Jump to content

Early Voting 2018 thread of doom


Pancho

Recommended Posts

10 minutes ago, Nice Guy Eddie said:

I think we're in unknown territory with whether this is more like 2014 or 2016.  Obviously the numbers will be much higher than 2014 but it's not as high as 2016 either.  Still it's fun to look into the numbers. 

I was listening to Nate Silver (538) talk about early voting and its affect on his election models.  He said that he gives them 0 weight when he aggregates polls.  He said the one difference that early voting makes is that if someone tells a pollster that they voted early for X, that person will be given more weight than just labeling them a "likely voter."  but overall he thinks early voting doesn't necessary demonstrate an advantage of one party over another.   He said that he would take bets if someone can take the early voting numbers and predict different results from his models.

 

The EV turnout so far is 11% smaller than 2016 and more than 100% larger than 2014.  I can't think of any historical model to use for the composition of the electorate other than 2016.  Without an Obama at the top of the ticket, 2008 and 2012 models don't work.  Black and youth turnout reverted to historic norms when he wasn't on the ticket even while he was in office (2010 and 2014).  And they remained in the historic normal range in 2016 also.  It doesn't leave many models to choose from.

Link to comment
Share on other sites

Between 2016 and 2018 many of the largest Texas counties have 5% more registered voters. Due to death snd moving another 5% have been replaced, so there are more than 10% new voters. If you go back to 2014 make it 20% or more people who will vote in 2018 who were not a county resident or not registered in 2014.

Any polling model that does not account for the demographics related to 20% of the voters will be flawed.

  • Like 1
Link to comment
Share on other sites

17 minutes ago, Celery Man said:

I think Nate’s overarching point was that polls already account for it and people use it to see what they want to see. Also I think he’s grouchy about laypeople using high turnouts to declare polls and models meaningless or flawed.

The internals and crosstabs tell me that some polls are flawed.  And a lot of experts are insulted by the very idea of being questioned by laymen and just dismiss questions.  I feel very confident the data going into his models is flawed to a significant degree.  So the models could be perfect and the output off considerably.

Link to comment
Share on other sites

Hillary is not running and places like Austin are changing fairly rapidly.  Texas as a whole has not only seen whites drop below 45% of the population in the past 15 years, but seen a large influx of folks from more liberal states.  

I think a lot of pollsters are going to get several races wrong, or will at least be way off on the numbers.  

Link to comment
Share on other sites

3 hours ago, Celery Man said:

I think Nate’s overarching point was that polls already account for it and people use it to see what they want to see. Also I think he’s grouchy about laypeople using high turnouts to declare polls and models meaningless or flawed.

If they use Republican and Democratic turnout percentages from 2014 or 2016 they are highly flawed, and a lot of polls do.

Link to comment
Share on other sites

1 hour ago, Js1 said:

34k on a Sunday in Harris County isn’t bad. Sunday is traditionally the slowest EV day of the cycle.  Already over half a million voters 

I think we solved the Harris County mystery.  I spoke with a Houston teacher who said she hadn’t voted because she can’t make the polls. Turns out for the first week that have 8-4:30 voting.  Inflexible job means you can’t get there.  Second week is 7-7.  Should see a big bump in Harris this week because of that. 

 

They anticipated the demand in 2016 and had polls open late the whole EV period   Thus the unfavorable comparison to 2016 in the charts.

 

article about it:

https://www.google.com/amp/s/www.chron.com/news/houston-texas/houston/amp/Harris-County-continues-early-voting-surge-13331159.php

  • Like 1
Link to comment
Share on other sites

Do you guys think this huge spike in statewide turnout will bring any tangible benefits to the Dems?

fletcher vs culberson = Harris county ?  Win?

Allred vs sessions = Dallas county ?  Win?

jones vs Hurd= Repubs are spending there again. Competitive now?

getting even one seat in Texas would boost the Dems chances of taking the house by a bit.

