Early & Absentee Voting Megathread - Build the Freiwal

[RI]

I get why people are doing it, but I wish there were more comparisons to 2016 rather than 2014. This is going to be a high-turnout midterm, and the patterns we'll see will be much closer to the 2016 patterns.

[RI]

Ballots were mailed out yesterday in WA.
As of 10/17, 3,411 ballots had been returned in King County.

I still haven't gotten mine #EastCoastProblems

Haven’t gotten mine either and I live in King

Clearly a concerted campaign to disenfranchise Atlasians.

I've got another data point to add to this troubling trend: I haven't received my ballot either.

Neither has my sister, and she lives in Seattle.

Overseas, out of state, service member ballots?

Ballots to the general public were mailed out on 10/17 which means they would arrive on 10/18 or 10/19

Here in Whitman County they aren't even mailing out ballots until today.

[RI]

It's perfectly plausible and perhaps expected that the Dem early vote in NV might be frontloaded this year.

[RI]

The numbers in Arizona are brutal -
http://www.arizona.vote/early-ballot-statistics

Republican landslide in Maricopa county, 117,000 to 75,000.

As others have pointed out, these numbers are worthless unless you include how they compare to previous years.

Overall AZ is currently 44.7 R - 32.4 D - 22.4 U. I don't have daily numbers, but the final 2016 distribution appears to be 40 R - 34 D - 25 U while 2012 was 43.1 R - 32.5 D - 24.3 U.

[RI]

https://www.nbcnews.com/politics/politics-news/republicans-outpacing-democrats-early-voting-key-states-nbc-news-finds-n922881

The guy who runs TargetSmart just slammed NBC on twitter for misinterpreting his data.

Dem voter file firm slams story unfavorable to Dems.

[RI]

That second graph is crazily similar for 2016 and 2018.

[RI]

Not sure if it's been posted yet, but TargetSmart has uploaded EV data by state, age and other factors and is displaying its models (which again, don't be like NBC and confuse them with vote share). I thought there were maps on there too, based on his tweet image, but I haven't found them yet:

https://targetearly.targetsmart.com/

So basically all the absentee votes in ND so far are elderly Republicans?

Edit: also the maps are here: https://targetearly.targetsmart.com/map.html

[RI]

I wonder what the % is so far for 1st time or atypical voters in both urban and rural areas. If reported, that would be the stat to look at as early voting progresses right?

TargetSmart suggests ~9% of TX voters so far had either never voted or voted infrequently. No idea about urban/rural.

State	New/Infrequent
AZ	18.8%
CA	21.1%
CO	14.4%
FL	17.0%
GA	18.4%
IA	15.4%
IN	13.3%
MI	6.8%
MN	17.5%
NC	15.0%
NJ	18.5%
NV	18.6%
OH	15.9%
TN	15.5%
TX	9.0%
WI	14.0%

[RI]

I wonder what the % is so far for 1st time or atypical voters in both urban and rural areas. If reported, that would be the stat to look at as early voting progresses right?

TargetSmart suggests ~9% of TX voters so far had either never voted or voted infrequently. No idea about urban/rural.

State	New/Infrequent
AZ	18.8%
CA	21.1%
CO	14.4%
FL	17.0%
GA	18.4%
IA	15.4%
IN	13.3%
MI	6.8%
MN	17.5%
NC	15.0%
NJ	18.5%
NV	18.6%
OH	15.9%
TN	15.5%
TX	9.0%
WI	14.0%

Well below the average. I’ll be curious to see if that holds as (presumably) more of the EV shifts to in person rather than mail

The problem with the TX TargetSmart data (and TX Early voting in general) is that it is very decentralized. Each county clerk is basically its own fiefdom and keeps its own data (and stores it in different ways). So the early vote data has to be collected separately from each county. Particularly for smaller counties, that is not possible because there are so many of them. That means - in contrast to states like GA or NC with early voting and more centralized election administration - it is pretty arbitrary as to what is included and what isn't included, and what days it gets updated. You can see that on the map view with the TX data, if you go forward and back in time to see when it is coming in from different counties. Because of the difficulty with collecting data for many rural counties, the available data skews somewhat towards urban/suburban counties.

Right now it is about 75% mail ballots, and the mail ballots are virtually all cast by seniors (about 97.5% of them), because there is not no excuse absentee ballots by mail except for people over age 65. That also biases it somewhat to people with good vote history.

I agree that the TX number seems rather low, but I'm not sure you can pin it on either of those things. Being just urban/suburban areas should bias the new/infrequent number upward if Beto's actually bringing in new voters. If the bias is just because of elderly mail voters, you'd expect a similar low number somewhere like ND, but even ND is 13.3%.

