Early Voting Tracker- Elections Lab

[Huckleberry 32147]

3 hours ago, scottsins said:

Has this been addressed yet by anyone on the “other side” of this analysis? I read it when I woke up to take a piss, and it’s making it difficult for me to go back to sleep now.

It's in the we don't know shit about fuck pile. The 2020 and 2024 elections are going to have different early versus election day voting profiles for the one big obvious reason. Before COVID, Democrats voted on election day and Republicans voted early (generally speaking those were their preferences). How much are we moving back toward that in 2024? Nobody knows.

I will say that the movement isn't huge but the more rural counties out of the top 15 counties in Texas are losing steam faster than the urban ones the last few days. Everyone is, but there used to be 4 counties ahead of their compressed 2020 EV rate and 3 of them were Trump counties in 2020. Now there is only one.

[Bozo_Casanova 25522]

11 hours ago, sushihorn said:

I think cultural attitudes towards marriage are a major reason for the rural-urban contrast. 

Highly correlated, sure. I'm skeptical about "major reason".  My hypothesis is that the partisanship of married people is an artifact of rural populations being older.

[bluto 4800]

…The state will provide four additional absentee return reports (Friday, Monday, Tuesday x2), and the trend remains the same.

Both parties are in line to return ballots at nearly the same rate, and with the surge of Republican requests down the stretch, a revision from a Dem +480k expected lead in absentee returns must be revised down to Dem +430k.

Absentee requests are now closed (except for in-person absentee ballots cast in Bucks County), and the final Pennsylvania electorate is tracking toward an environment of D+1.5 - R+2.5.

Slightly larger bands because so much of the vote comes on election day and presents more possibilities. 

[bluto 4800]

That same guy Pruser is modeling az/nv to be steep and daunting hills for dems. 

[Valmy77 12519]

Just bad news everywhere huh? Damn.

[Js1 82402]

I guess we should just quit and let it happen. 

[Huckleberry 32147]

[aggie08 20215]

This is decidedly NOT the good vibes thread.

[Js1 82402]

5 minutes ago, aggie08 said:

This is decidedly NOT the good vibes thread.

Far too much baselining to 2020.  STOP DOING THAT

[Nice Guy Eddie 13276]

The low Texas enthusiasm always tells me how ripe Texas is for a Democrat wave to take the state. However there is a problem that loud Dem enthusiasm always encourage GOP get out the vote. Beto scared many casual GOP voters to show up to vote for Cruz in 2018. Allred is probably running into that as well. 

[SydneyCarton 47100]

I can't even open this thread anymore. 

[immamac 41360]

highest turnout in recent history (possibly ever) in Texas for EV and people think that's a bad thing lol.

[Huckleberry 32147]

Yeah I'm not sure if mine is being interpreted as bad news but it isn't. In fact the last few days have seen more of a slowdown the redder the county is. This very well may be the election day vote cannibalization that has been discussed.

Edited 11 hours ago by Huckleberry

[Js1 82402]

2 minutes ago, Huckleberry said:

Yeah I'm not sure if mine is being interpreted as bad news but it isn't. I'm fact the last few days have seen more of a slowdown the redder the county is. This very well may be the election day vote cannibilazation that has been discussed.

Also that 2020 EV was 3 weeks long.  Texas is doing quite well, considering one less week of early voting. 

[sushihorn 3215]

12 hours ago, StassneyHorn said:

What states ask for relationship status? What states update this? Are they cross referencing marriage and divorce records?

I’m sure it’s that at the minimum. Google profiles are incredibly detailed and they are hardly the only one out there. I’d bet the database at Target Smart starts with an anonymized version sliced by precinct to match with SoS data released. 

[StassneyHorn 6410]

I'm trying to tell you TS is ass my dude

[Dahobbs 17095]

58 minutes ago, sushihorn said:

I’m sure it’s that at the minimum. Google profiles are incredibly detailed and they are hardly the only one out there. I’d bet the database at Target Smart starts with an anonymized version sliced by precinct to match with SoS data released. 

Assuming that Target Smart has correct demographic information, which I most certainly question, comparison to 2020 is really misleading because of clear change in early voting patterns due to the pandemic. 

