Talk Elections

http://mobile.nytimes.com/2016/11/08/upshot/trump-and-clinton-tied-in-final-upshot-poll-of-north-carolina.html?referer=https://t.co/b6K2oygAcz

Clinton 44%
Trump 44%

this is going to be the closest state! Trump barley wins in the end.

Pretty bad. Wasn't their last one Clinton +7?

Told you.


Someone who wants to complain about 538's "stupid" trend lines? ::)



It was. Tightening :)

Shy Trumpistas coming home

But not all of them  :)

This is more in line with reality.

I thought it would be tight. The difference between this poll and others was the white margin, and now it's in line with other polls.

Crosstabs from Sienna:

http://files.constantcontact.com/9c83fb30501/5aa2333d-0f5b-49b3-92d6-dfcc5a1c58f6.pdf
http://files.constantcontact.com/9c83fb30501/3603bf1b-a0f6-4234-ba77-34eeadff55d0.pdf


Republican retain control 47%
Democrats take control    45%

Educated white women ----> Trump :)





()

That graph just says White. It doesn't break down by education and gender together. Clinton is, in fact, winning the educated voters.

It was not what I said.

Educated white women --coming home--> Trump

And they are probably are coming from Johnson.

... Let me try this again.

Graph says White, which includes White men and White women. Not Educated White women, not Educated White men, not White women, not White men, but ALL White voters.

You CANNOT say that Educated White women are "coming home" to Trump based on that.

As much as I approve of Nate Cohn's valiant attempts to try to build a model based on actual turnout data, rather than polls, this poll exemplifies why that is a flawed project. There's no reason to believe that the early vote shifted to Trump over the past two weeks...

So compared to Sept

Trump share grew by
+6 white
+3 black (but blacks are just 22% of voters)
-1  men
+8 women
+2 no college
+8 college or more.

And you still think that could happen without educated women coming home? Yeah, teoretically it could be some weird flows that could explain it like Trump losing among no-col men, but gaining among non-col women and col-men ::)

I guess we can put to rest that Sweden and other countries like it have the best education system in the world!

This makes sense in an election where the Democrat wins nationally by 3-5 points

I can confirm that the Swedish education system is really awful, especially in math and statistics and especially over the last decade or so.

No. GOTV will put her over the top.

But we are best at genus science!!!!111

But not just educational system has worsened, dear Gustaf. So did police (no-go-zones), health and elderly care. Thank political correct left :)

And what about our liberal court of law? New trend among rapist: they claim they did it while sleeping... and a lot of them were freed by Swedish liberal court. Thank political correct left :)

I wounder if it has something to do that Sweden switched from labour migration to asyl migration from Arabic and African countries. Especially over the last decade or so ::)

http://www.migrationsverket.se/download/18.2d998ffc151ac3871599d0d/1462790765011/Beviljade%2Buppeh%C3%A5llstillst%C3%A5nd%2B1980-2015.pdf

No?

This is definitely more believable than Clinton +7. I think she wins by 1-3% here.

"Mrs. Clinton leads among voters who have already cast their ballot by a nine-point margin, 49 percent to 40 percent. But Mr. Trump leads among those yet to vote by 17 points, 52 percent to 35 percent."

This implies that about 65% of their sample has already voted early.

63% according to crosstabs :P

They've updated the model:

Clinton +1.1 is the call.

http://www.nytimes.com/interactive/2016/upshot/north-carolina-early-vote-tracker.html

Now it is look much better. Is it base on two polls, or just latest one?

EDIT: OK, it based on all 3, but the latest poll is sort of basis.

Nate Cohn's model is a great idea but the wild changes that this poll has had on his early vote estimate suggests that the only way to implement this idea soundly is if the NYT gave him a budget that's ~10 times larger or so. I think this only works if he conducts polls every day or so with the specific intent of garnering a huge sample of early voters that's more stable. We really shouldn't be seeing fluctuations in the early vote that are this huge: these people already voted!

His revised model will probably be much closer to the final result but I'm left feeling like this will be by coincidence rather than the soundness of this whole project.

I couldn't find the actual poll-- can you post the link to the crosstabs? Thanks.

Here

Thank you!

And updating the numbers:


It look like Ross is running slightly ahead of Clinton and Cooper in EV, despite trailing overall. Interesting!

You are a confirmation of it all by yourself :P