Hillary Clinton has a slim lead over Trump. In Georgia.

[Abbazorkzog]

We're not anywhere near a one party regime. There's been no sign that Democrats are getting close enough to have a real shot at the House yet. And Democrats would need to take a ton of Senate seats this year to not be favored to lose it in 2018 with such a favorable Republican map. Republicans also control the vast majorities of state legislatures and governorships.

Would even those be negatively affected by the political and economic connections of the Clintons? Their tentacles run deep across the geopolitical spectrum, and with a second Clinton Presidency their power and influence could increase three-fold, not to mention her war and foreign policy would be virtually unopposed no matter if it is Republicans or Democrats who control this, that or the other. A Clinton Presidency would effectively merge the two parties into one on war and foreign-policy. Again, victory for the Neocons, no matter which way you spin it.

 

[JANFU]

Is it?

He had a fairly solid lead last month (+8), and polls broken out for Texas aren't common (and thus likely to be less accurate).

Trump is now doing well mostly with older non-college educated whites, mostly men, who are already largely Republican. Hispanics in Texas have also voted for Republican candidates in substantial numbers in the last few elections, e.g. 40% for Romney, nearly 50% for Bush in 2004 - do you think Trump will do that well with that constituency?

He is plummeting with other groups that, even in Texas, lean Dem. If he starts losing white women, he's in it deep, as there isn't any other group where he is likely to pick up votes.
The Cuban portion of the Hispanic vote total in Florida has dropped significantly. The Republicans had a warp on them. Now with the Cuban portion shrinking, they are moving to HRC.

Clinton actually winning Texas does seem like a stretch. But just making Texas competitive will be a serious problem for the Trump campaign.... one among many.

HRC with up to a 4.5 with 538, RCP 2.7
http://projects.fivethirtyeight.com/2016-election-forecast/
RealClearPolitics - Election 2016 - Florida: Trump vs. Clinton
Then we have 24 Republican Senate seats up. It appears the Senate may change hands. Doubtful the House will.

 

[JANFU]

Seems highly unlikely. Texas is a step up from Arizona and Georgia. It was close to 10% more Republican than they were in 2012. Even if Clinton were to keep this level of support, I think her max is probably 2008 Obama - Indiana + Missouri, Arizona, and Georgia.

Latest from 538, and having watched them over the past few weeks or longer, HRC keep on with positive momentum. Since Trump went off the rails, her numbers keep climbing.
He has a difficult job getting those voters back.
http://projects.fivethirtyeight.com/2016-election-forecast/?ex_cid=rrpromo

Arizona is in play???

 

[Redress]

It's probably smart to not draw conclusions about how this election will end. But I do think we can start looking at what it actually is right now. Clinton's convention bounce has been bigger than any candidate's in 20 years. And Trump's support is lower than any candidate's has been at any point post-convention than any candidate since 1996.

We certainly can't claim that Trump's dead yet because things happen, but I do think these mean something.

How much is convention bounce, and how much is Trump's Very Bad Week? How long will the bounce last, and how far down will they sink once the bounce ends? Just call me Debbie Downer, and am not playing along with any of the storylines any more, they all get weird in the end. I predict a Gainax Ending to this election.

 

[Mycroft]

Here we go again...

The liberal lemmings see a doctored poll and think it reflects reality.

I don't know which is more pathetic...that people blindly believe this stuff...or that the crooked pollsters believe they can get away with it.

Busted! Crooked Media Plays With Polling Numbers To Pretend Georgia Is Hillary Country

 

[roughdraft274]

Here we go again...

The liberal lemmings see a doctored poll and think it reflects reality.

I don't know which is more pathetic...that people blindly believe this stuff...or that the crooked pollsters believe they can get away with it.

Busted! Crooked Media Plays With Polling Numbers To Pretend Georgia Is Hillary Country

Lol yea yea yea the polls are always wrong. You're missing the point of the thread. Trump needs Georgia to be a sure thing so he doesn't have to spend time there. That it's even close is troublesome for him.


Sent from my iPhone using Tapatalk

 

[Mycroft]

Lol yea yea yea the polls are always wrong. You're missing the point of the thread. Trump needs Georgia to be a sure thing so he doesn't have to spend time there. That it's even close is troublesome for him.


