@Nate_Cohn's Threads
Nonvoters do not seem poised to save the Democrats in the six key battleground states, based on an analysis of more than 600 nonvoters in our Times/Siena polls.
Despite their demographics, they are not much likelier to back Democrats than voters
nytimes.com/2019/11/07/ups…
Despite their demographics, they are not much likelier to back Democrats than voters
nytimes.com/2019/11/07/ups…
The results are consistent with an analysis of national polling of unregistered voters and Times/Siena '18 data over the summer
nytimes.com/2019/07/15/ups…
nytimes.com/2019/07/15/ups…
Our results do offer some good news for Democrats among nonvoting Hispanic voters, who back the Democratic candidates by huge margins and could propel the party forward in Arizona or perhaps even Texas
But outside of Arizona, Democrats are deadlocked with nonvoters
But outside of Arizona, Democrats are deadlocked with nonvoters
We went in-depth on the persuadable voters in our surveys, who said they were really open to voting for either Trump or the Democrat.
They weren't what I expected
nytimes.com/2019/11/05/ups…
They weren't what I expected
nytimes.com/2019/11/05/ups…
We also, probably wisely, rejected two semi-serious proposed nicknames for two kinds of swing voters: the HBO conservatives and the MMA (or WWE) populists
The latter is a real revelation to me: young, men, often nonwhite, kind of the Trump show, doesn't understand or even resents elite PC culture, does not like Warren, but is in no way conservative on the issues and still wants big change
New Times/Siena polls show Trump highly competitive in the six closest states carried by the president in '16:
Biden+2, Biden 46 to Trump 45
Even, Sanders 45, Trump 45
Trump+2, Trump 46, Warren 44
2016 result in these states was Trump+2, 48 to 46
nytimes.com/2019/11/04/ups…
Biden+2, Biden 46 to Trump 45
Even, Sanders 45, Trump 45
Trump+2, Trump 46, Warren 44
2016 result in these states was Trump+2, 48 to 46
nytimes.com/2019/11/04/ups…
Crosstabs and methodology here.
Unlike most state polls, we're weighted by education
Unlike most national polls, we're weighted (and stratified for that matter) on party registration, so we have the right number of reg Dems/reg GOP in each state
github.com/ndcohn/battleg…
Unlike most state polls, we're weighted by education
Unlike most national polls, we're weighted (and stratified for that matter) on party registration, so we have the right number of reg Dems/reg GOP in each state
github.com/ndcohn/battleg…
The main difference between our polls and recent national polls, to me, stems from two things.
1) The EC-battleground gap, which was 4 points in '16 and could easily grow in '20, esp among RVs. This is basically what we thought the result should look like, back in July
1) The EC-battleground gap, which was 4 points in '16 and could easily grow in '20, esp among RVs. This is basically what we thought the result should look like, back in July
I think the Canadian election result is pretty interesting, and useful to think about in the context of the various American debates over the Electoral College / popular vote / US House
As you may know, Trudeau was reelected and fairly comfortably. But it appears the conservatives won the most votes, reversing what we've seen for the center-left in the United States (and the UK on Brexit, if it had been by constituency rather than popular vote)
So what happened? One big part is that conservatives are dominant--truly dominant--in Albert and Sask., where they win 2/3 of the vote and beat the Liberals by a huge margin.
Voters in the states likeliest to decide the presidency support the House impeachment inquiry but still oppose impeaching the president and removing him from office
nytimes.com/2019/10/21/ups…
nytimes.com/2019/10/21/ups…
A crucial sliver of the electorate--7 percent of voters--fall into the "inquiry, not remove" camp. They're relatively young, independent, perhaps a hair cynical about politics, and less likely to follow the news very carefully
Voters who disapprove of the president split along lines of political engagement. The president's less engaged opponents aren't ready to remove the president
The thing is that this is how it was seen heading into the election: the 'midwestern firewall,' Bain, outsourcing, the autobailout, etc. Then the exits hit, Hispanics become the story, and Sean Hannity wants immigration reform a week or two later
But the exits were wrong about it all. Wrong about Obama's weakness with white voters, wrong about the age and educational composition of the electorate, wrong about the Hispanic turnout surge
nytimes.com/2016/06/10/ups…
nytimes.com/2016/06/10/ups…
I think you can make a case that this seemingly small thing--the 2012 exit poll and over-interperation of it--is really a major cause of a lot of what's gone haywire in American life over the years that followed
Something that has interested me in recent months: Trump's approval rating has increased a bit in live interview polls, both surpassing the online polls (for the first time) and reaching the highest levels of term
nytimes.com/2019/08/07/ups…
nytimes.com/2019/08/07/ups…
I often see a 2016-based view of 2020 in my replies. They take Trump '16 support, and then say Dems can win with incremental changes like higher black turnout or a better candidate.
I don't think that assumption works, in part bc Trump is also stronger than he was in '16
I don't think that assumption works, in part bc Trump is also stronger than he was in '16
And again, none of this means Trump is favored. His ratings are well under 50. He'll probably lose if he's at 45/55 with RVs and I'm pretty agonistic on his chances over the long term.
