Jump to content

2020 Texas Early Voting Numbers


StassneyHorn

Recommended Posts

17 minutes ago, BrickHorn said:

Huck - can you do a county-level regression between cumulative early voting and partisan split from 2016 or 2018? I get the sense that bluer counties are seeing higher early turnout.

Ran it for today's totals only, won't be doing it every day of course.

As of right now there's actually higher turnout for higher Cruz 2018 vote share. Galveston and Brazoria are 2 of the 3 highest turnout counties so far and went heavy for Cruz in 2018. On the other hand, the correlation is extremely weak. Adjusted R Square of 0.067 so we're talking statistically insignificant. And, of course, I'm only looking at 15 out of 254 counties.

Edited by Huckleberry
Link to comment
Share on other sites

2 hours ago, Huckleberry said:

image.png.c8474c35298d1d499d1f80787b980b70.png

image.thumb.png.4c61418fd2e1f28d9f89b5b2092ec1ff.png

 

 

 

I'll tell you, what stands out to me here is Galveston, Denton, Brazoria and Collin counties at 23, 21, 19, and 19%, respectively, while Montgomery County is bringing up the rear at >13%.

If I were to craft a narrative, it would be that there are a bunch of college-educated suburbanites (particularly women) who are very eager to vote against Donald Trump, presaging very poor results for him in Collin/Denton/Brazoria/Galveston while Montgomery is going to remain just as Republican as it ever was, but enthusiasm is way down.  But that's me pulling shit out of my ass. 

You would expect Montgomery to look a lot like the other suburban counties, and it just doesn't.  Montgomery County's turnout is lower than Hidalgo and El Paso counties, for chrissakes.  That's just odd.

And yeah--I'm sorry, but I'm fascinated by Montgomery County.  It's the King of the Red Counties.  We have plenty of counties that are going to go 75% for Trump.  But of those counties, Montgomery is by far the biggest.  It will provide more net votes to Trump than any other county, guaranteed.  So I really am watching it closely for turnout.

Fun question--which county is going to provide the most net votes for Biden.  In 2016, it was Dallas (just shy of 200,000 net votes for Clinton).  But this go around, I could see it being Harris or Travis.

 

Link to comment
Share on other sites

2 minutes ago, Ghost of LL said:

Fun question--which county is going to provide the most net votes for Biden.  In 2016, it was Dallas (just shy of 200,000 net votes for Clinton).  But this go around, I could see it being Harris or Travis.

Using turnout to date and 2018 Cruz/O'Rourke vote share (ignoring third parties), current net Biden votes:

HARRIS: 66,983
DALLAS: 70,930
TARRANT: 1,215
BEXAR: 31,866
TRAVIS: 70,542
COLLIN: -7,678
DENTON: -9,687
EL PASO: 35,054
FORT BEND: 7,954
HIDALGO: 22,639
WILLIAMSON: 2,105
MONTGOMERY: -21,867
GALVESTON: -10,634
BRAZORIA: -8,115
CAMERON: 8,540
Link to comment
Share on other sites

Was surprised to see that HRC only won bexar county by 13% in 2016. That is a real opportunity for improvement this time in terms of total votes and the actual margin. 

I assume that as turnout goes up, the margin will as well. Just need a full force effort to get San Antonio to the polls the same way Houston is.

Link to comment
Share on other sites

3 minutes ago, Huckleberry said:

Using turnout to date and 2018 Cruz/O'Rourke vote share (ignoring third parties), current net Biden votes:

HARRIS: 66,983
DALLAS: 70,930
TARRANT: 1,215
BEXAR: 31,866
TRAVIS: 70,542
COLLIN: -7,678
DENTON: -9,687
EL PASO: 35,054
FORT BEND: 7,954
HIDALGO: 22,639
WILLIAMSON: 2,105
MONTGOMERY: -21,867
GALVESTON: -10,634
BRAZORIA: -8,115
CAMERON: 8,540

For a net of roughly Biden +264.000?

I'd take that.

Link to comment
Share on other sites

You have to assume that you're going to get a negative net in the accumulation of all of the little redneck counties that will drag that overall number down from the big counties. No idea what the number is though. Huck would have to run every single county.

Edited by BradInATX
Link to comment
Share on other sites

1 minute ago, Ghost of LL said:

For a net of roughly Biden +264.000?

I'd take that.

Good head math, it's a hair under 260K.

So yes, if these 15 counties are voting with the exact same REP/DEM split that they did in 2018's Senate race, right now in those 15 counties only Biden has a 260K advantage after three days of early voting and including mail-in ballots through yesterday.