Edited by Voldemort86
Link to comment
Share on other sites

2 hours ago, Voldemort86 said:

Do you guys think this huge spike in statewide turnout will bring any tangible benefits to the Dems?

fletcher vs culberson = Harris county ?  Win?

Allred vs sessions = Dallas county ?  Win?

jones vs Hurd= Repubs are spending there again. Competitive now?

getting even one seat in Texas would boost the Dems chances of taking the house by a bit.

Given this turnout Beto is still in good shape, although about 3 points or so worse than before the Kavanaugh hearings. I think he will be up by enough in the 26 largest counties, to hold off the rural vote. Hillary won the big 26 counties by 262,000, and Beto should win them by 850,000 plus, based on demographics and current turnout. Of the 12 largest counties Democrats should win 9, the Republicans 2, with Collin a toss-up. By 2020 Democrats should win 11 of the 12 as long as they have a decent candidate.

The congressional districts are too hard to call with county data alone because of the way they are gerrymandered, except MJ Hegar looks good since her district is Williamson and Bell counties and Wilco is way ahead of Bell in voter turnout. If MJ wins I expect a few other Democrats will win congressional seats in Texas, but I have not taken the time to break down the districts.

 

  • Like 1
Link to comment
Share on other sites

38 minutes ago, RayDog said:

Given this turnout Beto is still in good shape, although about 3 points or so worse than before the Kavanaugh hearings. I think he will be up by enough in the 26 largest counties, to hold off the rural vote. Hillary won the big 26 counties by 262,000, and Beto should win them by 850,000 plus, based on demographics and current turnout. Of the 12 largest counties Democrats should win 9, the Republicans 2, with Collin a toss-up. By 2020 Democrats should win 11 of the 12 as long as they have a decent candidate.

The congressional districts are too hard to call with county data alone because of the way they are gerrymandered, except MJ Hegar looks good since her district is Williamson and Bell counties and Wilco is way ahead of Bell in voter turnout. If MJ wins I expect a few other Democrats will win congressional seats in Texas, but I have not taken the time to break down the districts.

 

Which 2 counties of the biggest 12 do you expect Beto not to win?

Edited by Texas73
Link to comment
Share on other sites

8 minutes ago, Texas73 said:

Which 2 counties of the biggest 12 do you expect Beto not to win?

Hillary didn’t do well in Tarrant, Collin or Denton.  

I have to think Beto love/Cruz hate is hurting Culberson the most.  His district went Hillary in 2016 even though he won easily in 2016. It appears to be a coin flip now.  With every voter that Beto pushes towards the poll in that district, it would seem they would more likely vote for fletcher over culberson.  

I’m sure Betos GOTV effort will result in some gop votes, you would have to guess they’re knocking on doors of people either who have expressed an interest in voting for him or at least reside in a heavy Dem area.

hurd would seem like the perfect candidate to lose in this midterm but the few polls show him as relatively safe.

edit: js1 is right.  Beto ain’t winning Montgomery. I Forgot they were in the top 12.

Edited by Nice Guy Eddie
Link to comment
Share on other sites

2 minutes ago, Texas73 said:

I can't believe Hurd keeps winning. Being an incumbent must be big factor.

I'm not buying anything on Hurd until the results comes in.  It's a hard to poll district and Siena/NYT polled it before Latinos seemed to engage in the last few weeks (based on their now polling)

Link to comment
Share on other sites

Just now, Hugo Stiglitz said:

My uncle is a Hurd voter.  He checks a lot of traditional conservative boxes. 

Hurd looks to be his own guy in Congress unlike many.  He’s probably stood up to trump more than any republican in Congress especially on the Wall topic. Surprising that trump has never once pushed back but then again maybe trump listened to his aides that he should leave Hurd alone if they want to keep the seat.

Link to comment
Share on other sites

1 minute ago, Nice Guy Eddie said:

Hurd looks to be his own guy in Congress unlike many.  He’s probably stood up to trump more than any republican in Congress especially on the Wall topic. Surprising that trump has never once pushed back but then again maybe trump listened to his aides that he should leave Hurd alone if they want to keep the seat.