But all of this strikes me as pretty important for the way to think about his position
But all of this strikes me as pretty important for the way to think about his position
Trump's Electoral College edge endures heading into 2020, and it could grow further--potentially allowing him to win the presidency while losing the national vote by 5 points or more nytimes.com/2019/07/19/ups…
That huge split between the Electoral College and the popular vote would be unprecedented and it is not necessarily likely.
But it is imaginable, given the president's resilience in Wisc. and Fla., and the potential for a historic turnout amid a racially polarized campaign
But it is imaginable, given the president's resilience in Wisc. and Fla., and the potential for a historic turnout amid a racially polarized campaign
The president's resilience comes down to two crucial but iconoclastic regions: Miami-Dade County and the Milwaukee sububrs. There, the president's approval rating seems to be holding firm or, in Miami-Dade, may be higher than his '16 vote share.
Democrats, only:
Do you consider yourself very liberal, liberal, or a moderate / conservative?
Do you consider yourself very liberal, liberal, or a moderate / conservative?
Finally, Democrats only:
Do you consider yourself a socialist or a Democratic socialist?
Do you consider yourself a socialist or a Democratic socialist?
One more.
Democrats only:
Do you agree or disagree with the following statement: political correctness is a problem in the US?
Democrats only:
Do you agree or disagree with the following statement: political correctness is a problem in the US?
We got a glimpse of conditions that yield a viable third party bid in 1992, when Perot was tied or even ahead in polls ahead of the DNC.
Here, CBS/NYT in 6/92: B32, P30, C24
Bush app: 34%
Clinton fav/unfav: 16/40
Perot fav/unfav: 26/20
By all measures, status quo less conducive
Here, CBS/NYT in 6/92: B32, P30, C24
Bush app: 34%
Clinton fav/unfav: 16/40
Perot fav/unfav: 26/20
By all measures, status quo less conducive
(I would also argue that Schultz is not as well positioned to take advantage of whatever opening exists as Perot was, but that's a separate question)
Of course, we haven't had many vigorous third party bids.
But it makes sense to me that you'd want to see *both* Trump approval and likely Dem nominee favorability under at least 35, and probably lower, before there was a real opening
But it makes sense to me that you'd want to see *both* Trump approval and likely Dem nominee favorability under at least 35, and probably lower, before there was a real opening
I'm so upset right now. Infuriating game
Forget the Q4 refs. The offensive play calling is outrageous. Wilson is the best player on the field and he wasn't even given a chance to win this game
I really can't remember a more upsetting game. I really don't get upset easily, but "shuffle" on the playbook might have been better offensive playcalling
The missing NC-9 vote nytimes.com/2018/12/07/ups…
A minor note here--this data will differ somewhat from most other analysis in that I've removed all of the absentee voters who wound up voting by other means (in person on election day or in early voting) or who appear to have change counties since they applied
Interestingly, some (perhaps all, didn't go through) voters who submitted affidavits about people picking up their absentee ballots wound up voting by other means (which makes sense, since clearly they realized something was up)
In general, CHS enjoying a stronger turnout in white areas than Roy Moore did, but Espy not as obviously getting a higher turnout than Jones in black areas (though fewer black counties completed).
Probably relatedly, CHS also running ahead of an ALSEN baseline in white areas
Probably relatedly, CHS also running ahead of an ALSEN baseline in white areas
And as the turnout picture shapes up, Espy appears to run even farther behind his 'true' targets than the ALSEN based targets we started with. Mid-single digits with an ALSEN-like turnout can quickly turn into an 8 point defeat or worse if the turnout picture isn't so favorable
And here's a quick look of what we've seen RE: turnout so far. Not much out of majority black counties, but none of it is keeping up with the more robust returns from white areas
As far as I'm concerned, we're basically in wait and see mode until we get a handful of completed counties.
In the mean time, the enlarged CHS margin in DeSoto and narrow Espy lead in Warren are cause for cautious GOP optimism. But you can't say much until you're at 100% here
In the mean time, the enlarged CHS margin in DeSoto and narrow Espy lead in Warren are cause for cautious GOP optimism. But you can't say much until you're at 100% here
Taking the broader view besides Warren/Desoto, I'd still argue that Espy is running a few points behind his targets. But again, who knows what a Dem win looks like here, turnout is quite uncertain, and the picture within county is pretty tough due to racial segregation
The other thing I'd note, though, is that the 'target' I'm using here is mainly based on the Alabama Senate race. Can certainly imagine an Espy road that involves even higher black turnout than in AL (which was very very good), and weaker performance in white rural areas
In writing this article on polling I took a cursory look at how individual pollsters fared. I thought YouGov really stood out. A very good year. CNN, on the other hand, stood out on the other end of the spectrum.