A comparison to 2022 is in my mind more interesting as it shows a greater voter share for Asian/Pacific Islander, Black, and Hispanic voters and a lower White share (66.6% in 2024 versus 70.1% in 2022 for early vote 6 days out). The early vote share for whites in 2022 decreased over the last 6 days to 68.5%. And on election date that percentage shrank further to 66.9%. 

Another interesting data point, in 2020 absentee ballots favored Democrats (which was a change from what had previously occurred). According to TargetEarly, only 19.5% (332,000 of 1,699,000) of the those white absentee voters from 2020 have voted early so far. On the other hand, 46% (377,000 of 813,000) of the white election day voters from 2020 have already voted. Assuming that election day vote favored Trump, this may be additional evidence that Trump is cannibalizing election day voters in the early vote. It would also suggest that Harris has a lot more votes still available. The question there is whether this shows reduced enthusiasm (I doubt that) or a shift back to election day voting (more likely in my opinion). Oh, and 5.6% (248,000) of the white early vote so far was unregistered in 2020. And 4.5% (197,000) were registered in 2020 but didn't vote in 2020 (or apparently 2022). 

 

 

Edited 10 hours ago by Dahobbs

[alphahydro 332]

fascinating stuff.  

obviously super anecdotal, but I voted by mail/early for the first time this year.

only reason - waiting in line in 2020 was kind of a pain in the ass. 

i was probably more "enthusiastic" to vote in 2020 - willing to wait in line to make sure my vote wasn't lost in the mail (even though I live in a non-swing state). 

anyway, point is - very hard to read into early voting numbers.  if everyone was like me, a surge in early voting numbers would actually signal a drop in enthusiasm. 

and, i think, for some demographics, that might be the case.  and in other demographics, a surge in early voting signals the opposite. 

bottom line for me is we def don't know shit about fuck. 

my gut feeling/intuition about the election is almost certainly wrong.  i'm surrounded by lots of white, college educated trump 2020 voters, and I can tell you that the enthusiasm among them has dropped considerably.  but will that be a prominent factor in this election?  idk. 

 

[WBT 20132]

What is the point of this thread again?  Is there anything showing EV/election day voting split by party to be predictive election to election?

[immamac 41360]

13 minutes ago, WBT said:

What is the point of this thread again?  Is there anything showing EV/election day voting split by party to be predictive election to election?

it's just tracking the EV from various different sources not just Texas but nationwide. In the end it doesn't make a shit because it's not true vote counts it's just vote volume and no one knows how any demo is voting this election, even registered republicans are a complete fucking mystery. 

[aggie08 20215]

14 minutes ago, WBT said:

What is the point of this thread again?

Misery 

[alphahydro 332]

anyway, just my thoughts -

each of us have some insight into the dynamics of a sliver of the electorate.  but it's hard to say which slice is going to be the deciding factor in 2024.  if we knew today which demographic was going to be key in 2024 - and if we could parse the EV data by demographic, we could prolly draw some better conclusions. 

for me, if i had to guess, in my county, trump's numbers will be lower than they were in 2020 due to 

• drop in enthusiasm among college educated whites
• increase in enthusiasm among women
• aging out of former elderly trump supporters

and each of these trends are only important on the margins.  if i could parse the EV data for my county by these demographics, i could prolly test my guess a bit. 

but i have no idea what's going on in the important swing counties.  might come down to enthusiasm among evangelicals - which i have no insight into. 

reminds me of 2016.  Jake Sullivan, who grew up in Minn., then a young upstart in Hillary's close circle, was begging Hillary to campaign in the upper midwest.  

He was def telling friends that she was gonna lose the election because the polling was wrong in teh upper midwest.  everyone else in her inner circle thought he was wrong.  RIP. 

he had some insight into the dynamics there that the other people in her close circle didn't, despite having tons of data and experience, etc. 

https://www.politico.com/magazine/story/2016/12/2016-presidential-election-10-moments-trump-clinton-214508/

Quote

Jake Sullivan, Clinton’s policy director—a brainy and nervous former State Department aide who took on an increasingly important political role as the campaign ground on—was the only one in Clinton’s inner circle who kept saying she would likely lose, despite the sanguine polling, according to his friends. He was also the only one of the dozen aides who dialed in for Clinton’s daily scheduling call who kept on asking if it wasn’t a good idea for her to spend more time in the Midwestern swing states in the closing days of the campaign.

