2018 Congressional Generic Ballot and House Polls Megathread - the original

[Skill and Chance]

Poliquin gonna get BTFO.

https://www.centralmaine.com/2018/02/23/new-poll-finds-trump-unpopular-poliquin-in-tight-race/

Trump at 52% disapproval in ME-2

Interesting.  He won a majority there, didn't he?

[Skill and Chance]

Even as Trump's approval increases in the new Monmouth Poll, the Generic Ballot remains essentially the same:

RIP Limo

Looking at this and the special election results, it seems clear that there is a percentage of the population that basically thinks of Trump as a 3rd party.  The continuing divergence between the GCB, which points to Dems just barely taking the House with 220-some seats and the average of the 2017-18 special elections, which points to an absolute blowout with Dems getting 250-some seats is fascinating.

[Skill and Chance]

https://twitter.com/MurphinDC/status/992465323088646146

Trump won this suburban Raleigh district 53-44.

Ya see, while the generic ballot is closer than expected, the individual race polling and fundraising numbers seem to be more accurate.

Possible explanations:

1. Modeling the wrong electorate: Trump voters will stay home and it will be a blowout like the average of the specials
2. Modeling the wrong electorate: Trump voters will turn out much more, like they did in GA-06, which means only a few districts flip nationally, mainly Clinton blowout seats like FL-27
3. Utah, Texas, Atlanta, etc. are coming home to Republicans vs. 2016 and inflating their GCB numbers, but the Midwest is coming home to Democrats and that is where most of the swing seats are.
4. Republicans gaining ground in Dem strongholds like VRA districts vs. 2016, inflating their GCB numbers while suburban seats swing even harder against them.

IMO it looks like some combination of 1 and 4.  #2 is possible, but given how the polls missed big on intensity of base turnout in each of 2012/14/16, I doubt it.  I don't really buy #3 at all, as rural Trump/Romney/every Republican since Reagan +20 districts are swinging hardest in the specials. 

I do think declining turnout/margin with minority voters will be a medium-long run problem for Dems.

[Skill and Chance]

Why do people keep saying Dems are losing ground with minorities because of one mediocre pollster? The Special Elections are saying the complete opposite.

Well, they clearly won't do as well as they did among minorities as Hillary, because most democratic candidates will not be running against someone as racially charged as Donald Trump, and they tend to have lower midterm turnout. Hillary pretty much maxed out among minorities, with the exception of 2020 Trump during a recession vs a candidate with strong minority appeal.

Notice how the biggest special election swings have been coming from districts that are mostly white (PA-18, AZ-08, Wisconsin, etc.). IIRC Tipinerni did about the same as Hillary among Latinos despite doing way better overall, and got less Latino turnout. And you can see how dems underperformed Hillary in Milwaukee in the Supreme Court election, specifically in the majority black areas.

Yeah, but Democrat have problems with minority turnout every midterm. It's not losing ground if it's the same exact problem we had in 2014 and 2010.

IMO it's not so much losing ground as they have already hit their ceiling sooner than expected during 2012-16.  The fundamental Dem mistakes in 2016 was thinking they could improve on Obama's margins with non-white voters.

Interestingly, post-Trump Republicans also seem to be hitting their ceiling with white voters sooner than they expected.  As pointed out further upthread, this has be the biggest source of problems for them in the 2017-18 specials.

Overall, racial polarization seems reaching its limit sooner than expected.

[Skill and Chance]

States like Nevada have become way more diverse since the 1990s, but democrats aren't doing much better in Nevada today than they were in the 1990s...

Yes, it is kind of amazing how that has worked out. There is good reason to think that Democrats will have consolidated their power over NV after the Nov elections, but from 2000-now, it's not really been a good time for them. They have had the state House locked down mostly since the 1930s, but that's it. They haven't controlled the Governors mansion for 20 years. That is pretty bad, all things considered. They did better pre-1998.

Edit: I wonder if the inability of Nevada Democrats to consolidate power sooner is due to the post-Reagan shift of WWCs to the Republican Party. NV's changing demographics has helped balance it out, but there seems to be a similar pattern at play here. Or maybe Republicans just had a good run of it, with the electorate generally being too swingy for Democrats to lock down.

