2018 Midterm Election Thread

[B-Fly]

I'm surprised Tester's name never comes up as a Presidential candidate. I think he would have a pretty broad and deep appeal.

[Teenwolf]

The last two polls by Emerson (B+) and Quinnipiac (A-) 50-47 51-46 I still don't see a win for Beto but the range is now moving the other way. I think it will be fun to see where it winds up.

If polling has it within 5 pts on election day, I bet Beto wins. He has a ton of support from folks that don't count as likely voters in polling, so 5 pt margin could be close enough to take it.

Did you see the LA Times polling? They were consistently the most favorable poll for Trump in '16, and they're an outlier again... this time they have Democrats up 57-40 on the generic ballot. If they're closer to the results than the rest of the polls again, you're looking at a bloodbath.

[chancellor]

Don't Libertarians pride themselves on their rugged individualism...why would they switch over to a candidate that a quitter recommends?

It'll depend on how they react to the reason their candidate dropped out. I suspect most, though not all, will move that way, but I also doubt it'll be enough to beat Tester. But I suspect the race became a lot closer.

[B-Fly]

If polling has it within 5 pts on election day, I bet Beto wins. He has a ton of support from folks that don't count as likely voters in polling, so 5 pt margin could be close enough to take it.

Did you see the LA Times polling? They were consistently the most favorable poll for Trump in '16, and they're an outlier again... this time they have Democrats up 57-40 on the generic ballot. If they're closer to the results than the rest of the polls again, you're looking at a bloodbath.

I guess I view anyone's reliance on the most favorable outlier poll for their team as mostly wishful thinking. That's why, despite that there have been elections with unexpected results and you can point to an outlier polling firm that came closest to "getting it right" that time, I tend to lean heavily on the outlets like 538 that have a clear methodology for processing all of the various polling inputs.

[frae]

If polling has it within 5 pts on election day, I bet Beto wins. He has a ton of support from folks that don't count as likely voters in polling, so 5 pt margin could be close enough to take it.

Did you see the LA Times polling? They were consistently the most favorable poll for Trump in '16, and they're an outlier again... this time they have Democrats up 57-40 on the generic ballot. If they're closer to the results than the rest of the polls again, you're looking at a bloodbath.

I saw it, but it is such an outlier I don't know what to do with it. In 2016 they basically had the results at 46.8 - 43.6 in favor of Trump whereas the avg was 48.5 - 45 Clinton. So basically a flip whereas this is taking the +9 Democratic average and almost doubling it. I am all for good news, but that result would shock me.

As for Beto I'll keep my expectations in check especially when you see Cruz in both of those polls over 50, but I think has continued to campaign well and has enthusiasm for him about as high as a Democrat can in Texas.

[frae]

I guess I view anyone's reliance on the most favorable outlier poll for their team as mostly wishful thinking. That's why, despite that there have been elections with unexpected results and you can point to an outlier polling firm that came closest to "getting it right" that time, I tend to lean heavily on the outlets like 538 that have a clear methodology for processing all of the various polling inputs.

This, a lot of this. Even when everyone likes to point out that they got the 2016 election wrong 538 basically nailed the popular vote. The polling data had the lean in the midwest wrong, but that doesn't make the system bad.

[Teenwolf]

I guess I view anyone's reliance on the most favorable outlier poll for their team as mostly wishful thinking. That's why, despite that there have been elections with unexpected results and you can point to an outlier polling firm that came closest to "getting it right" that time, I tend to lean heavily on the outlets like 538 that have a clear methodology for processing all of the various polling inputs.

I agree, wishful thinking.

Do you agree with me that previous non-voters can be poll-breakers, as they aren't counted in polling of likely voters? This would include people too young to have voted in the last cycle, which if they even vote at a 30-40% clip with a 80-20 type of Dem lean... all those things make a difference. I think that a 5 pt polling lead is small enough for these factors to swing it.

[Teenwolf]

This, a lot of this. Even when everyone likes to point out that they got the 2016 election wrong 538 basically nailed the popular vote. The polling data had the lean in the midwest wrong, but that doesn't make the system bad.

I think it's important to look at the reason they got the midwest wrong. Seems that "unlikely voters" did the unexpected and pushed Trump in. That's why I'm so optimistic about certain Dem races which are relying on turnout from this target group (unlikely voters). Trump showed the viability of that path, in my mind.

[B-Fly]

I agree, wishful thinking.

Do you agree with me that previous non-voters can be poll-breakers, as they aren't counted in polling of likely voters? This would include people too young to have voted in the last cycle, which if they even vote at a 30-40% clip with a 80-20 type of Dem lean... all those things make a difference. I think that a 5 pt polling lead is small enough for these factors to swing it.

