Early Voting thread.

[Adam Griffin]

Interesting chart of early voting history in Georgia. 2020's absentee ballot numbers really look like an aberration. Interesting that while in person and total early voting is going gangbusters absentee voting is at it's lowest level since 2014.

Important context:

Previously, counties would start mailing out ballots in late-August/early-September, but GAGOP restricted this post-2020 to no earlier than October 10 this year

Most people have only had their mail ballots in their hands for a week or so at this point. It's absolutely going to result in nerfing mail vote potential (as was the intent - especially in future years where the first Tuesday after the first Monday isn't so late in November, and basically shaves another week off, leaving only 3 weeks between when the ballots are mailed out by the counties and Election Day). Barring a weird calendar year like this, mail ballots in the future won't go out until the first day of in-person early voting begins.

However, there are currently an additional 172k mail ballots/applications outstanding in GA in addition to the ~55k returned. I imagine the final ABM number will be somewhere in the 170-190k range (in 2018, there were 223k returned ABM ballots; that was the first midterm ABM became somewhat popular due to the Abrams campaign pushing it heavily).

Just for reference (ABM #s and winner by margin):

2020:   1315294  Biden +30
2018:   223576   Abrams +24
2016:   207716   Trump +2
2014:   107023   Perdue +13
2012:   212695   Romney +17

It's funny, because prior to 2016, ABM was the GOP's best category of voting in terms of percentage of support (better than provisionals, better than early in-person, & even better than Election Day: largely because it was a practice only used by senior citizens at the time).


I'm also publishing daily EV updates/breakdowns in the Georgia megathread if anybody is interested. I usually update late in the evenings or in the early morning of the following day (data tends to be released for the day by SoS between 9-11 PM ET).

[Adam Griffin]

Georgia:

2022 Day 5 total EV compared to Day 5 total EV in:

2018, % of Electorate
+3 Black
-7 White*
+1 Female
-2 65 & up

2020, % of Electorate
+2 Black
-1 White
+2 Male
+17 65 & up

My estimates suggest the total EV at this point is around 4 points more Democratic than 2018 and 1 point more Republican than 2020 (margin) when taking an average of support levels for these four groups in 2018 & 2020.

And just for reference, the final EV figures (ABM + AIP) for the past three elections:

2021 Warnock   +13.76
2020 Biden       +5.73
2018 Abrams    +1.61

*I would take the white shift with a grain of salt, given the ever-increase of voters being labeled as "other/unknown" when first registering as opposed to their actual race or ethnicity. I imagine the real number when assigning everybody properly is probably 4-5 points less white than 2018 rather than 7-8

[Adam Griffin]

Georgia:

2022 Day 5 total EV compared to Day 5 total EV in:

2018, % of Electorate
+3 Black
-7 White*
+1 Female
-2 65 & up

2020, % of Electorate
+2 Black
-1 White
+2 Male
+17 65 & up

My estimates suggest the total EV at this point is around 4 points more Democratic than 2018 and 1 point more Republican than 2020 (margin) when taking an average of support levels for these four groups in 2018 & 2020.

And just for reference, the final EV figures (ABM + AIP) for the past three elections:

2021 Warnock   +13.76
2020 Biden       +5.73
2018 Abrams    +1.61

*I would take the white shift with a grain of salt, given the ever-increase of voters being labeled as "other/unknown" when first registering as opposed to their actual race or ethnicity. I imagine the real number when assigning everybody properly is probably 4-5 points less white than 2018 rather than 7-8

If it's more black than both 2018 and 2020, how does that mean a lower Democratic advantage?

With regard to 2020: slightly more male & far fewer young voters/far more older ones (the latter of which would be expected in any midterm EV). The estimate is a simple amalgamation of all 4 groups' pulls/tugs on the electorate relative to 2018 & 2020. Keep in mind those comparisons are cumulative EV as of Day 5 of in-person in those years (I just included the final/total EV margins for 2018 & 2020 as a separate footnote).

In reality, I suspect that the 65+ crowd thus far is actually a bit blacker than 2018 - possibly so for 2020 as well - as it's hard to see how the electorate is both blacker and older without this being true.