Hurd votes with Trump 95.7% of the time.

Don't make shit up.  He's the Jeff Flake of the House.  "Oh I'm so very concerned about Trump's rhetoric. Blah blah blah" *votes with Trump*

His 2 votes against the AHCA (that was going to pass regardless) and Trump's immigration bill doesn't make him his own guy.

Link to comment
Share on other sites

22 minutes ago, Nice Guy Eddie said:

Hillary didn’t do well in Tarrant, Collin or Denton.  

I have to think Beto love/Cruz hate is hurting Culberson the most.  His district went Hillary in 2016 even though he won easily in 2016. It appears to be a coin flip now.  With every voter that Beto pushes towards the poll in that district, it would seem they would more likely vote for fletcher over culberson.  

I’m sure Betos GOTV effort will result in some gop votes, you would have to guess they’re knocking on doors of people either who have expressed an interest in voting for him or at least reside in a heavy Dem area.

hurd would seem like the perfect candidate to lose in this midterm but the few polls show him as relatively safe.

edit: js1 is right.  Beto ain’t winning Montgomery. I Forgot they were in the top 12.

yeah beto doesn't win montgomery, would be great if he peeled off some votes there though.

I haven't done any door knocking recently, but I think the plan is supposed to be at this point they're reconnecting/establishing voting plans with people who have indicated that they would or might vote for Beto.

Link to comment
Share on other sites

40 minutes ago, Js1 said:

Hurd votes with Trump 95.7% of the time.

Don't make shit up.  He's the Jeff Flake of the House.  "Oh I'm so very concerned about Trump's rhetoric. Blah blah blah" *votes with Trump*

His 2 votes against the AHCA (that was going to pass regardless) and Trump's immigration bill doesn't make him his own guy.

Show me any congressman that doesn’t vote with his party’s president >90%?  I will help you. That person doesn’t exist. 

Link to comment
Share on other sites

1 minute ago, Nice Guy Eddie said:

Show me any congressman that doesn’t vote with his party’s president >90%?  I will help you. That person doesn’t exist. 

Walter Jones (R- NC) - 49.3%%

Justin Amash (R - MI) - 54.8%

Thomas Massie (R-KY) - 62.4%

  • Like 1
Link to comment
Share on other sites

As Js said:  We won't know about Hurd until about 10 PM CST on Nov. 6.  That district is one of the true toss up district's in the whole nation.  Neither the Dems nor the GOP have won there since it was court redrawn by more than 3 points.  Now Hurd is up 15 points?  That just defies any statistical evidence.  I mean it's possible, but it would be an extreme tail event that heralds a fundamental politically and demographic shift.

Link to comment
Share on other sites

2 minutes ago, Bateshorn said:

As Js said:  We won't know about Hurd until about 10 PM CST on Nov. 6.  That district is one of the true toss up district's in the whole nation.  Neither the Dems nor the GOP have won there since it was court redrawn by more than 3 points.  Now Hurd is up 15 points?  That just defies any statistical evidence.  I mean it's possible, but it would be an extreme tail event that heralds a fundamental politically and demographic shift.

You know when they talk about certain races that look likely/safe and then "unexpectedly" flip in a wave year?

TX-23 is exactly that kind of race.  Only a few polls here and there, extremely difficult to poll, will hinge solely on Hispanic turnout, etc.

Link to comment
Share on other sites

5 hours ago, Voldemort86 said:

Do you guys think this huge spike in statewide turnout will bring any tangible benefits to the Dems?

fletcher vs culberson = Harris county ?  Win?

Allred vs sessions = Dallas county ?  Win?

jones vs Hurd= Repubs are spending there again. Competitive now?

getting even one seat in Texas would boost the Dems chances of taking the house by a bit.

We are still well below the 2016 turnout levels.  Unless we get above that, I expect most of the above normal turnout to be people who voted in 2016 but who normally wouldn't vote in a mid-year election.

