaffiliated), which makes up a good chunk of the mix and, with hope, is breaking way more D than R this year.
Iowa
Like Arizona Iowa allows in-person absentee voting. And like Arizona both in-person and mail-in ballots received are counted together.
|
Early Ballot |
Requested |
Sent |
Received |
% Received |
|
Dem |
177,100 |
176,477 |
102,882 |
45.9% |
|
Rep |
149,917 |
149,640 |
80,932 |
36.1% |
|
No Party |
82,748 |
82,527 |
39,171 |
17.5% |
|
Libertarian & Other |
2,208 |
2,201 |
1,019 |
0.5% |
|
Total |
411,973 |
410,845 |
224,004 |
|
An additional feature that Iowa’s site shares with Arizona is that the early vote count is broken down by congressional district. Very helpful.
Georgia
Despite Brian Kemp’s and the Republican’s best efforts at cheating people out of their ability to vote, Georgia still offers early in-person voting in various locales, along with mail-in absentee balloting (a means of voting that traditionally favors Republicans but that also has Kemp worrying over Democratic strength). Of course, the state still needs to make it difficult because—although they separate in-person and mail-in ballot totals—they do not disaggregate the data by party. They do provide the top 5 counties in terms of turnout, which is updated on an ongoing basis
Total Cast |
741,090 |
|
top 5 Turnout counties |
Total vote |
% total |
In-Person |
|
|
Fulton |
98,950 |
13.4% |
Voted |
633,362 |
85.5% |
Gwinnett |
50,270 |
6.8% |
Absent/Mail-In |
|
|
DeKalb |
50,107 |
6.8% |
Returned |
107,728 |
14.5% |
Cobb |
29,395 |
3.9% |
Outstanding |
129,401 |
|
Henry |
20,518 |
2.8% |
The likes of TargetSmart and the NBC Lab have to do a deeper dive and make some inferences about the make-up of the early electorate in instances like these. Are their assumptions accurate given Georgia Democrats unprecedented early GOTV?
Texas
Texas started strong out of the gate with early voting on Monday. Texas is rather restrictive when it comes to who may vote by mail so the bulk of early voting in the Lone Star State must be done in-person. Like Georgia and the next two states the results are not broken down by party, so party-line enthusiasm must be inferred.
Total cast |
1,187,007 |
|
5 largest Counties |
Total vote |
% Total registered |
In-Person |
|
|
Harris |
181,916 |
7.8% |
Cast |
955,694 |
80.5% |
Dallas |
136,982 |
10.3% |
By-Mail |
|
|
Tarrant |
107,360 |
9.6% |
Returned |
231,313 |
19.5% |
Bexar |
83,179 |
7.6% |
|
|
|
Travis |
83,162 |
10.7% |
By Day 2 the early in-person vote has already begun to swamp the vote-by-mail numbers, which is to be expected. I’ve also included the top 5 largest counties in terms of registered voters and their relative performance in terms of early vote returns as a percentage of their overall registration numbers. Travis County is Austin and even though they are dealing with epic flooding and an ongoing water boil order they are leading among top 5 Texas cities in terms of percentage of registered voters getting to the polls early. That’s voter energy!
Tennessee
Although not as restrictive as Texas, Tennessee also has a list of specific statutory criteria to determine who may vote by mail-in ballot. Tennessee also combines the in-person vote totals with the mail-in ballots returned and does not break down the early vote by party. What they do provide, cumulatively and broken down by county, is a comparison of early vote from the previous 2 comparable elections, the 2016 Presidential election and the 2014 Midterms.
|
2018 voters |
|
total |
difference |
% change |
counties ahead of pace |
|
515,957 |
2016 |
609,021 |
-93,064 |
-15.3% |
3 |
|
2014 |
137,687 |
+378,270 |
+274.7% |
95 |
Clearly enthusiasm for early voting is way up compared to the last midterm election in 2014, but lags compared to the 2016 Presidential election. Not sure how to interpret that. I wonder how many Trump voters composed the early vote in Tennessee that helped drive him to a 26 point win in 2016? (I bet the people at TargetSmart could tell me).
Montana
Montana has in-person absentee voting whereby you can apply, receive, fill-out, and drop off an absentee ballot at a county election office. It will be opened and counted on election day with the rest of the mail-in ballots (as opposed to being immediately scanned, banked, and ready for an immediate vote dump as soon as the polls close on election day). Montana provides a daily updated spreadsheet of the ballots that have been requested and received broken down by county. That’s it.
|
number sent |
number received |
% received |
|
407,993 |
134,586 |
33.0% |
Indiana
Indiana has in-person absentee voting as well as mail-in absentee allowed according to statutory criteria. Indiana’s Secretary of State does not seem to aggregate county level absentee ballot return data. It appears to be available through some county election boards but that’s too much for me to try to compile.