Georgia senate seats runoff(s) megathread

[ProgressiveModerate]

https://docs.google.com/spreadsheets/d/1CHaYVrkw7GVGga4Jna_d4Kau8ESLYHCEKO9-INR_P9o/edit?usp=sharing

Made a more detailed calculator which outputs 609k for Ds and 508k for Rs at the moment. Here, you can specify how many votes were cast absentee, in person early, and on election day.

The full model (on a seperate spreadsheet) currently expects a Trump +9.43 electorate which would essentially guarentee a Perdue and Loeffler win, but it's not quite not done yet and needs to be refined because it assumes a disproportionate amount of the remaining vote will be in person

It's down to Trump +8.67 today using the same methodology. This number is likely to decrease as we get closer to election day; we'll see if it flips.

Trump + 7.87 today. This number is almost certainly going to continue to go down, we'll just have to see if it becomes Biden positive

[ProgressiveModerate]

This is probably a nothingburger, but I also thought the cal cunningham texts were a nothingburger, and they probably cost him some votes.

I think it depends upon how blown up this becomes. If it never blows up it will have no meaningful impact on the outcome. If it blows up throughout the state it could cause serious problems for him. However, if it blows up "too much", it'll become some sort of political football and ultimately have very little impact on the final outcome.

With Cal Cunningham, it's truly impossible to know how much it cost him, if at all. My guess based on his underperformance of Biden was maybe a 1-2% shave, but that could make all the difference in a close race.

[ProgressiveModerate]

https://twitter.com/JackPosobiec/status/1341567038834270209

lol.. why are you replying with some random Posobiec tweet, that has nothing to do with the subject at hand.

Watch the clip and you'll see it's literal bodycam footage.

Because you debase your own credibility as an Atlas poster when you rely on known purveyors of fake news for your information.

How have you jumped to the conclusion that I rely on Pobosiec for my information..? Because I linked one tweet which contained the clip which is being discussed..

Atlas at large doesn't have any real degree of credibility, so I don't have far to fall in that sense.

Ultimately, quoting random people on twitter isn't the best source but TBF, there's so many on this forum who do it. Overall, I wish we would stop quoting random tweets from random people on this forum overall, not specifically targeting you, but thought I would bring it up while we're on the subject.

[ProgressiveModerate]

https://twitter.com/JackPosobiec/status/1341567038834270209

lol.. why are you replying with some random Posobiec tweet, that has nothing to do with the subject at hand.

Watch the clip and you'll see it's literal bodycam footage.

Because you debase your own credibility as an Atlas poster when you rely on known purveyors of fake news for your information.

How have you jumped to the conclusion that I rely on Pobosiec for my information..? Because I linked one tweet which contained the clip which is being discussed..

Atlas at large doesn't have any real degree of credibility, so I don't have far to fall in that sense.

Ultimately, quoting random people on twitter isn't the best source but TBF, there's so many on this forum who do it. Overall, I wish we would stop quoting random tweets from random people on this forum overall, not specifically targeting you, but thought I would bring it up while we're on the subject.

I would agree.. in this instance, I only quoted Pobisec as he had the clip of the bodycam on hand.

But, that one post has clearly exposed me as relying on purveyors of fake news for my information. My reputation is ruined..  

Yeah, I think you're fine here for the sake of showing the clip, which is definately something very relevant in this race

[ProgressiveModerate]

Nate Cohn, whom i respect more than any other data yoda, is tweeting daily now that the EV data is beyond dem’s wildest dreams

We’ll see

I'd say right now it looks pretty good, but it also remains to be seen how many Rs show up on Election Day too. Maybe more than in november are straying away from any type of early voting. Who knows

My model also projects that the final turnout has about a 38% chance of being a Biden majority electorate + another 32% for Biden plurality, the real question is if a Biden leaning electorate will be enough to win. Perdue/Loeffler + 5, 6, maybe even 7 is still possible but at this point it's pretty clear Ossoff and Warnock aren't going to get completely blown out of the water.

[ProgressiveModerate]

Perdue/Loeffler + 5, 6, maybe even 7 is still possible

No it isn't. Why would Ossoff/Warnock do worse than John Barrow and Lindy Miller? That's literally not going to happen.