Link to comment
Share on other sites

1 minute ago, sushihorn said:

We are still well below the 2016 turnout levels.  Unless we get above that, I expect most of the above normal turnout to be people who voted in 2016 but who normally wouldn't vote in a mid-year election.

We are seeing reports of sizable numbers voting with no voting history.

Link to comment
Share on other sites

4 minutes ago, SuingToGetAMessageBoard? said:

We are seeing reports of sizable numbers voting with no voting history.

Yeah, the big counties were showing an average of 7 to 8% with no previous history. That exceeds population growth since 2016 for those counties by quite a bit. 

Link to comment
Share on other sites

5 minutes ago, SuingToGetAMessageBoard? said:

We are seeing reports of sizable numbers voting with no voting history.

this. I have had more conversations with people who have never voted in a midterm (or even early) before because they have had questions about the process and such. I've never had conversations outside my close friend group about these types of elections before. It is definitely a different demographic.

Link to comment
Share on other sites

If Beto actually outright won each of the top 12 counties save for Denton and Montgomery it would be over. That group includes Tarrant, Williamson (both R+9, similar to the state as a whole) and Collin Country at R+16. If he won Collin, Denton (R+20) which is similar demographically and has UNT, would be very close to a draw. Montgomery will still be R plus infinity but that isn't enough at that point. That would eliminate a big portion of the structural Republican lead, and allow the big urban counties to easily out vote the rest.

I don't have a model or anything but from anecdotal evidence that seems wildly optimistic. I do hope y'all are right though.

Sent from my SM-G920V using Tapatalk

Link to comment
Share on other sites

4 minutes ago, gmr548 said:

If Beto actually outright won each of the top 12 counties save for Denton and Montgomery it would be over. That group includes Tarrant, Williamson (both R+9, similar to the state as a whole) and Collin Country at R+16. If he won Collin, Denton (R+20) which is similar demographically and has UNT, would be very close to a draw. Montgomery will still be R plus infinity but that isn't enough at that point. That would eliminate a big portion of the structural Republican lead, and allow the big urban counties to easily out vote the rest.

I don't have a model or anything but from anecdotal evidence that seems wildly optimistic. I do hope y'all are right though.

Sent from my SM-G920V using Tapatalk
 

I am a little worried about how much Harris is lagging.  I posted something the other day saying how they shorten early voting hours in week 1 and expand them in week 2, so it might go up, but it would be nice to already have banked those votes.

Edited by SuingToGetAMessageBoard?
Link to comment
Share on other sites

9 minutes ago, SuingToGetAMessageBoard? said:

I am a little worried about how much Harris is lagging.  I posted something the other day saying how they shorten early voting hours in week 1 and expand them in week 2, so it might go up, but it would be nice to already have banked those votes. 

Harris halfway to a million voters already with the shortened EV hours already and had 34k turnout on the slowest day of the cycle.  I think it'll ramp up this week.  A lot of people complained about lines last week can now go before/after work much more easily.

Link to comment
Share on other sites

18 minutes ago, Dahobbs said:

Yeah, the big counties were showing an average of 7 to 8% with no previous history. That exceeds population growth since 2016 for those counties by quite a bit. 

Not sure how meaningful that is.  Even if the population were completely static and had 100% turnout every year, you'd still have 4% with no voting history every cycle from 2 years of people turning 18. 

Link to comment
Share on other sites

3 minutes ago, sushihorn said:

Not sure how meaningful that is.  Even if the population were completely static and had 100% turnout every year, you'd still have 4% with no voting history every cycle from 2 years of people turning 18. 

not sure your math checks out.  births - deaths = 2% annual growth?

 

edit:  I am dumb.

Edited by SuingToGetAMessageBoard?
Link to comment
Share on other sites

1 minute ago, sushihorn said:

Not sure how meaningful that is.  Even if the population were completely static and had 100% turnout every year, you'd still have 4% with no voting history every cycle from 2 years of people turning 18. 

