Talk Elections

http://www.nytimes.com/interactive/2016/10/30/upshot/florida-poll.html

Trump 46%
Clinton 42%

Hmmm - interesting.

That's not good at all (their modeling is good). However they got an 11 point divergence between NC and FL which is very weird

Yikes, this is not good

But the Hillary supporters told me she has it "in the bag" ...

It is real...

It's funny that the commentary around this, whilst arguing the numbers are possible, is very quick pour cold water as this being definitive.

Explain ::)

Wow... the race looks like it really is getting closer.

It is pre-Weiner-gate poll. Junk! :D

4-way: T+4
Trump  46%
Clinton 42%
Johnson 4%

2-way: T+3
Trump  48%
Clinton 45%

Senate: R+9
Rubio    51%
Murphy 42%

Hmmm. No. :)

We've got an NBC poll of Florida coming in 39 minutes so i'll wait.

Knew it ::)

And NC

Eek.

Something seems off though when their model is getting such wildly diffrent results from NC and FL.

NC is off ::)

Honestly, the minority support for Trump do seem very generous. We'll see.

I must say I did not expect this....

I kinda figured Selzer was up to something.

Latino Decisions have it 63-23 amongst Latinos. This poll is 59-30 but also shows a large swing back to Trump amongst Cubans. Might be tied to the Senate race.

She doesnt need Florida but I want her to win it.

Intresting...




Whites are coming home?

New Poll: Florida President by Siena Research Institute on 2016-10-27 (https://uselectionatlas.org/POLLS/PRESIDENT/2016/polls.php?action=indpoll&id=1220161027079)

Summary: D: 42%, R: 46%, U: 5%

Poll Source URL: Full Poll Details (https://www.siena.edu/news-events/article/trump-takes-four-point-lead-over-clinton-46-42)

Hmmm... perhaps registered Republicans are lying about their registration?

Siena/Upshot use a voter file. So they know voters' registrations. As I understood.

IIRC, Cohn mentioned one of the downsides of this type of polling is that it fails to pick up on new registrants, which would miss a recent surge in Hispanics registration. That alone probably wouldn't yield a Trump +4 lead though.

That's correct. They use voter registration file so it won't pick up the most recently registered voters.

But nonetheless, 4 point lead is huge.

If we're counting Upshot as gospel (and I do trust their model a good deal), then NC is just as much a lock for Clinton.

Never Trump Hispanics are coming home. Grand!

Next thing is Ana Navarro voting Trump.
And next thing after that - Torie voting Trump. :)

Pretty obvious outlier

This could be very well true as Hill has received some really bad publicity in the last few days.

I think I agree - not even Gravis is showing a Trump lead in Florida.

That should be a negligible effect though. The biggest difference in this poll is that it weights to party registration, which it notes, lots of other pollster don't. Remains to be seen how accurate this will turn out.

IDK, what type of bad publicity you are talking about but this poll was coducted    OCT. 25-27.

Which could imply that no "movement" occured [yet], and other polls would show the same results if weighted "properly". If this method is accurate and noice didn't cause it etc.

I did. Trump is somehow still ahead in the Early vote after almost a week. The underlying model of a lot of these other polls does not look right.

Did they mention an exact number for their early voter subsample?  I didn't see it in the article.

I havent deep delved into that, but that's definitely a point of calibration. They do have the ultimate poll at R+2.

It's roughly about even at this point in time. I think Trump has a 30K lead as of yesterday.

Definitely seems R-friendly. I doubt that FL will trend strongly R this year. In fact, it's a prime candidate to trend D.

ok what the f**k

I'm regretting ignoring my gut feeling and instinct (and everything I've been saying in posts) and putting Florida in the D column on my official prediction.

guys

I just don't understand where a Trump lead would come from is all - hispanics are turning against him, and there hasn't been any polling showing Trump doing any better than Romney with white voters.

GOP still has some room to grow here with midwestern working class transplants. Trump's NC problem is that Romney maxed out the white non-college vote in 2012.

This is what we get for complaining about all the C+4 polls out of Florida.