[Adam Griffin]

https://twitter.com/GabrielSterling/status/1583882601580072960

[Adam Griffin]

Day 6 AIP (+ABM) Voting: 87,831 voters cast ballots Saturday, for a grand total of 816,884 votes.

Breakdown of Saturday's voters:

Code:

46170	White	52.57%
28153	Black	32.05%
2260         Asian	2.57%
2067 	Latino	2.35%
9181 	Other	10.46%

48675	Female	55.42%
38909	Male	        44.30%	
247          Other         0.28%

Total votes thus far:

Code:

407481	White	55.53%
238372	Black	32.63%
10175       Latino	1.50%
9723	        Asian	1.47%
63302	Other	8.87%

394622	Female	54.27%
333383	Male	        45.57%	
1048	        Other	0.16%

[Adam Griffin]

Day 6 AIP (+ABM) Voting: 87,831 voters cast ballots Saturday, for a grand total of 816,884 votes.

Breakdown of Saturday's voters:

Code:

46170 White 52.57%
28153 Black 32.05%
2260         Asian 2.57%
2067 Latino 2.35%
9181 Other 10.46%

48675 Female 55.42%
38909 Male        44.30% 
247          Other         0.28%

Total votes thus far:

Code:

407481 White 55.53%
238372 Black 32.63%
10175       Latino 1.50%
9723        Asian 1.47%
63302 Other 8.87%

394622 Female 54.27%
333383 Male        45.57% 
1048        Other 0.16%

A lot of R counties will be way down tomorrow, any idea what big african american pop counties will be open and if there is a larger thana verage souls to the polls push?

IIRC a given county only needs to staff 2/3 early vote Sundays?

Sunday voting is optional and up to each county whether it does it or not. All counties are now required to have 2 Saturdays of early voting (today and next week), as opposed to the 1 that'd have been next week in previous cycles.

These are the only counties I'm sure that have at least 1 Sunday (if not 2) of early in-person voting (roughly 30% of the state's population; 44% black, 39% white VAP):

[Adam Griffin]

I'm not seeing a list, even switched to Edge in case it was an extension.

You might have been logged out or unable to see the image I posted due to post count (not sure if images are restricted for new registrants), but:

Cobb
Fulton
Dekalb
Rockdale
Clayton
Athens-Clarke
Baldwin

These seven counties combined were Biden +44 (versus the rest of the state, which was Trump +19).

https://snipboard.io/kXMQCp.jpg

[Adam Griffin]

I'm not seeing a list, even switched to Edge in case it was an extension.

You might have been logged out or unable to see the image I posted due to post count (not sure if images are restricted for new registrants), but:

Cobb
Fulton
Dekalb
Rockdale
Clayton
Athens-Clarke
Baldwin

These six counties combined were Biden +44 (versus the rest of the state, which was Trump +19).

Okay so the big 5, minus Gwinnett. Interesting. No Chatham or Muscogee or Doughtery? Sad times.

I guess tomorrow will be very good proportionally for Warnock but a small overall increase.

Unless I'm reading Georgia votes wrong the same day in 2018 saw only 6000 votes?

Also yeah I had to remove the image link from your post since it kept telling me I wasn't allowed to post links.

Actually, it does appear Dougherty is open both Sundays, so that's #8.

Picked through this list for the notables, but I can't be bothered to siphon through all 159 counties individually and deal with their individualized ways of listing schedules (I assume counties that don't explicitly list Sunday dates for voting aren't offering it).

https://sos.ga.gov/advanced-voting-location-information

[Adam Griffin]

Yeah I follow the GeorgiaVotes site, would love to see a similar fancy one for NV.

Gonna be interesting to see the women/black vote share today. Churches are heavily female.

Sorry, I read that wrong at first: thought you were looking for one for Georgia. Whoops!

[Adam Griffin]

Day 7 AIP (+ABM) Voting: 20,744 voters cast ballots Saturday in Georgia, for a grand total of 837,628 votes.