It’s a fringe outcome on my model but a Trump +6 electorate is still below the 97th percentile for Rs in my model

[ProgressiveModerate]

If one side wins both runoffs, they will have gotten extremely lucky with winning narrow senate races to form this congress, especially if they both break narrowly for Republicans.

[ProgressiveModerate]

Honest question:

is there ANY good news for the gop in any of this ev data?

it looks dreadful.

-Polling is a virtual tie, and pretty much 90% of the time, a statistical tie in polling broke for Rs in 2020.
-Strongly D counties don't have all that much higher turnout than strongly R counties, if you're a Democrat, you prolly want a slightly bigger gap
-The holidays are coming up which will prolly lead to fewer EVs for at least a few days
-Never underestimate Rs ability to turn up on election day

[ProgressiveModerate]

Is early vote data actually worth reading into? People on here thought blue Texas was inevitable based on massive early voting turnout, only to be sorely disappointed.

To some degree yes. Biden did end up getting 1.5 million more votes than Clinton, it’s just that the Election Day turnout was also pretty massive. Heck, even Hegar got 1 million more votes than Clinton did. To me, it seems like the EV data can tell us a lot about D turnout with Republican turnout being more of a wild card that we won’t know to the day of.

Also, in a place like TX, there were very few breakdowns so we knew turnout was high and that urban counties were passing there 2016 total, but we didn’t know much else about who these people are

[ProgressiveModerate]

If one side wins both runoffs, they will have gotten extremely lucky with winning narrow senate races to form this congress, especially if they both break narrowly for Republicans.

Also, if both seats flip, it kind of destroys the "structural likely R senate" meme.  Biden won 25 states.  If Senate Dems can defeat GOP incumbents in Biden's narrowest state, they are in much better shape than anyone thought after 2018.

There is a error on your list

Lol I can’t find it and I’m going insane at this point. Any care to tel me what it is?

[ProgressiveModerate]

If one side wins both runoffs, they will have gotten extremely lucky with winning narrow senate races to form this congress, especially if they both break narrowly for Republicans.

Also, if both seats flip, it kind of destroys the "structural likely R senate" meme.  Biden won 25 states.  If Senate Dems can defeat GOP incumbents in Biden's narrowest state, they are in much better shape than anyone thought after 2018.

There is a error on your list

Lol I can’t find it and I’m going insane at this point. Any care to tel me what it is?

ND-SEN 2018 ; you have mixed Hoeven and Cramer.

Oops, thanks for letting me know I’ll change it ASAP

[ProgressiveModerate]

I agree the bolded states are sure losses, and the other 3 are not sure wins.  However, bouncing back in New England and to a lesser extent Michigan + alleviating CA/NY vote sink concerns in the medium/long run + Alaska and Kansas/Nebraska looking attractive by the middle/end of the decade + Florida potentially becoming a big state GOP vote sink by then are why things look more even now than in 2017 or 2019.  They will be in a hole come 2025, but it would be a lot more manageable with this coalition than the 75% Dem CA, 55% Dem TX, and 55% GOP Midwest/New England world it looked like we were careening toward in 2016/18. 

MT is definitely not a sure loss, and I wouldn’t completely write off Brown either although he’s almost certainly the underdog. I also don’t see NH/ME/MI/KS/NE being nearly as promising for Democrats as long as they actually control the White House (which they probably will until 2028 at the very least), although I agree on AK. The GA seats are probably gone for the GOP before the end of the decade regardless of which party occupies the White House.

Even with GA firmly in the D column, there are still a lot of potentially vulnerable Senate seats Democrats currently hold in competitive/-ish states (Tester, Manchin, Brown, Stabenow, Peters, Baldwin, Sinema, Kelly, Rosen, Cortez Masto, Hassan, Shaheen, King, arguably one or two of the NM/MN seats as well but that’s stretching it), so some combination of one or two Republican-friendly midterms in 2022/2026 + a rapid decline in split-ticket voting and/or a national GOP victory in 2024 could wipe most of them out. GA is definitely not going to be part of the GOP's path of least resistance after 2021.

