FL/OH/PA/NC-Quinnipiac: Tied in FL, Trump lead in OH, Clinton lead in PA/NC

[Seriously?]

The 2-way/4-way difference in Ohio is striking.

Eh. All MOE stuff, including the NC result.

Fundamentally, where the race is right now, this looks about right.

[Senator-elect Spark]

Clinton is not going to win NC. If Obama couldn't do it again in 2012, it's not happening.

What does his narrow loss in 2012 have to do with this? You realize the current trends in North Carolina actually favor Democrats, right? By all measure, the NC electorate will actually be slightly more favorable to Democrats this year than in 2012, due to demographic changes alone.

Just curious what exactly your justification is for your position. North Carolina is becoming more competitive, not less, and Trump is clearly not a shoe-in for that state based on the polls we have been seeing.

I agree with those statements about changing demographics. My justification for my position is that Hillary Clinton is not Obama and will not drive turn out as effectively in AA communities (Raleigh, Charlotte). Trump will gain more support among blacks than Romney did, simply because there is not an AA present on the ticket.

A) No reason to assume African American turnout will drop all that much.
B) White college voters are shifting towards Clinton, Trump already maxed out here with Romneys numbers among non-college whites
C) Ground game, ground game, ground game
4) HB2 and the general state of the NC GOP.

It will decrease and that's a fact. Both nominees are extremely unpopular and there is a reason to believe that turnout will be lower than 2012. Trump has ground to gain with college educated voters and I believe that he will in the end. He should lose that group narrowly.

[Erich Maria Remarque]

Nice!

Little by little


NC and Ohio are little bit off, but overall it seems like Clinton up 1-3 nationally.

[Virginiá]

I agree with those statements about changing demographics. My justification for my position is that Hillary Clinton is not Obama and will not drive turn out as effectively in AA communities (Raleigh, Charlotte). Trump will gain more support among blacks than Romney did, simply because there is not an AA present on the ticket.

Basically what Wiz said:

You must be logged in to read this quote.

The scenario you are proposing - that reduced African American turnout/support will tank Clinton in NC simply because she isn't black, that would require a pretty decent drop given the support Clinton has from other demographics. Such a drop is unlikely in just 1 election. Black turnout was already trending upwards before Obama, he just supercharged it. Given that studies show people who vote once tend to vote more and more, I'd be skeptical of any claims of the "2012->2004 electorate theory."

To emphasize on Wiz's other points - Clinton is beating the crap out of Trump with GOTV in NC, and her support from college educated whites is valuable as they have higher turnout rates than non-college whites. Romney got, what, almost 60% of college educated whites in 2012, and 74% of non-college? 42-38 is a significant narrowing, even if most of Trump's lost support went to 3rd parties or is undecided.

So I dunno. There are too many reasons to believe Trump is at a disadvantage here if Clinton keeps the race in NC close.

It will decrease and that's a fact.

Oh my god, no. You can't claim something like that as a fact. It's your opinion. The election hasn't even happened yet.

Both nominees are extremely unpopular and there is a reason to believe that turnout will be lower than 2012. Trump has ground to gain with college educated voters and I believe that he will in the end. He should lose that group narrowly.

Based on polls in the past that have correlated somewhat with turnout, there is actually reason to believe that turnout could be high, or remain at 2012 levels:

https://www.brookings.edu/blog/fixgov/2016/07/08/why-clinton-and-trump-may-increase-voter-turnout-in-2016/

[DrScholl]

Considering that Quinnipiac never produces results that are overly Democratic, these are good results.

[Illuminati Blood Drinker]

Wow, if Clinton is getting these numbers in rigged Quinnijunk polls, imagine what her numbers must really be! Trump is FLOUNDERING, folks. Sad!

[HillOfANight]

Considering that Quinnipiac never produces results that are overly Democratic, these are good results.

https://twitter.com/ForecasterEnten/status/773962753778323456
Harry Enten of 538 noted that the racial weighting seems a bit funky. 64% white, vs 71% in 2012.