I think the Trump era adds a wild card to predictions on turnout, and that we can't really know at this point whether that Trump factor (which I'd broadly define as greater fear/anger/demonization/catastrophizing of the "other side", greater tribalism, greater mistrust of media and institutions) will skew in favor of the Republicans or the Democrats in various critical races. I hope the Trump factor drives higher opposition turnout of voters not captured in the likely voter models, but I also worry that it could drive higher Trump supporter turnout among voters not captured in the likely voter models, even without Trump's name on the ballot, because Trump is largely signaling to his supporters that these midterms are a referendum on him, whether out of narcissism or calculated strategy.

[Gregg]

because Trump is largely signaling to his supporters that these midterms are a referendum on him, whether out of narcissism or calculated strategy.

It is both.

I would say it is the hail Mary for the House.

Then again I am not so sure that this is not the same strategy as all Presidents who speak for their parties employ except this is filtered through the Trump personality. And we haven't really seen anything like it.

[gcstomp]

Just listened to Oprah opening speech for endorsing Stacey Abrahms. Just powerful, uplifting, and motivating speaker. (Think she should have a talk show or something...) Oprah made it very clear she is a registered independent, she has voted both D and R, and she asked Stacey Abrahms if she could come and say a few words just 3 days ago.

When Oprah says you must vote, or you disrespect your family that has passed that never had the right to vote, and mixes in passages from Maya Angelou, as well as stories of an older resident putting on his best suit and walking 6 miles to polling place and being denied chance to vote, as many in Georgia are being denied. Oprah is a moving speaker, and I think she moves needle by herself. Abrahms opponent, who happens to be in charge of polls, who had a million voters removed under the use it or lose it statute, just major suppression taking place right now in a key state. Oprah made it clear she has zero interest in running for any office at any time, but felt strongly about coming out and showing support.

Just the contrast between this rare speaker who brings light, hope, positive energy, she contrasts so strongly with how Trump speaks, in ugly dark attack snippets of untruths. Listening to Trump you feel the fear he is selling, and you feel dirty for just having endured the fetid display.

[frae]

So while the Senate and the House of both settled in at 1/7 vs 6/7 the race for the Govenors mansions is still very interesting. 538 had a few races flip recently. The interesting part is that different states move in different models.

In lite version Ohio and Kansas turn blue on top of WI, FL

https://projects.fivethirtyeight.com/2018-midterm-election-forecast/governor/#lite

In the classic version Ohio Iowa and WI stay blue along with FL

https://projects.fivethirtyeight.com/2018-midterm-election-forecast/governor/

In the deluxe version we see NV move over with Iowa, OH, WI and FL

https://projects.fivethirtyeight.com/2018-midterm-election-forecast/governor/#deluxe

So all 3 models have WI, OH, and FL going for the democrats and FL going by about 4 points. The other races that move in and out of Toss up all go to the Democrats in at least one model except for Georgia. The other states to keep an eye on now seem to be South Dakota which in the lite version (polls only) is a toss up. The other state that could be worth watching is Alaska. Even though the last two I listed are not likely to be fair if the electorate moves to the right Tuesday Oregon and CT seem to be the two races Democrats are favored in that could move into play. They are both favored at about 80% likely to win in the classic model, but if I am listing Alaska and South Dakota if things break blue I wanted to be fair and list these.

With that said things finally do look good for Democrats somewhere other than the house. I am optimistic that wins at the Gov. level will help down ticket races especially in state houses where democrats need to start winning back seats. Here is a 538 article on Democrats trying to win Gov. races in red states...

https://fivethirtyeight.com/features/governors-update-democrats-could-win-in-some-very-red-states/

I'll wait until Tuesday to do final picks, but as of today I am optimistic about the house and the Gov races. The Senate is all about damage control now. A zero net gain for either side would be great by me and if they manage to hit everything and get it to 50-50 I'd have a party. I know some people are hoping for TX and TN, but for me the best scenario is hold onto everything that is lean or likely D. Keep FL, IN, and Mizzou. Flip AZ and NV. I don't see any polling data to give me hopes in ND so maybe you can convince me the next best case scenario is some insane number of young voters do something in TN or TX. For now 51-49 I'm cool, 50-50 I am over the top happy.

[Bernie Brewer]

So all 3 models have WI... going for the democrats...

I agree that all three models have this, but unless I'm wrong, I am not seeing Evers beating Walker. I have lots of family in Wisconsin and there just doesn't seem to be a lot of excitement in favor of Evers. Walker rises from the dead more often that Dracula!

Chance, what are you seeing, you live there?

[revo]

Latest poll in Texas (Change Research, Nov 2) has Cruz & Beto at 49-49.

[onejayhawk]

Latest poll in Texas (Change Research, Nov 2) has Cruz & Beto at 49-49.

Kansas may very well choose a Democrat for Governor. They have a history of doing it. In a more liberal state (like Iowa) he would be a Republican.

J