No. Hispanics are coming home.

In fact, no. Whites are coming home.

Florida's lost, world's ending, Clinton stands no chance, we're finished. :P

()

Look at the early vote numbers. Obviously, things could significantly change today since it's likely to be the heaviest pro-D day with all the church-sponsored stuff, but Trump slightly ahead between absentees and early vote. Unless this somehow changes over the next few days, you're looking at an R+ electorate in Florida, which favors Trump.

Republicans traditionally vote on Election Day there. Democrats use their padded lead to stave off any Republican advance.

Okay, everyone calm down (including the Trump supporters). Consider the following from the poll write up:


Recall a previous Upshot feature where different pollsters were given the same Florida dataset (http://www.nytimes.com/interactive/2016/09/20/upshot/the-error-the-polling-world-rarely-talks-about.html) and calculated the margin. The two pollsters weighing on party registration predicted 2-4 points higher votes for Trump relative to the one who didn't, holding likely voter screens equal.

The very good writeup also mentions two possible poll biases toward Trump:
- Oversampling of Republicans who support Trump rather than staying home, since weighing was done on party registration.
- The usual likely voter screen critique. Specifically, the voter screen may have worked against minorities sampled in the poll: if Latino or Black voters don't respond that they're very likely to vote, the Cuban-Americans crowd out other votes.

I will say about Florida, Trump has been spending an insane amount of time campaigning there, to the detriment of his campaign elsewhere. The presence of a candidate in a state can temporarily impact polling.

Florida is a must win state for Trump.

Cubans are coming home to Trump.
That's pretty clear.

Tender and all of these Trump supporters need calm down. There is no way Trump will get 11% of the Black vote (You can only dream). Also, Trump is doing somewhat worse among Latinos than the 59-30 margin obviously.

Stop it. Hillary is winning FL (albeit a bit close).

OK. We just need to wait 9 more days and see for sure.

I'm not saying Trump will win Florida by a big margin. I think the best he can hope for is a 1-2% win.

I think this is the reasonable best case scenario for Trump in FL

Ugggg.

yeah, trump can win FL.

and if he is not, we won't be able to see it before election day.

Is it possible turnout taps out eventually? Is there evidence registration has surged there?

I think in this case, Cohn's Likely Voter model isn't going to fare so well when the votes are counted.

Party Reg, he has it at D+1. The Early Vote right favors the Republicans and Republicans traditionally show up on election day in Florida.  Obviously, we'll have a lot more color after today, traditionally one of the biggest D days in the Early vote cycle.

I think we'll all need to reexamine some of these Southern States once the Souls to the Polls numbers kick in tomorrow and SoS sites report the numbers.

I'm not going to geek out just because of a few polls, although data points appear to be converging that potentially Clinton's margins in NC could well outperform her FL margins.

If things weren't bad enough for Democrats, these polls were taken before the FBI reopened investigation.

Thanks for that hard hitting analysis.

FL was always going to be a Trump state, he always got his highest % there, sometimes polling as high as 47-49%.

Has gotta be a concern troll.

The ignore button is right there.

I have no need for a safe space.

Ljube, Cubans are not coming home to Trump. What Trump may be doing, if he is winning in Florida, is overperforming with older, formerly Democratic-voting whites. There are business organizations in Miami that have endorsed every Republican nominee since the 1960s, including Goldwater, and they have uniformly backed Hillary Clinton in this election. It isn't happening.

I don't know why you don't think 11% of the black vote is attainable in an election where the Democrat candidate is white. Traditionally, the Republican share of the black vote is somewhere between 8-12%. With Obama not running, you'd think there'd be a return to the mean.

While all this is true, Trump is not a "generic R" for African-Americans giving his birtherist smearing of Obama over the last several years and his record of discrimination in housing.

What might be happening is reduced democratic AA turnout with Obama off the ballot to the tune of 1-2 percentage points of the electorate back to normal levels when GOP candidates traditionally got about 10% of the AA vote. But Trump is a pretty bad fit for them so I'd expect him to get about 6% of the AA vote with reduced turnout.