I played around in Excel, and matching 2012's racial demographics is a 1 point Trump lead, and assuming 70% is a tie.

[Erich Maria Remarque]

Wow, if Clinton is getting these numbers in rigged Quinnijunk polls, imagine what her numbers must really be! Trump is FLOUNDERING, folks. Sad!

Love you!

Considering that Quinnipiac never produces results that are overly Democratic, these are good results.

You too!

[Person Man]

Down 4 in North Carolina in a Quinnipiac poll isn't impressive. Being down 4 against Trump in Ohio isn't either...Interesting map..

[Erich Maria Remarque]

This takes NC to slight Rep on the Atlas map.

Q-poll takes it right back

https://twitter.com/ForecasterEnten/status/773962753778323456
Harry Enten from 538 seems to think Suffolk has it weighted better than Quinnipiac, 69% white vs Q's 64% white (it was 71% in 2012).

I expect red hacks to be as unskew'ish as they were towards CNN

[Matty]

NC electorate in 2012 was 71% white, this poll expects 64%.

No way in hell that happens.

[Erich Maria Remarque]

Considering that Quinnipiac never produces results that are overly Democratic, these are good results.

https://twitter.com/ForecasterEnten/status/773962753778323456
Harry Enten of 538 noted that the racial weighting seems a bit funky. 64% white, vs 71% in 2012.

I played around in Excel, and matching 2012's racial demographics is a 1 point Trump lead, and assuming 70% is a tie.

Nate Cohn (Upshot NYT) agrees:

You must be logged in to read this quote.

So shell we unskew it like CNN? Invisible Obama taught me, that demographical unskewing is not really an unskewing. Right, Invisible Obama?

With that said, unskewing was totally based on the concept of party ID and nothing else. If you are talking about the actual demographics of the election, then that is where polls can end up being wrong. I'm talking about what Gallup did when they assumed that the 2012 electorate would look like 2000. It has nothing to do with party and everything to do with demographics.

Let's do it! Or does it apply to Trump friendly polls ONLY?

[Sprouts Farmers Market ✘]

NC electorate in 2012 was 71% white, this poll expects 64%.

No way in hell that happens.

Quit unskewing. North Carolina is clearly trending in this direction. It'll be up to Trump to re-establish himself with the so-called 'educated' vote like he has in several other states...just not here yet. I don't think he will pull it off, but that's his only hope. Not changing the demographic battle. Fight with what you got.

[DrScholl]

Considering that Quinnipiac never produces results that are overly Democratic, these are good results.

https://twitter.com/ForecasterEnten/status/773962753778323456
Harry Enten of 538 noted that the racial weighting seems a bit funky. 64% white, vs 71% in 2012.

I played around in Excel, and matching 2012's racial demographics is a 1 point Trump lead, and assuming 70% is a tie.

Nate Cohn (Upshot NYT) agrees:

You must be logged in to read this quote.

So shell we unskew it like CNN? Invisible Obama taught me, that demographical unskewing is not really an unskewing. Right, Invisible Obama?

With that said, unskewing was totally based on the concept of party ID and nothing else. If you are talking about the actual demographics of the election, then that is where polls can end up being wrong. I'm talking about what Gallup did when they assumed that the 2012 electorate would look like 2000. It has nothing to do with party and everything to do with demographics.

Let's do it! Or does it apply to Trump friendly polls ONLY?

You really should learn how to spell, it shall, not shell. Better yet, just go away and never post again.

[NOVA Green]

As long as Clinton can hold Pennsylvania in the east, she should be OK as long as North Carolina and Florida is iffy at best for Trump.

It's tough to imagine a realistic map in which Trump wins nationally, but loses PA (and VA). The only one I can think of involves the unlikely sweep of NC, IA, OH, FL, NV, and NH (with a bonus ME2 if a tie-breaker is necessary)

[Erich Maria Remarque]

You really should learn how to spell, it shall, not shell. Better yet, just go away and never post again.