Link to comment
Share on other sites

10 minutes ago, BradInATX said:

You have to assume that you're going to get a negative net in the accumulation of all of the little redneck counties that will drag that overall number down from the big counties. No idea what the number is though.

In 2018 O'Rourke won those 15 counties 3,176,551-2,340,295 (57.1%-42.1%)

Total results were Cruz 4,260,553-4,045,632 (50.9%-48.3%)

So Cruz won the rest of the state 1,920,258-869,081 (68.3%-30.9%)

I think the target has to be running up a 1M vote lead in those 15 counties for Biden versus the 836,000 vote win O'Rourke took. That's if you want to think there's any chance for Biden in Texas. Then on top of that you have to hope Trump fatigue keeps his win total in the other 239 counties flat instead of growing by voters scared the Republicans might lose Texas.

Edited by Huckleberry
Link to comment
Share on other sites

2 minutes ago, Nice Guy Eddie said:

seems to me that it has to soon drop off somewhat significantly. We can't expect 5% of the voters to vote each day for 18 days in a row.

Huck’s graph seems to indicate the trend is steady participation throughout, with a Sunday dip due to less open hours and a procrastinator spike at the end. 

Edited by Pato del Muerto
Link to comment
Share on other sites

2 minutes ago, Huckleberry said:

In 2018 O'Rourke won those 15 counties 3,176,551-2,340,295 (57.1%-42.1%)

Total results were Cruz 4,260,553-4,045,632 (50.9%-48.3%)

So Cruz won the rest of the state 1,920,258-869,081 (68.3%-30.9%)

So he needs to be + a million or so in the large counties to have a chance, and he’s a quarter there after 3 days?

Link to comment
Share on other sites

3 minutes ago, Pato del Muerto said:

So he needs to be + a million or so in the large counties to have a chance, and he’s a quarter there after 3 days?

Ha, that's exactly the target I edited into the above post after you quoted it. I think so, that's where he wants to be. Obviously 1.25M would make things better for his chances.

Edited by Huckleberry
Link to comment
Share on other sites

6 minutes ago, Huckleberry said:

In 2018 O'Rourke won those 15 counties 3,176,551-2,340,295 (57.1%-42.1%)

Total results were Cruz 4,260,553-4,045,632 (50.9%-48.3%)

So Cruz won the rest of the state 1,920,258-869,081 (68.3%-30.9%)

I think the target has to be running up a 1M vote lead in those 15 counties for Biden versus the 836,000 vote win O'Rourke took. That's if you want to think there's any chance for Biden in Texas. Then on top of that you have to hope Trump fatigue keeps his win total in the other 239 counties flat instead of growing by voters scared the Republicans might lose Texas.

Can you  PM me or just post here a daily total of votes in those counties combined? The big ones?

Link to comment
Share on other sites

Just now, Huckleberry said:

That's what I'm doing.

Should be up in the 10-11am hour each day.

Where can I get a CSV of that bottom table? You have a link? And where is the old data by day coming from? I'm trying to put something together that shows a daily "ahead or behind" what a winning pace would look like.

Link to comment
Share on other sites

2 minutes ago, BradInATX said:

Where can I get a CSV of that bottom table? You have a link? And where is the old data by day coming from? I'm trying to put something together that shows a daily "ahead or behind" what a winning pace would look like.

I've been going to each of those county's websites each day, but after 3 days the State of Texas site is finally keeping pace:

https://earlyvoting.texas-election.com/Elections/getElectionDetails.do

Around 10-11 each morning that should be updated. For the first couple of days it was wildly off for some counties but now looks close enough for this kind of analysis. That would be awesome if you could track all 254 counties in a spreadsheet using that site and their 2018 vote splits. Or if you don't want to then I would end up doing it. Ha.

Edited by Huckleberry
Link to comment
Share on other sites

10 minutes ago, Pato del Muerto said:

Huck’s graph seems to indicate the trend is steady participation throughout, with a Sunday dip due to less open hours and a procrastinator spike at the end. 

Isn't there an extra week of voting this year that should impact the daily %s? I'm all for an 80-90% turnout but it just doesn't seem likely.

Link to comment
Share on other sites

1 minute ago, Nice Guy Eddie said:

Isn't there an extra week of voting this year that should impact the daily %s? I'm all for an 80-90% turnout but it just doesn't seem likely.

Yes, it will tail off next week I would think. Then we'll have the usual early voting spike right before it ends. I do think the extra 6 days will increase overall voter turnout, but certainly not by 24-30% (normal average of 4-5% of registered voters voting every available early day). Then you'll see a much smaller Election Day turnout.