Breakdown of Sunday's voters:

Code:

8586 	White	41.39%
8561 	Black	41.27%
739          Asian	        3.56%
509 	       Latino	2.45%
2349 	Other	11.33%

12026	Female	57.97%
8639 	Male	        41.65%	
79            Other         0.38%

Total votes thus far:

Code:

462237	White	55.18%
275086	Black	32.84%
12751       Latino	1.52%
12722	Asian	1.52%
74832	Other	8.94%

455323	Female	54.36%
380931	Male	        45.48%	
1374	        Other	0.16%

[Adam Griffin]

Amazing numbers today. GeorgiaVotes actually has more black than white voters. Also women vs men was much more heavily in favor of women and youth % cumulative increased another 0.5% at the expense of 65+.

All in all a fantastic day for what will arguably the lower turnout of the 2 Sunday Voting days.

For Georgia specifically, the difference between GAVotes figures on their chart and my figures is that their charts (as opposed to their main page figures) only count in-person votes (something like 18k today), whereas mine includes ABMs processed in the total (close to 21k today), so while the AIP+ABM total was very close overall with whites edging out blacks today, it does appear that the mail ballots were meaningfully more white and the in-person vote was slightly more black, which when combined explains the difference.

[Adam Griffin]

Could someone please tell me what the actual f@ck is going on in Georgia? WTF are Black voters almost a plurality of the early vote for today?

Counties with Sunday EV (thanks to axiomsofdominion for compiling this).

Sunday Counties: Biden +35; 43% Black, 39% White VAP; 52% of state's population
Elsewhere: Trump +35; 67% White, 21% Black VAP; 48% of state's population

[Adam Griffin]

Even with this do the numbers still look good for Ds?

They correspond with a genuinely competitive race: they look OK for Democrats, but I'm hesitant to say - with another 12 days of EV remaining - that they're "good". They certainly don't look "good" for Republicans. A lot can change, however.

I'm expecting 2.7-3.0m early votes by the time everything is done, with a total turnout of 4.2-4.3m. If these patterns can roughly hold, then we're on track for a 2021-redux via ED vote where Warnock has a fantastic shot at making a runoff worst-case scenario.

[Adam Griffin]

Here's a map showing turnout as of Sunday by county - as a share of "% of '18" as shown by Georgia Votes (still not entirely clear what this is a measurement of, as the numbers don't line up with total 2018 EV or 2018 EV on this comparable day). Colors are in gradients of 10 points (the lightest shade is <30% & the darkest shade is >100%).

DRA data available here.



I also thought it'd be interesting to see how each grouping of counties compares to their presidential margins in 2020. This is what it shows:



Other than Athens-Clarke & Richmond skewing the data by being sub-30%, there's a fairly clear trend (though Bibb and Dougherty being in the 40s at the moment breaks things a tad as well; the remainder of the category was Trump +39).

[Adam Griffin]

https://twitter.com/RalstonReports/status/1584467450057236480

[Adam Griffin]

Georgia:

Day 8 AIP (+ABM) Voting: 148,968 voters cast ballots Monday, for a grand total of 986,596 votes.

Breakdown of Monday's voters:

Code:

89002 	White	59.75%
41947 	Black	28.16%
2276         Asian	1.53%
2220 	Latino	1.49%
13523 	Other	9.07%

82125	Female	55.13%
66527 	Male	        44.66%	
316          Other         0.21%

Total votes thus far:

Code:

551239	White	55.87%
317033	Black	32.13%
14971       Latino	1.52%
14998	Asian	1.52%
88355	Other	8.96%

537448	Female	54.47%
447458	Male	        45.35%	
1690	        Other	0.18%

[Adam Griffin]

Georgia

Here comes the White Cavalry:

Day 9 AIP (+ABM) Voting: 136,733 voters cast ballots Tuesday, for a grand total of 1,123,329 votes.

Breakdown of Tuesday's voters:

Code:

84975 	White	62.15%
35234 	Black	25.77%
2170         Asian	1.59%
2037 	Latino	1.49%
12317 	Other	9.00%

76062	Female	55.63%
60384 	Male	        44.16%	
287          Other         0.21%

Total votes thus far:

Code:

636214	White	56.64%
352267	Black	31.36%
17168	Asian	1.53%
17008       Latino	1.51%
100672	Other	8.96%

613510	Female	54.62%
507842	Male	        45.21%	
1977	        Other	0.17%

[Adam Griffin]

I've been projecting that we will see 2.8-3.0m early votes cast in Georgia & an additional 1.2-1.6m Election Day votes, for an average total of 4.3m votes.