Will you be consistent and unskew this poll?

[GeorgiaModerate]

You really should learn how to spell, it shall, not shell. Better yet, just go away and never post again.

Will you be consistent and unskew this poll?

Are you trying to make his behavior match your expectations?  That sounds like you're unskewing it.

[DrScholl]

You really should learn how to spell, it shall, not shell. Better yet, just go away and never post again.

Will you be consistent and unskew this poll?

The poll speaks for itself.

[Erich Maria Remarque]

You really should learn how to spell, it shall, not shell. Better yet, just go away and never post again.

Will you be consistent and unskew this poll?

Are you trying to make his behavior match your expectations?  That sounds like you're unskewing it.

CONSISTENT.

Either you unskew all the polls [with "wrong" demographics] or you don't. It is that simple, bro
Othervise, it seems little bit... ehm.. pathetic. Do you really not think so

You really should learn how to spell, it shall, not shell. Better yet, just go away and never post again.

Will you be consistent and unskew this poll?

The poll speaks for itself.

But CNN

[Terry the Fat Shark]

I'm confused....I thought that Quinnipiac was supposed to be super biased and not to be trusted?

[The Other Castro]

I'm confused....I thought that Quinnipiac was supposed to be super biased and not to be trusted?

See the numerous discussions about the shift in RV to LV.

[NOVA Green]

I'm confused....I thought that Quinnipiac was supposed to be super biased and not to be trusted?

See the numerous discussions about the shift in RV to LV.

For all of the controversy regarding their shift from RV to LV, overall they are still an A- Polling firm based on 538 data and use a mix of landlines and cell phones, with an 87% accuracy rating and a slight 0.7% Rep House bias.

These are considered better than most of the post-labor day polls we have seen thus far, so although we can argue about the sample size of certain populations, there's no reason to throw these in the trash as junk polls.

[Associate Justice PiT]

I'm confused....I thought that Quinnipiac was supposed to be super biased and not to be trusted?

See the numerous discussions about the shift in RV to LV.

For all of the controversy regarding their shift from RV to LV, overall they are still an A- Polling firm based on 538 data and use a mix of landlines and cell phones, with an 87% accuracy rating and a slight 0.7% Rep House bias.

These are considered better than most of the post-labor day polls we have seen thus far, so although we can argue about the sample size of certain populations, there's no reason to throw these in the trash as junk polls.

     Quinnipiac has a solid reputation going back a decade and is among the best of the uni polls. They're a welcome breath of fresh air among the many actual junk polls we have to deal with.

[NOVA Green]

I mean, the NC poll is obviously too Democratic-friendly and I doubt that PA will be 9 points more Democratic than OH, but overall this looks pretty plausible.

Agreed that NC looks to be too Dem friendly, but makes sense that OH and FL would be close to tied with Trump's national polling improvements in the past couple weeks.

PA does look on the verge of being out of reach for Trump and rapidly on the way of becoming a VA or CO (Virtually impossible) because of White educated voters in the Philly Burbs and Republican/Indie suburban women.

[Erich Maria Remarque]

Compared to their polls from    JUL. 30-AUG. 7
Also LV:

Florida
Clinton - 43% (+0)
Trump - 43%  (+0)
Johnson - 8% (-1)

Ohio
Trump - 41% (-1)
Clinton - 37% (-7)
Johnson - 14% (+6)

Pennsylvania
Clinton - 44% (-4)
Trump - 39%  (+0)
Johnson - 9% (+2)
[/quote]

So, the shift in Florida is too Clinon-friendly, while shift in Ohio is too Trump-friendly. Othervise Clinton >>> Johnson.

It is pretty much consistent with national polls. They show, that Trump was gaining 1-2% since Khan controversy (Clinton +8-9), while Hillary was losing 3-4%. Though Johnson was very stable in nat.polls...