Major 538 shift in favor of Democratic control of House

[Awsi Dooger]

Jumped 3% to 77.6% likelihood, and average gain of 36 seats

Both are high water marks, since the chart was posted several weeks ago at 75% debut. Earlier today it was 74% and 34 seats. The prior high I remember was 75.3% and 35 seats. I've seen as low as 71.5% and 33 seats.

Must have been sparked by the widening gap in recent generic House polling

https://projects.fivethirtyeight.com/2018-midterm-election-forecast/house/?ex_cid=rrpromo

[RandySF]

1. The generic ballot polls are catching up with other factors

They are a lot more bullish on Gina Ortiz Jones than I am.

[George Eliot]

2. I learned in 2016 not to count my chickens before hatching.

Even 538 can be wrong. I'm still susceptible to thinking that Trump has more support than many are willing to admit. I'll wait and see.

[RandySF]

3. They were right in 2016

Hillary won the popular vote as predicted and wrote there was a chance of Trump winning the EV while losing the PV.

[George Eliot]

7. Unfortunately we don't use the popular vote.

And the three million were Californians. Wrong is wrong even though people thought she had in the bag. I hope I'm wrong. Don't know what you think of Salena Zito but she was right about the 2016 results. You just can't underestimate the loyalty these people have to Trump. It is irrational.

[leftynyc]

13. Sorry, but that's a

moronic talking point. I could just as easily say he would have lost he popular vote by MILLIONS MORE if only Texas didn't vote.

[George Eliot]

15. Yes, it could be any populated area. But that's called rationalization.

Last edited Tue Sep 4, 2018, 06:28 PM - Edit history (1)

But the fact is such areas do not reflect the United States as a whole. If you want presidents to just represent major cities - mostly on coasts - then change the rules. I wouldn't be against that. But our system recognizes middle America. So it is disingenuous to keep saying "she won..." because she didn't win.

BTW, Texas doesn't hold a candle to California.

[leftynyc]

18. I never said that so you can

stop shoving words into my mouth. Your implication was that she only won the popular vote because of California - who are still Americans and it matters not at all where they come from. Same as the large republican population in Texas are still Americans. The fact is MILLIONS more voted for Hillary. NO state reflects the US as a whole. None.

[Hekate]

20. Well, that's an interesting link: one article inside avers millions of illegals voted. Thanks for...

...providing that eye-opener. Or did I misunderstand?

[Awsi Dooger]

4. I think it is too high but I'll take any positive movement

Much better to force the other side to overcome higher odds. Let them unskew the polls while we rely on normalcy.

Also, Trump's approval rating on 538 is now 40.1%, which is the lowest since 40.0% on April 6

[grantcart]

8. You can count your eggs before they hatch

1) 40 Republican Congressman retiring before they get defeated is a record

2) we aren't running 'geeric candidates'. As Randy has documented in excruciating detail these new candidates, mostly highly educated women but some outstanding young men Lamb, O'Rourke and others give us reason not to be complacent but confident, enthusiastic and determined to help.

[Awsi Dooger]

10. Key sentence regarding Gillum's win in Florida

I don't think this received enough attention, in terms of new voters and enthusiasm on the Democratic side:

https://www.nytimes.com/2018/08/29/us/politics/andrew-gillum-florida-governor.html

"Tuesday’s results showed that a quarter of Democratic ballots cast during Florida’s early voting period came from people who had never voted in a primary before, according to the state party."

[Awsi Dooger]

5. Generic gap on 538 is 10.5%

49.3% to 38.8%

Appears to be the highest gap since early January

https://projects.fivethirtyeight.com/congress-generic-ballot-polls/?ex_cid=rrpromo

[SunSeeker]

6. K & R

[Old Crank]

9. The Economist a couple of weeks ago had their prediction at

about 75 % also.
We still need to get out and vote and drag your friends out to vote. Or the predictions won't matter.

[lindysalsagal]

11. It's been stuck around 73 for weeks. Good move.

[Awsi Dooger]

12. Now above 80%

I never thought I would see that. Another uptick of nearly 3%

Specifically at 80.3% and 37 seats

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

[lindysalsagal]

14. Wow!

[Awsi Dooger]

17. Small dip to 79.3%

Seats down to 36

Still a great turn of events the past few days. It may not seem like much but for reference purposes 71-72% is the NFL money line equivalent of a 5.5-6 point favorite in an NFL game while 79-80% is equivalent to 9.5-10 point favorite.

Small shifts in that range are very meaningful. That's why when Comey dropped Hillary from 80% likelihood on 538 down to 65-72% it was a massive downturn in expectancy, far beyond what was understood by voters and most pundits/analysts.

[Scurrilous]

16. K&R