[RI]

The WA big 5 as of 10/23 (except for Snohomish, which reports in the morning the previous day's totals):
                  Ballots:   Turnout:
King:           52,776   4.09%
Pierce:         27,133   5.52%
Snohomish:  17,140   3.68% (9/22)
Spokane:     32,297   10.09% (wowza!)
Clark:          17,898   6.36%

....

That Spokane number is INSANE

LD 4 (Spokane Valley and east) has the highest return rate, although LD 3 (Spokane proper) is not far behind. The rural areas to the north and south of the county are lagging behind.

[RI]

I highly, highly doubt CMR loses.

[RI]

McDonald is a hack, end of story.

[RI]

McDonald is a hack, end of story.

To be fair, anyone saying that these numbers must mean someone will win, rather than being suggestive of certain possibilities, is a hack.

Yes, but I've interacted with him personally as well, and let's just say he has a massively overinflated ego and doesn't play well with others.

[RI]

The WA big 5 as of 10/23 (except for Snohomish, which reports in the morning the previous day's totals):
                  Ballots:   Turnout:
King:           52,776   4.09%
Pierce:         27,133   5.52%
Snohomish:  17,140   3.68% (9/22)
Spokane:     32,297   10.09% (wowza!)
Clark:          17,898   6.36%

King:             123,689   9.59%
Pierce:             43,015   8.75%
Clark:              28,081   9.96%

[RI]

The WA big 5 as of 10/23 (except for Snohomish, which reports in the morning the previous day's totals):
                  Ballots:   Turnout:
King:           52,776   4.09%
Pierce:         27,133   5.52%
Snohomish:  17,140   3.68% (9/22)
Spokane:     32,297   10.09% (wowza!)
Clark:          17,898   6.36%

King:             123,689   9.59%
Pierce:             43,015   8.75%
Clark:              28,081   9.96%

Where are you getting the results? sos.wa.gov is, uh, not user friendly (or maybe I’m a moron)

From individual county auditors.
I'll have a multi-county update later tonight (+Thurston, Kitsap, Yakima, and Whatcom)

You might want to check out this page I just found: https://wei.sos.wa.gov/agency/osos/en/press_and_research/PreviousElections/2018/General-Election/Data/Pages/default.aspx

[RI]

The WA big 5 as of 10/23 (except for Snohomish, which reports in the morning the previous day's totals):
                  Ballots:   Turnout:
King:           52,776   4.09%
Pierce:         27,133   5.52%
Snohomish:  17,140   3.68% (9/22)
Spokane:     32,297   10.09% (wowza!)
Clark:          17,898   6.36%

King:             123,689   9.59%
Pierce:             43,015   8.75%
Clark:              28,081   9.96%

Where are you getting the results? sos.wa.gov is, uh, not user friendly (or maybe I’m a moron)

From individual county auditors.
I'll have a multi-county update later tonight (+Thurston, Kitsap, Yakima, and Whatcom)

You might want to check out this page I just found: https://wei.sos.wa.gov/agency/osos/en/press_and_research/PreviousElections/2018/General-Election/Data/Pages/default.aspx

Very cool! Not convenient for the casual user, but maybe I'll play around with it later to add to my return tracking.

Looks like there have been 303,283 votes cast as of 10/23, if I imported into Excel right.

It looks like the files are not cumulative, so you have to export both the 10/22 file and the 10/23 file.

[RI]

Votes cast so far as % of 2014 total vote
1. Tennessee, 53.1%
2. Arizona, 47.8%
3. Nevada, 45.1%
4. Montana, 42.3%
5. Texas, 38.7%
6. Georgia, 36.8%
7. North Carolina, 32.5%
8. New Mexico, 31.1%
9. Florida, 29.0%
10. Colorado, 21.0%

Votes cast so far as % of 2016 total vote
1. Montana, 31.2%
2. Tennessee, 29.1%
3. Arizona, 27.7%
4. Georgia, 22.8%
5. Nevada, 21.9%
6. Utah, 20.4%
7. New Mexico, 20.1%
8. Texas, 20.0%
9. North Carolina, 20.0%
10. Florida, 18.2%

Map of TargetSmart's partisanship numbers:

Map of % of 2014 (color ramp +2 shades from normal):

[RI]

TargetSmart's latest updates in AZ, IN, MT, and OH were all slight improvements for the GOP. FL was completely neutral. NJ and RI favored the Dems.