This is why PR and DC statehood need to happen. If not Democrats will be blocked out of the senate for at least two decades, despite probably winning the Popular Vote in every nationwide senate election (even if you don’t include one of the California candidates).

I wouldn’t go as far as to say that. Things can change relatively rapidly; remember in 2008 when Republicans fell deep into the minority and worried they would lose the body for a political generation? The real issue for Ds was they didn’t have the traditional 2 wave years in their favor to rebuild their senate majority, they only had 1 because Trump lost re-election. Had Trump won in 2020, they could’ve got into the mid 50s in 2022.

Furthermore, 2020 actually spelled some reviving news for Ds in the senate. In 2016 and 2018, people thought the main re-alignments were urban-rural, but now that seems to be shifting into an argument educational divide, and there are a lot of well educated small states, especially in the North East and West. The real fear was Democrats would only gain GA/TX/NC but lose the rust belt and north east, but that doesn’t seem to be where we’re heading.

It seems like they will still be slight underdogs in the senate, but it seems to me their current problems won’t get worse than they are now. The real issue for the was the break in the 2-term Presidents rule depriving them of a wave year to build up downballot.

[ProgressiveModerate]

In the past few days, my model has grown more confident that we’re dealing with at least a Biden plurality electorate. Even giving the GOP the benefit of the doubt, assuming 90% GE turnout and a complete halt in early votes, with E-day votes breaking how they did in the GE, that still only yields a Trump + 5 electorate. The median outcome is about Biden + 3 (+2.87 to be exact), which is right around the inflection point in the Perdue - Osoff Senate race. These will likely be close, and while Perdue and Loefeller could still win by larger than expected margins (+7) like the poster above said, it’s becoming an increasingly more unlikely scenario statistically. These next few days, I’ll keep updating how the early vote data impacts the model as the holidays roll around.

[ProgressiveModerate]

https://twitter.com/CNN/status/1342494520605368325?s=20

Well we know a ton of money didn't do sh-t in South Carolina and it damn sure didn't in rural, white, undereducated Kentucky. Let's see what it does in a D-trending Biden +0.3 state with a Black base, dense Black suburbs, significant Hispanic and Asian population, along with college-educated white suburbs that have embraced mainstream Dems like Abrams and Biden.

The 2020 election made 1 thing very clear
Georgia is not like South Carolina and Kentucky at all..

Jaime Harrison and Amy mcgrath never stood any chance on those states
In mcgrath case polls showed that she was doomed even as some dems kept foolishly throwing money at her


Ossoff and Warnock on the other hand Actually have a will real chance to win in Georgia
even if the both go on lose it will not be by the kinds of margins that we was in South Carolina and Kentucky

I don’t think money necessarily switches many votes, but can certainly help with outreach to turnout voters which will be key in this race. When a party over saturates a race with money, in a state that leans the other way, it nationalized the race and increases polarization. While I don’t think he would’ve won, I think Harrison could’ve come closer to Graham had he not gone crazy with money in the last couple months.

[ProgressiveModerate]

What do you guys think of this county map?

Will try to make an interactive spreadsheet with several sliders which will output county results based upon what you personall;y think turnout will be, sort of like a swing o' meter (The map above is a 0.88% R win)

[ProgressiveModerate]

Seriously, who are these 65k + voters who have thus far voted in the Georgia runoffs who didn't vote in the GE? Some may be people who since turned 18, but that isn't enough to explain all of it, plus I doubt every single person who since turned 18 has been registered.

[ProgressiveModerate]

Seriously, who are these 65k + voters who have thus far voted in the Georgia runoffs who didn't vote in the GE? Some may be people who since turned 18, but that isn't enough to explain all of it, plus I doubt every single person who since turned 18 has been registered.

Maybe some could be Dem-leaning folks who don't pay a lot of attention to politics and thought GA was a titanium R state, so voting in the GE would be useless. But they saw GA vote blue and know it's possible for a Dem to win.

Hmm, interesting theory. Generally, swing-states tend to have slightly higher turnout than non-swing states, so now that from the public's perception, GA now has potential to go blue, their vopte can actually matter. This is also probably the first time in a long time GA has been this heavily targetted; throughout the 2020 cycle, both campaigns had it as a lower tier priority, so maybe just better outreach of making the election known.