 

[StassneyHorn 6410]

18 minutes ago, WBT said:

What is the point of this thread again?  Is there anything showing EV/election day voting split by party to be predictive election to election?

Women voters are literally murdering men in the important swing states and PA's HUGE early vote bank by party (although information could be slightly dated due to changed changing political party)

 

Some people started talking about models. First person to show any kind of math at all, bonus points if written in Python, will win 6 month subscription to get their own Google Data Analytics certification. As site admin I can make this happen.

[aggie08 20215]

1 minute ago, immamac said:

even registered republicans are a complete fucking mystery.

To a point. I think even as 8% defection would be amazing (since every defection essentially counts as two votes), but registered Republicans flocking to the polls at a ~91% Trump vote clip is still bad news.

[Pato del Muerto 51914]

My takeaway is that you can’t accurately model a novel event, which this election seems like it might be. But you won’t really know until the election is over how novel it ends up being. 

Edited 8 hours ago by Pato del Muerto
You can’t really dust for vomit

[SydneyCarton 47100]

1 hour ago, WBT said:

What is the point of this thread again?  Is there anything showing EV/election day voting split by party to be predictive election to election?

 

1 hour ago, aggie08 said:

Misery 

It's defiinitely miserable, but it is effectively keeping the hopeium and the doom effectively separate. 

[gsoda3 3187]

3 hours ago, immamac said:

highest turnout in recent history (possibly ever) in Texas for EV and people think that's a bad thing lol.

Last night KXAN said there were 1mm more Rs than Ds in Texas who early voted.  What's the expected number?

[Js1 82402]

1 minute ago, gsoda3 said:

Last night KXAN said there were 1mm more Rs than Ds in Texas who early voted.  What's the expected number?

That can't be 100% quantified.  We don't have party registration.  And looking back at past primary participation is dumb - most of the time in TX, the GOP primary is the general election.  Voters in TX just pick one to vote in at primary season - no registration, nada.  Dems and indies vote in GOP primary in TX all the time. 

Edited 7 hours ago by Js1

[Huckleberry 32147]

5 minutes ago, gsoda3 said:

Last night KXAN said there were 1mm more Rs than Ds in Texas who early voted.  What's the expected number?

@bolverk's data would be better here but seems unlikely. I have estimated 2.54mm to 2.10mm in favor of Ds from the top 15 counties. Using 2020 vote splits this would estimate a .56mm lead for R right now. 53.9% to 46.1% out of votes cast for one of the two. Which seems a lot more reasonable. 

[Pescado_Rojo 19194]

13 minutes ago, Js1 said:

That can't be 100% quantified.  We don't have party registration.  And looking back at past primary participation is dumb - most of the time in TX, the GOP primary is the general election.  Voters in TX just pick one to vote in at primary season - no registration, nada.  Dems and indies vote in GOP primary in TX all the time. 

This. I vote in the R primary a lot in mid-term years because in many of the county and judicial elections it may as well be the general. A lot of the races in the general, the R winner is unopposed. 

[Js1 82402]

2 minutes ago, Pescado_Rojo said:

This. I vote in the R primary a lot in mid-term years because in many of the county and judicial elections it may as well be the general. A lot of the races in the general, the R winner is unopposed. 

I know a lot of Dems who voted this year against Trump by voting for Haley. 

[Pescado_Rojo 19194]

1 minute ago, Js1 said:

I know a lot of Dems who voted this year against Trump by voting for Haley. 

I voted in the D primary so I could vote for Roland Gutierrez. 

[sushihorn 3215]

4 hours ago, Js1 said:

Far too much baselining to 2020.  STOP DOING THAT

I don’t see a better baseline, unfortunately. Midterms are different and very few states had large EV in 2016 or earlier. That’s why I’m starting to compare to 2020 final vote instead of EV. 

[Js1 82402]

3 minutes ago, sushihorn said:

I don’t see a better baseline, unfortunately. Midterms are different and very few states had large EV in 2016 or earlier. That’s why I’m starting to compare to 2020 final vote instead of EV. 