As of Tuesday, GA's raw in-person early vote is 87% of what it was on the same day (EV Day 9) in 2020. If those EV proportions remain steady and are reflected in ED as well, that points to an electorate of 4.35m voters. That's just a tad shy of the turnout during last year's Senate runoffs.

[Adam Griffin]

EV gender breakdown (as of Tuesday):

[Adam Griffin]

Georgia

Day 10 AIP (+ABM) Voting: 132,444 voters cast ballots Wednesday, for a grand total of 1,255,773 votes.

Slight uptick in non-white vote compared to yesterday across all groups, but not enough to offset overall increase in white EV share. Females continue to see their share of the vote increase day-by-day.

Breakdown of Wednesday's voters:

Code:

81745 	White	61.72%
34339 	Black	25.92%
2201         Asian	1.66%
2017 	Latino	1.52%
12142 	Other	9.18%

73811	Female	55.73%
58326 	Male	        44.04%	
307          Other         0.23%

Total votes thus far:

Code:

717959	White	57.17%
386606	Black	30.79%
19369	Asian	1.54%
19025       Latino	1.52%
112814	Other	8.98%

687321	Female	54.73%
566168	Male	        45.09%	
2283	        Other	0.18%

[Adam Griffin]

Georgia

Day 11 AIP (+ABM) Voting: 127,087 voters cast ballots Thursday, for a grand total of 1,382,860 votes.

Similar trends compared to Wednesday, albeit a larger drop-off in whites & greater increase in black and other non-white vote. White margin among the total electorate increased by 0.7 points (compared to 1.1 points between Tuesday & Wednesday).

If history is any indicator, then we're likely reaching the zenith of white EV share - typically the second week of EV is the strongest among white voters, while black voters post their best turnout rates in Weeks 1 & 3.

Breakdown of Thursday's voters:

Code:

77333 	White	60.85%
33636 	Black	26.47%
2261         Asian	1.78%
2060 	Latino	1.62%
11797 	Other	9.28%

71802	Female	56.50%
55058 	Male	        43.32%	
227          Other         0.18%

Total votes thus far:

Code:

795292	White	57.51%
420242	Black	30.39%
21630	Asian	1.56%
21085       Latino	1.52%
124611	Other	9.02%

759123	Female	54.90%
621226	Male	        44.92%	
2511	        Other	0.18%

[Adam Griffin]

Florida? He [Trump] ought to be fearing wipeouts in WI, MI & PA.

It doesn't matter much: most Democrats don't have a shot in hell of winning Florida. It's time for Democrats to learn and to write off FL (and OH while we're at it), and start putting money into the swing states of the modern era. For the cost of what we spend targeting these 2 useless states, we could target GA, AZ & make serious attempts in TX (or shore up efforts in NC).

FL is #2 on my "States Democrats Should Stop Dumping Tens of Millions Into Because of Tradition, Only to Lose" list (OH is obviously #1), but 2020 may be too soon for GA to surpass it. Under neutral conditions, I definitely believe it will by 2024, though.

Imagine thinking NC is more deserving of investment at this point than GA.

Also:

STOP 👏 SPENDING 👏 IN 👏 FL

Even the Biden campaign itself admitted spending in FL was risky and potentially not worth it. I guess now that they're ahead meaningfully, they think it deserves targeting.

FL is a perpetually-close state but that doesn't mean it's a perpetual swing state. Its size and persistence on being a net-unfriendly state along demographic lines for Democrats is why the GOP has won 80% of statewide/federal races there in the 21st century; why Trump having no campaign there yet Hillary having the most advanced campaign in the history in the state still led to his win; why even Obama's '12 win there merely took him from a comfortable 303 EVs to 332; why any marginally-competitive presidential election this century has seen the GOP win it.  Additionally, there's no Senate contest and its metro areas are so scattered that I've personally coined it "the Ohio of the South". If Biden wins FL, he's won the election long since; if he doesn't and his campaign is on-point, he can still win (especially with states like AZ, GA & NC) comfortably without it.