Don't forget the impact of retirees in NV as well.  It's a similar story in AZ and FL of how Republicans came back from 1996. 

[Skill and Chance]

Reuters/ISPOS

Democrats: 40% (+2)
Republicans: 37% (-/-)

Democrats improving.

Strangely, the amount of Dems and Reps surveyed are the same... So I highly doubt that it's accurate. I think we all expect more Democrats to vote on Election Day than Republicans.

https://www.ipsos.com/sites/default/files/ct/news/documents/2018-05/2018-reuters-tracking-core-political-05-23-2018.pdf

That wouldn't flip the House, but contrary to the CW, I think D+4-5 would. 

[Skill and Chance]

CBS/YouGov, May 24-30, 24759 registered voters including 5693 in 64 "competitive and likely competitive" districts.  The overall topline is D 43, R 38.  Applying a "multilevel regression and post-stratification model", they estimate the final tally at 219 D, 216 R...but with a MoE of +/- 9 seats.

The 64 districts are AR02, AZ01, AZ02, CA07, CA10, CA21, CA24, CA25, CA39, CA45, CA48, CA49, CO06, CT05, FL07, FL18, FL26, FL27, GA06, IA01, IA02, IA03, IL06, IL12, KS02, KS03, KY06, ME02, MI08, MI11, MN01, MN02, MN03, MN08, NC09, NC13, NE02, NH01, NH02, NJ02, NJ03, NJ05, NJ07, NJ11, NV03, NV04, NY11, NY19, NY22, OH01, PA01, PA05, PA06, PA07, PA08, PA17, TX07, TX23, TX32, UT04, VA02, VA07, VA10, and WA08.

This is roughly what I think will happen.

[Skill and Chance]

I know some of you dream of 2006. But what if 2010 was a realigning year with regard to how folks vote in Congressionial elections.  Have you  put that possibility into your analysis?

Republicans already had a realignment in the 1980s, which finally took effect downballot in 1994 and Congress has had a Republican lean to it ever since. 2006 and 2008 were exceptions to this - Dem waves, which toppled Republican majorities. 2010 was simply a backlash to Obama that returned Republicans to control that they were bound to get eventually anyway. If you look at the popular vote totals of elections since the 90s and the general state of the states right now, not a whole lot has changed at the macro level even if the electorate has been shifting underneath.

In fact, if anything, Democrats have actually made more progress in terms of expanding their "baseline" Congressional power. Clinton's 2% win won almost as many Congressional districts as Obama 2012, despite having almost half the PV margin. If she got Obama's 3.7%, she likely would have won at least a bare majority of districts.

This is underrated.  Based on the specials and polling, the areas most likely to "trend" Republican this year are majority-minority seats with rural or manufacturing influence.  That would substantially unpack the national Dem vote vs. the Obama-Clinton era. 

[Skill and Chance]

I know some of you dream of 2006. But what if 2010 was a realigning year with regard to how folks vote in Congressionial elections.  Have you  put that possibility into your analysis?

Republicans already had a realignment in the 1980s, which finally took effect downballot in 1994 and Congress has had a Republican lean to it ever since. 2006 and 2008 were exceptions to this - Dem waves, which toppled Republican majorities. 2010 was simply a backlash to Obama that returned Republicans to control that they were bound to get eventually anyway. If you look at the popular vote totals of elections since the 90s and the general state of the states right now, not a whole lot has changed at the macro level even if the electorate has been shifting underneath.

In fact, if anything, Democrats have actually made more progress in terms of expanding their "baseline" Congressional power. Clinton's 2% win won almost as many Congressional districts as Obama 2012, despite having almost half the PV margin. If she got Obama's 3.7%, she likely would have won at least a bare majority of districts.

I think you’ve hit the nail on the head here - the parties really haven’t shifted much at all in the last twenty or so years, and while the general political climate has shifted back and forth the individual districts have moved based on population dynamics, the simple fact is that the core constituencies have remained the same.

2010 was a massive wave year for the Republicans, but it’s not like they won in places they simply hadn’t even come close in previously. Just Rs dominated in places that they had either won in at the presidential level or had come close in 2004 or 2008.

Wouldn't that just mean Dems are out of luck in normal years until there is some major realignment?  I mean, other than maybe a 221 seat House majority, what is there for them in a narrow PV win right now?