tbh it might just be RDD state polling in general. personally, i'm not even sure how you do it anymore, and it was at best a pretty mixed set of results there
here are all the pollsters with at least 5 separate races over the last 3 weeks
State polls in 2018: pretty good, but still off in ways highly reminiscent of 2016 nytimes.com/2018/11/21/ups…
It's a little hard for me to look at this in our own polls, mainly because they're CD level. But tbh I don't really see much evidence of the same pattern in the upper-right (we did underestimate Dems in the southwest, tho)
To take one high-profile example: we're basically the only public poll that had Blackburn ahead (+14) by more than her eventual margin of victory (+12). One of our least accurate polls was a GOP-biased poll in MN08. Overall, no bias in polls of states where polls were D in '16
The result in UT-04 clinches IA-03 as the closest of our final polls, at least by simple margin. Very close though!
one aside on this chart: if you had asked me on 11/6 to rank our final polls (like, finished on 11/4) from most to least error, i probably would have put them in approximately the opposite order of what actually happened
i was looking at our IL14 poll like we'd just be lucky for underwood to win by a hair, let alone by 5
Is there anyone out there who understands the origin of the estimate that there are only ~400 absentee ballots to be reconsidered in Gwinnett in GA-7? @conorsen @bluestein
I don't really see support for the estimate in the absentee file, which, as I interpret it, implies there are no fewer than 480 ballots in this category and potentially hundreds more. But I could be wrong
With Woodall's lead down to around 500, and these ballots expected to break overwhelmingly to Dems, the difference between 400 and 800 ballots in this category is pretty significant
The absentee ballot birthdate decision in Gwinnett + remaining provisional ballots keep GA07 very interesting. Best Dem shot at a 40th seat
There were 1048 rejected absentee ballots in GA7.
Hard to tell how many due to birth year. About half are 'insufficient oath information,' which could include birth year. 285 are explicitly 'yob missing'
Hard to tell how many due to birth year. About half are 'insufficient oath information,' which could include birth year. 285 are explicitly 'yob missing'
Should note that I last pulled the absentee file the day before the election. So there's a chance it's a bit higher now.
Dems had some decent showings in farm-belty parts of the Midwest (IA, KS, SD, nw OH), but had mediocre to outright bad showings in traditionally Dem, old industrial/union parts of the Midwest. Metro areas pretty bad for GOP across region, of course
Dems came closest to putting it all together in PA. They got their suburban margins. Western Pennsylvania was great for Dems in gov/sen/pa16 house. NE PA was ok for GOP, but not great like IL12/MN8/northern Ohio/rural mich
Also GOP erosion in south-central PA (york/dauphin/lancaster/cumberland) is easy to overlook, but it's part of a long term trend and it could be a real issue the GOP going forward
I think one of the most important, outstanding questions about this election, with big implications for 2020, is what went wrong for Democrats in Miami-Dade County
Clinton win Miami-Dade by 29; Obama won it by 23; in 2014, an overall bad year for Democrats, Crist won it by 19 with a turnout of just 521k.
At the moment, Nelson and Gillum are up just 21 points with turnout at 800k
At the moment, Nelson and Gillum are up just 21 points with turnout at 800k
In a lot of ways, Democrats did what they needed to do to win Florida. That Pinellas margin should be good enough. The Seminole/Duval wins are really impressive.
Orange/Osceloa were bound to be bad for Dems in a midterm, I'd say.
But I wouldn't say Miami-Dade was inevitable
Orange/Osceloa were bound to be bad for Dems in a midterm, I'd say.
But I wouldn't say Miami-Dade was inevitable
Senate map in 2020 opens a little for Democrats if Texas in play again, but still tough. If Dems lose FL, they'd need 3 for 50/50; 4 if Dougie Jones can't pull it off twice. 5 if Trump wins.
CO, AZ, NC, ME, IA, TX, GA seem like top 7 D shots. Not easy. washingtonpost.com/opinions/beto-…
CO, AZ, NC, ME, IA, TX, GA seem like top 7 D shots. Not easy. washingtonpost.com/opinions/beto-…
deleted prior tweet bc i forgot about AZ special
a couple of tweets about MT if Bullock runs, which is a fair thought. seems more plausible than iowa imo
Caution to analysts who use exit polls: many big shifts in this year's exit polls are due to methodological changes, not just actual changes among voters
As I've said a million times, the exit polls have long been biased towards having too many well-educated and young voters. This year, they've made changes to try and fix this.
As a result of the methodological change, the college educated share of the electorate in the exit poll is 10 pts lower than 2014. In reality, this was a better educated electorate, but the exit polls were just that biased by this measure before.
Catching up on last night's vote count. The late mail ballot margins put Democrats on track for victory in all of CA39, CA45, and CA10, supposing those margins continue, as expected.
Won't be called for a while, but might not even be all that close in some of these races
Won't be called for a while, but might not even be all that close in some of these races
That would put the Democrats at +38 in the House; +39 if you think they're favored in ME-2 ranked choice (and I think they are)
And CA48 is obviously done. I haven't even been looking at it. Projectable, as far as I'm concerned