[RI]

Big round of TargetSmart updates:

D improvement
AK, ID, KS, VT, WV

R improvement
AR, CA, CO, DE, GA, IA, LA, MD, NC, NM, NV, OR, PA, TN, UT

No change despite new votes
FL, IL, IN, MI, NE, TX, WY

Top ten states as percent of 2014 total (Estimated plurality)
1. Tennessee, 58.6% (65.2% R)
2. Arizona, 54.6% (54.2% R)
3. Nevada, 51.9% (47.6% D)
4. Texas, 51.2% (54.5% R)
5. Montana, 47.3% (46.8% R)
6. Utah, 47.2% (51.7% R)
7. Georgia, 41.0% (52.8% R)
8. New Mexico, 37.5% (57.5% D)
9. North Carolina, 36.0% (51.2% D)
10. Florida, 33.8% (49.9% R)

[RI]

Indiana:



Marion County absentee voting higher than in 2016 presidential election

This appears to be the case just about everywhere in Indiana.

[RI]

What is the early voting pattern for AZ? Is it like NV, with early vote for Ds coming in on the weekend, or is there a quirk to it like Rs flooding in near the end?

Not every registered R in AZ is a Pub voter, check the 20% gap between Pub early voter advantage and the end results in the Arizona special early in 2018. Especially since E-Day voting isn't all that meaningful. Its an area where change is happening faster then ID change.

There are some parts of AZ with ancestral DINOs as well (Greenlee is the most obvious, but not the only place).

[RI]

Why is Mineral County, NV ancestrally Dem?

Hawthorne Army Depot is the primary employer. The primary population lived in Babbitt, a planned community of government-owned duplexes from WWII until the 1980s. Most of Hawthorne's operations have been contracted out now to private companies.

[RI]

Your daily TargetSmart update. Seem like the Dems had a good weekend:

D improvement
AK, AZ, CA, CO, FL, GA, MA, MD, NC, NE, NM, NV, OH, SD, TN, TX, VA

R improvement
AR, IN, MN, UT, WV

No change despite new votes
IL, MT

Top ten as percent of 2014 total vote
1. Texas, 69.9% (53.7 R)
2. Nevada, 62.1% (48.5 D)
3. Tennessee, 61.8% (64.4 R)
4. Utah, 57.0% (52.0 R)
5. Arizona, 55.5% (54.2 R)
6. Montana, 52.9% (46.7 R)
7. New Mexico, 46.6% (57.8 D)
8. Georgia, 46.3% (51.5 R)
9. Florida, 42.9% (49.7 R)
10. North Carolina, 38.7% (51.8 D)

New feature allows for a comparison against 2014 at this many days out (large percent increases likely mean there were very few early votes at all in 2014):

State	% Ch Votes	% Swing
Alaska	+4%	R+6.8
Arkansas	-1%	D+0.2
Arizona	+53%	D+3.3
California	+24%	D+7.8
Colorado	-8%	D+10.8
Delaware	+143%	D+13.6
Florida	+45%	D+8.4
Georgia	+170%	D+3.5
Iowa	-1%	D+3.2
Idaho	+70%	D+5.5
Illinois	+64%	D+7.8
Indiana	+178%	D+6.8
Kansas	+66%	D+8.8
Louisiana	+44%	R+12.4
Maryland	+87%	D+6.1
Maine	+41%	D+13.0
Michigan	+47%	D+0.8
Minnesota	+190%	R+1.7
Montana	+53%	D+1.6
North Carolina	+244%	R+0.4
North Dakota	+36%	R+1.7
Nebraska	+83%	R+12.1
New Jersey	+166%	D+6.1
New Mexico	+105%	D+15.5
Nevada	+113%	D+14.5
Ohio	+30%	R+0.1
Oregon	-5%	D+6.7
Pennsylvania	+135%	D+13.6
Rhode Island	+44%	R+3.5
South Dakota	+113%	D+4.5
Tennessee	+371%	R+3.7
Texas	+171%	D+7.3
Utah	+187%	D+12.3
Virginia	+201%	D+13.5
Washington	-9%	D+7.4
Wisconsin	+64%	D+4.5
West Virginia	+76%	R+3.8
Wyoming	+70%	D+7.9
National	+75%	D+3.3

[RI]

TargetSmart is hot garbage. They do some of the worst polling I've ever seen, and regularly miss races by 10-20 points. Their modeling has been equally useless in past elections, so I wouldn't give them the time of day.

Maybe, but the overall patterns square with polling. Big Dem swings in the West and Northeast with smaller/mixed swings in the Midwest and South (excluding TX/FL).

[RI]

Spokane County is no longer the top return county in WA:

1. Lincoln, 43.4%
2. Jefferson, 41.1%
3. Pend Oreille, 36.9%
4. Garfield, 36.59%
5. Mason, 35.5%
6. Chelan, 35.3%
7. Columbia, 34.6%
8. Spokane, 34.5%
9. Ferry, 33.4%
10. Lewis, 32.5%

6 of the top 10 are in WA-05.

Returns in King County are pretty consistent except in WA-09, where turnout is way behind.