Quinnipiac National: Clinton +7/+10

[Erich Maria Remarque]

Were junk when it was showing Trump up!

Getting good now

[Erich Maria Remarque]

men:
Trump 42%
Clinton 35%
Johnson 14%
Stein 6%

women:
Clinton 54%
Trump 33%
Johnson 6%
Stein 3%

Intresting. In head-to-head there are 10% of men (5% someone else + 5 DK/NA) and just 4% of women (3% + 1%) that don't vote Trump/Hillary.
In 4 way there are 23%!!! of men (14% Johnson + 6% Stein + 3 DK/NA) and just 12% of women (6% +3% + 3%) that don't vote Trump/Hillary.

It is hard to believe that men vill vote in such proportions for third party candidates 

[Erich Maria Remarque]

In most years a likely-voter screen 'favors' whiter, older, more educated people, and that means Republicans almost as a rule. Not this year!

This year as well!

[Erich Maria Remarque]

Hispanics will be at least 10% of the electorate(as 2012 had that) and probably 11 or 12 being as they'll likely be pissed off this time. Whites been dropping 2% each national presidential election since 1980, so 70% looks likely.

http://ropercenter.cornell.edu/polls/us-elections/how-groups-voted/how-groups-voted-2012/

I heard that a lot of whites gonna be voting for Trump because they are pissed off.

[Erich Maria Remarque]

So Quinnipiac isn't junk anymore?

Did you ever listen to why some of us had problems with QU? It wasn't a blanket statement.

I did!

QU continues to be ridiculous ...

But now it suddenly stopped

[Erich Maria Remarque]

So Quinnipiac isn't junk anymore?

When was it junk?

When Dems didn't like its result

[Erich Maria Remarque]

So Quinnipiac isn't junk anymore?

Did you ever listen to why some of us had problems with QU? It wasn't a blanket statement.

I did!

QU continues to be ridiculous ...

But now it suddenly stopped

So, you obviously didn't.

What did change? Besides RV -> LV

[Erich Maria Remarque]

Ok.... not sure all of the pros and cons on the "Q-Poll" argument, but Nate Silver/ 538 gives them an overall A- rating, which is actually pretty decent for polling organizations.

Is there something I'm missing here that y'all are arguing about?

This all started WAY back in October/November, which seems like a lifetime ago. QU's national and some state polls had a demographic model more in line with a 2004 electorate (which is how Gallup came undone in 2012) with significant increases in the white vote, and pretty sizeable drops in the Hispanic vote (ie, dropping about 1/3). Part of the reason why QU has had much worse numbers for Clinton, generally, this season was their demographic model. Now, combined with a LV screen and a much more likely demographic model they've moved more in line with the conventional wisdom.

It's cute to respond to those issues with "something something, hack "

They're the reason why I've treated QU with a pinch of salt for this season.

Nice try, my uneducated hack!

Firstly, you seem not understand how random polling works. The crosstabs can generally not be perfect even after weightening. You cannot do anything about, but increase sample size, which is costly.

Secondly, QU does not report demographic before weighteing.

Thirdly, compare with Pew latest poll, they've got White 73%, Blacks 12%, Hispanics 8% vs QU's 73%, 11%, 8%. You didn't complain. There are a lot of other polls with "strange" crosstabs. But only those that shows Trump-friendly numbers've got screened. Rasism!

Latest RV from Marist, having C+12. Whites 71%, Blacks 10%, Hispanics 12%, Other 7%. Hispanics at 12%? No one quetiones their demographic model.

Then there are pollster that have Whites at 69% (73%-74% in 2012), no one cares.

But QU is junk!!!!!

[Erich Maria Remarque]

Let me translate into your language

Btw did you see me praise this poll? I said it's now in line with conventional wisdom, not that I trust it.

Then the only problem with your statement, that this is not true

How should cross tabs look like according to conventional wisdom? You talking about cross tabs, right?

% of total voters cast
2012: Whites 73.7%, Blacks 13.4%, Hispanics 8.4%
2008: Whites 76.4%, Blacks 12.3%, Hispanics 7.4%
2004: Whites 76.0%, Blacks 11.1%, Hispanics 6.0%
2000: Whites 80.7%, Blacks 11.7%, Hispanics 5.4%
https://www.census.gov/prod/2013pubs/p20-568.pdf

Note, that 2012 were pretty abnormal, since amount of Whites votes cast dicrease with 2mln, even though amount of Whites increased. Does conventioal wisdom assume that 2016 will follow 2012-trend?

[Erich Maria Remarque]

Let me translate into your language

Btw did you see me praise this poll? I said it's now in line with conventional wisdom, not that I trust it.

Then the only problem with your statement, that this is not true

How should cross tabs look like according to conventional wisdom? You talking about cross tabs, right?

% of total voters cast
2012: Whites 73.7%, Blacks 13.4%, Hispanics 8.4%
2008: Whites 76.4%, Blacks 12.3%, Hispanics 7.4%
2004: Whites 76.0%, Blacks 11.1%, Hispanics 6.0%
2000: Whites 80.7%, Blacks 11.7%, Hispanics 5.4%
https://www.census.gov/prod/2013pubs/p20-568.pdf

Note, that 2012 were pretty abnormal, since amount of Whites votes cast dicrease with 2mln, even though amount of Whites increased. Does conventioal wisdom assume that 2016 will follow 2012-trend?

Why the hell wouldn't it?

One of the best things about no longer being a mod is I can use the ignore button, as I'm about to do, on posters like you who border between troll and useless.

Because it usually does not. The trend on average follows the demographic change.

The words hurt, you know I will miss you with all my heart