Is that really true? Because we know that 18-year-olds don't vote. And I certainly don't think newly minted ones make up 4% of the electorate. 

Link to comment
Share on other sites

2 minutes ago, Dahobbs said:

Is that really true? Because we know that 18-year-olds don't vote. And I certainly don't think newly minted ones make up 4% of the electorate. 

That's fair.  They do turn out at a lower rate than older people.  Let's say half as likely.  That's still a significant chunk of the new voters.  Dallas Co grew 1.5% per year since 2010 and many counties grew faster than that.  Those 2 factors should be a large majority of the effect.

Link to comment
Share on other sites

29 minutes ago, SuingToGetAMessageBoard? said:

Why is it only the 30 biggest counties reporting?  Are they they only ones required to report?  

The TX Sec of State site is showing the daily results for the 30 counties with the largest voter registration.   In previous years they appear to only have shown the top 15.

I don't know if the other ~220 counties are required to report daily counts to the Sec of State office or not.  I looked at a few smaller county websites, and they all definitely do not report the #s on their own websites.  Some of the websites practically looks like they belong on Geocities.

Link to comment
Share on other sites

11 minutes ago, Dahobbs said:

Is that really true? Because we know that 18-year-olds don't vote. And I certainly don't think newly minted ones make up 4% of the electorate. 

Looks like natural growth is about 211,000 per year, or about 1.2% of the voting age citizen population of 17,500,000 in Texas.  So two years gets 2.4%.  18 -24 year olds vote 38% in 2012 (presidential), compared to 49% / 63% / 69% for the older age brackets.  So compared to older voting, it looks like something like 1.8% would be the natural increase in new voters in two years.

Edited by SuingToGetAMessageBoard?
Link to comment
Share on other sites

2 minutes ago, sushihorn said:

That's fair.  They do turn out at a lower rate than older people.  Let's say half as likely.  That's still a significant chunk of the new voters.  Dallas Co grew 1.5% per year since 2010 and many counties grew faster than that.  Those 2 factors should be a large majority of the effect.

I'd really like to see historical data on this. We are seeing up to 10% (Travis) being new voters with no history at all. If that was a typical trend, you'd expect our voter base to completely cycle every 10 years or so. There is no way that happens. 

Link to comment
Share on other sites

Also, I feel we are still a ways away from a D winning Tarrant.  
Yeah, even Tarrant and Williamson being flipped both seem like reaching. Closer, sure, but overall their demographics still aren't that D friendly.

Sent from my SM-G920V using Tapatalk

Link to comment
Share on other sites

1 minute ago, Dahobbs said:

I'd really like to see historical data on this. We are seeing up to 10% (Travis) being new voters with no history at all. If that was a typical trend, you'd expect our voter base to completely cycle every 10 years or so. There is no way that happens. 

Especially considering this is a midterm.  I would guess first time voters in a midterm is tiny.

Link to comment
Share on other sites

1 minute ago, Dahobbs said:

I'd really like to see historical data on this. We are seeing up to 10% (Travis) being new voters with no history at all. If that was a typical trend, you'd expect our voter base to completely cycle every 10 years or so. There is no way that happens. 

Travis Co is growing 3% annually.  It's addition more then turnover.

Link to comment
Share on other sites

29 minutes ago, sushihorn said:

Travis Co is growing 3% annually.  It's addition more then turnover.

I don't mean so much as in replacement, but that the total makeup of voters changes such that every 10-15 years or so the "old voters" (not in age, just meaning with previous history) are outnumbered by the new ones. There is no way that happens in Texas.  But at 10% that would have to be the case. 

Link to comment
Share on other sites

1 hour ago, SuingToGetAMessageBoard? said:

Also, I feel we are still a ways away from a D winning Tarrant.  

I live in tarrant. Unfortunately, it always goes red no matter what. Probably the best you could hope for is that a ton of soccer moms show up, vote for Beto, and Cruz wins here by only 5 or 10 points.

Edited by Voldemort86
Link to comment
Share on other sites



×
×
  • Create New...