I do wonder though if you can flip the same logic on it's head and say that conservatives who once took the state for granted now feel more of an obligation to GOTV. As someone from NYC, I have stumbeled across some who take NY's status as a blue stronghold for granted.

[ProgressiveModerate]

https://twitter.com/WinWithJMC/status/1342126489383403521?s=20

https://twitter.com/WinWithJMC/status/1342116169961140230

I'm feeling really positive about the EV stats. The last week of early voting should skew younger and Blacker and Fulton, DeKalb, and Gwinnett having early voting this weekend is helping us pad our numbers. Perdue/Loeffler are going to need Election Day to make up more of the vote than it did in November or win 2/3 of the ED vote both of which seem daunting. We will see......

I am too, and my current estimates the final electorate will have a 79% chance of being at least plurality Biden. However, Perdue still outperformed Biden, and the e-day vote will make things tight. Will it be enough to put Perdue back ahead? That is yet to be seen. These seem like they are going to be close.

Estimate of current votes casts for the Perdue - Ossoff race is Ossoff + 8.9





Will also note that in order for Perdue to win my model estimates he will need to net about 200-300k votes on election day assuming trends roughly continue. The % of the vote that will be will depend upon turnout.

Obviously my model is imperfect, and I personally think that the number he needs to net will be closer to 175k-225k range, but that's speculation on my part

Overall though, most data indicates the current vote totals haveing both Ds anywhere up from 6-12 points

[ProgressiveModerate]

I don't live in Georgia. but from what I've been reading and hearing about, it seems like Democrats are actually the more united party going into these runoffs. Really the only Democrats who aren't on board with Ossoff and Warnock seem to be those few fringe moderate voters who voted for Biden out of spite for Trump, and even then, many are splitting their ballot Warnock-Perdue. Outside that though, everyone from neoliberals to BernieBros seem to be united in their support for voting for Warnock and Ossoff.

While Republicans aren't in a terrible place in these races, it does seem like there's a bit of disunity within the party, mainly almost on the loyalty test of how loyal Perdue and Loeffler are willing to be to Trump. There seems to be a small yet loud presents of Trump voters who think Trump should've won who threaten to not vote in the runoffs unless Perdue and Loeffler are more aggressive in protesting the results, particularly those that are very "Trumpy". It's yet to see how many Republicans will actually boycott the runoffs; we'll truly never know, but it still creates tension within the party regardless.

Would be curious to see if anyone who's actually from GA right now can confirm or deny this

I still personally think Perdue and Loeffler are very very very slight favorites, but this does seem to be a potential downfall for them, and I do see a small window for a lopsided D result if a combination of Trump not directly being on the ballot and lack of trust of the establishment causes even just 5% of Republican voters to stay home who would've otherwise voted.

[ProgressiveModerate]

https://twitter.com/tbonier/status/1343228374257459202?s=20

RIP

Is that the Walmart data? They tend to be a bit iffy, but still interesting nonetheless. According to my stupid model, the e-day vote will end up being a larger share of the vote, even though currently the early vote seems to be more D than in the GE, which could be a sign Ds have an enthusiasm  edge. At the same time, that share is likely to be slightly less pro-R than it was in the GE based upon who the outstanding likely voters are. Again, nothing is finite since we don’t have any pure party registration data, but Democrats should feel good about the position they are in, though if they’re favored is a much harder question to answer

[ProgressiveModerate]

Literally everything seems to cancel out in these races; for every point about why a canidate may be favored, there is an equally strong counterpoint. Some examples:

Early vote data looks good for Democrats but the GOP will likely make a huge turnout on election day

Biden won Georgia but Republican's still got a plurality or majority of votes statewide in both Senate races and in the House PV

Warnock and Ossoff seem like better canidates and have the enthusiasm edge but ticket splitting likely works in favor of Perdue and Loefller

History favors the GOP as being more reliable voters in Georgia but Democratic drop-off in runoffs has been getting smaller and smaller and you even have some who threaten to boycott the runoffs now

Republicans largely outperformed polling this cycle, but Georgia was one of the few places where polling held up pretty well, even underestimating Ds in some cases

[ProgressiveModerate]

Today, Democrat's estimated lead is up to 9.7 points in my model, up from 8.9 yesterday. The GOP is going to need to net a lot of votes on e-day, though as others have pointed out, their strategy seems to be betting on the e-day vote.