Write off Florida. Write off Ohio. Even a sizable segment of Biden's campaign team wanted to do such, but special interests and deluded pundits stuck in the 00s won the day ultimately.

Do what I've been doing for the past 4 years and you'll sleep better at night.

See my screen name and past 5 years of posts. FL is too vast, too expensive and too R-leaning to continue subscribing to outdated "swing state" definitions. The Biden team almost came to the same conclusion last year, but CW unfortunately prevailed. Hopefully 2024 doesn't follow the same path.

[Adam Griffin]

Day 12 AIP (+ABM) Voting: 150,815 voters cast ballots Friday, for a grand total of 1,533,675 votes.

Friday's batch was less white and less black than Thursday's, with the other 3 groupings gaining ground. Slightly less female than Thursday as well. Total vote drops below 30% black for the first time this cycle.

Breakdown of Friday's voters:

Code:

91263 	White	60.51%
39208 	Black	26.00%
2884         Asian	1.91%
2807 	Latino	1.86%
14653 	Other	9.72%

84217	Female	55.84%
66169 	Male	        43.87%	
429          Other         0.29%

Total votes thus far:

Code:

886555	White	57.81%
459450	Black	29.96%
24514	Asian	1.60%
23892       Latino	1.56%
139264	Other	9.07%

843330	Female	54.99%
687395	Male	        44.82%	
2950	        Other	0.19%

[Adam Griffin]

Any chance somebody has the corresponding numbers for Day 12 in 2020? I know someone posted the numbers for the runoff at some point but those aren't quite as comparable (though still appreciated as a point of reference!).

I have the data - but only for in-person votes cast in those elections.

*Obviously that likely has a big impact on the 2020 figures given the sheer number of people who cast ballots by mail each day during that election (26% of the final statewide vote), and the fact that those ballots were skewed heavily-Democratic.

The figures for 2018 would be much more in line given that the percentage of votes cast by mail in that election (6%) is very close to the current figures for 2022 (9%):

2020*: 62.2% White, 23.0% Black
2018: 61.7% White, 27.2% Black

The white-black margin for 2018 Day 12 is identical down to the tenth (34.5 points) - though again, even the 2018 figures omit the relative handful of votes cast by mail that day.


EDIT: I found my update for Day 12 in 2020, but I didn't do breakdowns by race. 24% of all votes received on Day 12 (which in reality was Day 11 in 2020, as we only had 1 day of Saturday EV statewide in 2020 - today's) were mail ballots.

Georgia, Final Friday Update:, 214,549 votes were cast on Friday.

This includes 163,820 in-person votes and 50,729 mail ballots.

Total ballots cast in GA are now 1,625,250 in-person & 919,703 by mail, for a grand total of 2,544,953 (61.10% of 2016 total vote).

[Adam Griffin]

Georgia

Day 13 AIP (+ABM) Voting: 100,644 voters cast ballots Saturday, for a grand total of 1,634,319 votes.

Breakdown of Saturday's voters:

Code:

54099 	White	53.75%
30613 	Black	30.42%
2807         Asian	2.79%
2541 	Latino	2.52%
10584 	Other	10.52%

56571	Female	56.21%
43746 	Male	        43.47%	
327          Other         0.32%

Total votes thus far:

Code:

940654	White	57.56%
490063	Black	29.99%
27321	Asian	1.67%
26433       Latino	1.62%
149848	Other	9.16%

899901	Female	55.06%
731141	Male	        44.74%	
3277	        Other	0.20%

I'm going to start posting my projections of what I believe the two-way EV looks like. Unfortunately, I do not have statewide voter file access this year, which otherwise would give me greater certainty in how the vote breaks down, but I have decent confidence that these figures are within the proverbial ballpark.

Estimated Georgia EV Composition:
57.81% D (944,800)
42.19% R (689,519)
Net Margin: D+255279