Now that is fine. Don’t compare 2024 EV with 2020 EV but I’d say that the type of voting will be similar to 2022 EV - far less mail, more GOP voting early, more Dems voting on Tuesday 

Remember that Dems lost the early vote in NV in 2022 but Masto won the senate race 

Edited 7 hours ago by Js1

[Red Five 40114]

43 minutes ago, Js1 said:

I know a lot of Dems who voted this year against Trump by voting for Haley. 

(raises hand)

[sushihorn 3215]

1 hour ago, Js1 said:

Now that is fine. Don’t compare 2024 EV with 2020 EV but I’d say that the type of voting will be similar to 2022 EV - far less mail, more GOP voting early, more Dems voting on Tuesday.

We can pretty much rule out Ken Paxton winning by double digits this year. 

[Js1 82402]

7 minutes ago, sushihorn said:

We can pretty much rule out Ken Paxton winning by double digits this year. 

Well that’s one bright spot that can’t be dimmed 

[bolverk 28890]

2 hours ago, Huckleberry said:

@bolverk's data would be better here but seems unlikely. I have estimated 2.54mm to 2.10mm in favor of Ds from the top 15 counties. Using 2020 vote splits this would estimate a .56mm lead for R right now. 53.9% to 46.1% out of votes cast for one of the two. Which seems a lot more reasonable. 

edited to tag @gsoda3

As of yesterday, a little over 7.5 million Texans had voted.

• If we assume each county has the same partisan tilt today as it did in 2020, Kamala has 45.6% of the vote so far (3,425,767).
• If we go by my calculated guestimate for each county's evolving partisan shift since 2012, Kamala has 47.8% of the vote (3,591,567).
• If we go by Huckleberry's alteration that "dampens" the partisan shift, Kamala has 46.6% of the vote (3,501,382).

Right now, based on the partisan swing in each county and according to their turnout, Harris likely won't win Texas ranging from 45.6% to 47.8% of the vote. In 2020, Biden won 46.5%.

However, her odds are improving as more EV data come in. For example, Huckleberry's formula had Kamala winning 46.3%, while my formula had her with a 47.5% chance yesterday.

Below is the revised map of EV turnout. As of yesterday, 40.35% of registered Texans had voted. The overall pattern hasn't changed, but when I looked at the updated figures, it appears like some counties are plateauing and others are catching up. That applies to almost everywhere except amongst the worst laggards, who are falling further behind.

 

Edited 5 hours ago by bolverk

[quigley 1147]

4 hours ago, aggie08 said:

To a point. I think even as 8% defection would be amazing (since every defection essentially counts as two votes), but registered Republicans flocking to the polls at a ~91% Trump vote clip is still bad news.

Goal for Harris is > 6% flip of 2020 Trump voters. There was apparently internal Dem data that showed 9% flip if I recall (heard on podcast). If Harris hits 9% across the swing states, she wins easily per my recollection of the conversation.

[Dahobbs 17095]

54 minutes ago, bolverk said:

edited to tag @gsoda3

As of yesterday, a little over 7.5 million Texans had voted.

• If we assume each county has the same partisan tilt today as it did in 2020, Kamala has 45.6% of the vote so far (3,425,767).
• If we go by my calculated guestimate for each county's evolving partisan shift since 2012, Kamala has 47.8% of the vote (3,591,567).
• If we go by Huckleberry's alteration that "dampens" the partisan shift, Kamala has 46.6% of the vote (3,501,382).

Right now, based on the partisan swing in each county and according to their turnout, Harris likely won't win Texas ranging from 45.6% to 47.8% of the vote. In 2020, Biden won 46.5%.

However, her odds are improving as more EV data come in. For example, Huckleberry's formula had Kamala winning 46.3%, while my formula had her with a 47.5% chance yesterday.

Below is the revised map of EV turnout. As of yesterday, 40.35% of registered Texans had voted. The overall pattern hasn't changed, but when I looked at the updated figures, it appears like some counties are plateauing and others are catching up. That applies to almost everywhere except amongst the worst laggards, who are falling further behind.