As we well know, all it takes is massive e-day turnout for these seemingly large D leads to dissolve, but everyday the gap doesn't close means more votes the GOP will have to net.

[ProgressiveModerate]

^^^ So your model includes all the votes cast/returned up to 12/27?

My brother voted in person for Ossoff/Warnock today. He said the line was long (moved quick) and it was pretty much all Black people. We do have three other voting locations for the first time and we are a polarized county as far as where people live so that has a lot to do with the demographics of where people early vote.

Yes that is correct, it’s compiled all votes that have been cast.

My model also seems to support the strong AA turnout take, as you can see Dems having leads in many black belt counties that Biden ultimately lost to Trump.

[ProgressiveModerate]

VBM + IN PERSON EARLY
Nov GE: 4.03M (56.5% white, 27.7% black) ---> +28.8 white

.....

12/28: 2.13M (55.2% white, 31.6% black) —> +23.6 white

I think that Ossoff & Warnock have a greater than 50% chance of winning  (and Warnock will receive more votes than Ossoff)- But there are a few things in EV numbers that do worry me:

Age of early voters so far
Runoff EV over 56-years-old: 57% (General election EV: 46%)
Runoff EV under 34-years-old: 15% (General election EV: 21%)

I think Dems really need strong turnout from those under 34... which isn't happening yet.  Hopefully a lot of young ppl are going to vote & mail in ballots in the next 2 days (last minute)... but these stats are the most troubling for Dems imo.

Potential overestimating the Dem advantage based on white/black %
Although the Black % of EV has increased, and the "White-Black %" has decreased compared to the general election- I think this is potentially not as much of an advantage for Dems as it may seem.  Mainly b/c the Percentage of White voters has stayed about the same.  So if you look at all non-white voters... the number is basically the same as the General Election EV.  

Also- I assume a large % of "other race" voters tend to be young voters who are biracial/mixed-race... so if the under 34 voting % in down, that could partially explain why the Black vote is up, "other" voter is down, and white voter is the same.  But considering most non-white voters support Dems and majority of white voters support Republicans... the fact that white EV has remained the same as the general election, would seem to indicate that Dems may not have gained much of a real advantage regarding white/non-white EV%.

Yeah, I do think that the age gap if something to be concerned about as a Democrat, especially in a place like GA where younger people are significantly more likely to vote D than older people.

As for your point about the strong black numbers being a mirage, while that could be true and I get what you're saying, It does seem like black voters as an isolated group have strong turnout, even if other minority groups do not. There is a larger gap between majority black counties in the "black belt" and surrounding majority white counties in terms of % that have voted than there was in the GE. This could mean either that black voters really are having disproportionate turnout in these runoffs relative to white voters, or that the election day vote will be more white than it was in November (or something in between ofc). We saw that in the GE a few of these counties e-day vote was to the left of how the county overall, so I suspect that may flip.

However, even if you take the racial lens out of it, and calculate things based upon how they broke in the GE, it's really hard to find a way to get Democrats current vote lead under 8%. The real concern for me is that the e-day vote has potential to be even more lopsided towards the GOP as it was in the GE because more Democratic voters are voting early.

I'll second your point about the white vote, at least thus far, the white vote has skews from counties where the white vote likely voted to the left of the white vote statewide

[ProgressiveModerate]

https://twitter.com/Nate_Cohn/status/1343925218839056384

I have my doubts about it being a Biden + 20 electorate thus far, I could see the case for Biden + 14/15/16 but Biden + 20 seems like a stretch. If it really is anywhere close to that though, that's terrible news for the GOP as they will likely need to net 500k + votes. The early votes for Dems thus far are objectively better for Ds than they were in the general though, but that's not to say Perdue and Loeffler are underdogs.