Link to comment
Share on other sites

1 minute ago, Nice Guy Eddie said:

Isn't there an extra week of voting this year that should impact the daily %s? I'm all for an 80-90% turnout but it just doesn't seem likely.

Yes but the electorate knows that and presumably has spread their voting plans accordingly. Maybe a longer slow decay before the procrastinator spike starts as by definition that should start as late as possible no matter how long early voting runs. 
First few days are tracking well though. 

Link to comment
Share on other sites

Voted this morning in Killeen and it took an hour.  I arrived at the place right at 8 when it opened.  Easily 75%+ POC in line with me.  From what the lady in front of me said, this was a shorter wait than the day before when she came and tried to vote before having to leave to go to work.  

Link to comment
Share on other sites

2 minutes ago, Huckleberry said:

I've been going to each of those county's websites each day, but after 3 days the State of Texas site is finally keeping pace:

https://earlyvoting.texas-election.com/Elections/getElectionDetails.do

Around 10-11 each morning that should be updated. For the first couple of days it was wildly off for some counties but now looks close enough for this kind of analysis. That would be awesome if you could track all 254 counties in a spreadsheet using that site and their 2018 vote splits. Or if you don't want to then I would end up doing it. Ha.

I was going to do it way more roughly than that because I have about 30 minutes of this conference call before I have actual head-down metrics work to get back to that actually pays my bills. But you might be able to run with it. 

If we assume that the "other" counties stay flat (really we should account for population growth but I'm not doing that for hundreds of counties), we need the list of big counties to improve turnout by 25.7%. I think to be safe we need to shoot for 30% to account for growth and a little bit extra Trump enthusiasm.

That means 7,171,899.8 total votes in those big cities is where we need to be to have a chance - I'd be interested to see a running total of your big counties up toward a goal line of say, 7 million. If we get anywhere near that, with a little luck and a handful of Cruz->Biden voters, maybe it's realistic to think there's a shot.

Link to comment
Share on other sites

13 minutes ago, BradInATX said:

I was going to do it way more roughly than that because I have about 30 minutes of this conference call before I have actual head-down metrics work to get back to that actually pays my bills. But you might be able to run with it. 

If we assume that the "other" counties stay flat (really we should account for population growth but I'm not doing that for hundreds of counties), we need the list of big counties to improve turnout by 25.7%. I think to be safe we need to shoot for 30% to account for growth and a little bit extra Trump enthusiasm.

That means 7,171,899.8 total votes in those big cities is where we need to be to have a chance - I'd be interested to see a running total of your big counties up toward a goal line of say, 7 million. If we get anywhere near that, with a little luck and a handful of Cruz->Biden voters, maybe it's realistic to think there's a shot.

Don't think we need to account for population growth. We'll be using 2018 vote shares but the actual number of voters from 2020 so population growth is sort of baked in already.

And I've got all the data so I'll run it real quick right now.

Link to comment
Share on other sites

2 minutes ago, Huckleberry said:

Don't think we need to account for population growth. We'll be using 2018 vote shares but the actual number of voters from 2020 so population growth is sort of baked in already.

And I've got all the data so I'll run it real quick right now.

Right on.

On the bright side, Biden is outperforming Beto's polling from 2018 by a couple of points. If you bake those in, the numbers get a lot more manageable. If we theoretically give Biden the 2.2% that he is outperforming Beto's polls by, we'd only need to see about 10% increase in big-county voting (about 6 million total) to win, assuming small county turnout is flat.
 

I gotta get to it but I might mess around with this some more later.

Edited by BradInATX
Link to comment
Share on other sites

Using early voting turnout to date, and assuming that every county votes in the same ratio that they did in 2018 (ignoring 3rd party votes), the current total in Texas is:

Trump 1,350,121 (51.3%)
Biden 1,279,232 (48.7%)

You'll note that is a 2.6% margin. The 2018 Senate race was a 2.6% margin. Voter turnout location does not appear like it will be anything close to a deciding factor. Demographic shift, targeted voter turnout, and the candidates themselves would have to break Biden's way to get it done.

Edited by Huckleberry
Link to comment
Share on other sites

Just now, BradInATX said:

Also the election day splits (is election day redder?)
Too much for me to mess with right now

Right but this extremely simple analysis isn't affected by that. I'm assuming that the total votes in each county for the entire election split the same they did in 2018. If the Dem votes are tilted to coming in earlier and the Rep votes are tilted to later then in the end it doesn't matter. The problem with this analysis is that if Dems are more motivated than Reps we won't know until Election Day turnout is known. Because the conventional wisdom is that you're correct that this year Election Day will be more Rep whereas in previous years it was more Dem.

Link to comment
Share on other sites



×
×
  • Create New...