 

Per above, Dems should have a lot more votes left than GOP. If they show up, it gets real tight. 

[Valmy77 12519]

Some of those numbers are so weird. Like why would Hartley County have everybody voting early but Dallam is like...no fuck that.

[bolverk 28890]

10 minutes ago, Valmy77 said:

Some of those numbers are so weird. Like why would Hartley County have everybody voting early but Dallam is like...no fuck that.

Dalhart sits directly on the county line (hence the name). The more Anglo and wealthier side of town is in Hartley County (30% Hispanic), while the poorer, more Hispanic part of town is in Dallam County (52% Hispanic).

Edit: You can see the same patterns with Amarillo (Potter & Randall Cos.) and Midland/Odessa (Midland & Ector Cos.). This is why Texas desperately needs a Stacey Abrams-type who can credibly lead and organize our vast Hispanic population.

Edited 4 hours ago by bolverk

[bolverk 28890]

8 minutes ago, Dahobbs said:

Per above, Dems should have a lot more votes left than GOP. If they show up, it gets real tight. 

The rural areas jumped to a huge lead in Early Votes and are slowing down in gains, but it looks to me like the suburbs are still peaking.

[WBT 20132]

1 hour ago, quigley said:

If Harris hits 9% across the swing states, she wins easily per my recollection of the conversation.

Well no shit.  I find 9% incredibly hard to believe though.

[Huckleberry 32147]

Depends on what we are talking about. I think 9% of registered Republicans who vote is feasible. 9% of people who currently consider themselves Republicans is less likely. 

Edited 3 hours ago by Huckleberry

[ShaggyBevo RIP 3827]

Can’t tell how much of thread is Texas bs national vote.

Surprised that people think it close nationally. Harris is going to take 60% of white female voters.

The interesting thing to me is the 9-10 House seats in Texas’ major metropolitan area that could flip by the female vote.

[quigley 1147]

35 minutes ago, WBT said:

Well no shit.  I find 9% incredibly hard to believe though.

I'm just quoting what they said. 

I'm sorry if it's obvious to you that 9% is the threshold of comfortable win. I wasn't to me. Also, I have no idea why you find it hard to believe. They have internal polling showing this.

One thing that's absolutely certain is that we don't know shit about fuck. Public polling is a disaster, and early voting is reading shadows. I'm prepared to believe anything.

[quigley 1147]

39 minutes ago, Huckleberry said:

Depends on what we are talking about. I think 9% of registered Republicans who vote is feasible. 9% of people who currently consider themselves Republicans is less likely. 

This is basically right. They were talking about 9% of 2020 Trump voters.

[quigley 1147]

58 minutes ago, bolverk said:

Dalhart sits directly on the county line (hence the name). The more Anglo and wealthier side of town is in Hartley County (30% Hispanic), while the poorer, more Hispanic part of town is in Dallam County (52% Hispanic).

Edit: You can see the same patterns with Amarillo (Potter & Randall Cos.) and Midland/Odessa (Midland & Ector Cos.). This is why Texas and FLORIDA desperately need a Stacey Abrams-type who can credibly lead and organize our vast Hispanic population.

FIFY

[StassneyHorn 6410]

4 hours ago, sushihorn said:

We can pretty much rule out Ken Paxton winning by double digits this year. 

Well yea he’s not up for a vote.

One of the things I always love to do is go from top to bottom on who cancels out who, raw numbers. So Harris county cancels the bottom 203, Harris and Dallas raw cancels out bottom 222 and on and on. It’s usually been top 7-8 cancel out the rest of the state.

So knowing that, what do the numbers need to look like in the top 7-8 counties is a question I think naturally comes up. I only look at early voting and haven’t included Election Day numbers, I don’t know when the top 6 counties outvotes and went D.Should we find a predictor at top 5 counties or make it a list with counties over 200k? Things to consider

Until there is a successful breakthrough I haven’t thought of good “data” questions to look for/ask for. I think there’s a lot of focus on top 15 counties cause that’s a round number/inclusive number etc. There’s discussion of the “Texas spine” that goes SA to Austin to Metroplex” and then we just add Houston (I don’t have the heart to acknowledge ElPaso since they choose not to participate).