Sent from my iPhone using Tapatalk

How do you know it's close?

I mean, it's shown that the poll is bogus. Who knows...it could be that Trump has a 10 point lead.

But, you see, that's the whole purpose of a bogus poll...to sow doubt. To cause people to think things are troublesome for Trump.

Are you one of those people who blindly believe this stuff?

 

[Helix]

i doubt that Trump will lose a solidly red state like Georgia regardless of what the polls say right now.

 

[Anagram]

Here we go again...

The liberal lemmings see a doctored poll and think it reflects reality.

I don't know which is more pathetic...that people blindly believe this stuff...or that the crooked pollsters believe they can get away with it.

Busted! Crooked Media Plays With Polling Numbers To Pretend Georgia Is Hillary Country

That article is fundamentally flawed. Good pollsters don't weight by Party ID. They weight by demographic factors that aren't dynamic. Trying to re-weight polls by Party ID retroactively is what Unskewed Polls did in 2012 and what the Daily Kos did in 2014. The polls were right both times, and the unskewers were wrong.

And although in September 2014 they did show that the race was tied, the AJC's final poll of that cycle got almost the exact margin.

 

[Mycroft]

That article is fundamentally flawed. Good pollsters don't weight by Party ID. They weight by demographic factors that aren't dynamic. Trying to re-weight polls by Party ID retroactively is what Unskewed Polls did in 2012 and what the Daily Kos did in 2014. The polls were right both times, and the unskewers were wrong.

And although in September 2014 they did show that the race was tied, the AJC's final poll of that cycle got almost the exact margin.

You are saying one thing...and then saying another thing.

1..Good pollsters don't weight by Party ID.

2. Weighting by Party ID was done in 2012 and 2014...and was a good thing.

You should make up your mind.

In any case, weighing by Party ID isn't a bad thing, as long as it's done accurately and honestly. That was the point of the article. It showed how the pollster weighed dishonestly in order to give Hillary a bogus advantage.

 

[Anagram]

You are saying one thing...and then saying another thing.

1..Good pollsters don't weight by Party ID.

2. Weighting by Party ID was done in 2012 and 2014...and was a good thing.

You should make up your mind.

In any case, weighing by Party ID isn't a bad thing, as long as it's done accurately and honestly. That was the point of the article. It showed how the pollster weighed dishonestly in order to give Hillary a bogus advantage.

No I'm not. I'm saying that people didn't weight by party ID and that it has been accurate in 2012 and 2014. What Unskewed Polls, the Daily Kos, and your article did is take the reported Party ID numbers and tried to weight them. That created inaccurate results.

This is because Party ID is a self reported, dynamic characteristic. Unlike, race, age, or gender it changes throughout the cycle depending on how people are feeling about the candidates. Weighting by it, necessarily creates inaccurate results and that's why pollsters don't do it.

 

[Mycroft]

No I'm not. I'm saying that people didn't weight by party ID and that it has been accurate in 2012 and 2014. What Unskewed Polls, the Daily Kos, and your article did is take the reported Party ID numbers and tried to weight them. That created inaccurate results.

This is because Party ID is a self reported, dynamic characteristic. Unlike, race, age, or gender it changes throughout the cycle depending on how people are feeling about the candidates. Weighting by it, necessarily creates inaccurate results and that's why pollsters don't do it.

Dude...the pollster weighted their numbers based on Party ID. The article didn't weight anything. They simply pointed out what the pollster did.

Furthermore, almost ALL pollsters nowadays weight by Party ID. When they do it honestly and accurately, you get a good poll. When they do it dishonestly, you get this Georgia poll.

Polling firms contact registered voters. The number of voters who answer the phone and agree to the survey are put into a number called the “unweighted base.” The polling firm then applies their own “formula” based on who they think is going to vote. In the example below, 278 Republicans in Georgia answered the phone for a poll in May. The AJC’s polling firm thinks that’s too many Republicans, so they move that to 245, “adjusting” the numbers with an “average weighted response.”

2 1/2 months, later, the same polling firm substantially altered their “secret formula.”

 

[Anagram]

Dude...the pollster weighted their numbers based on Party ID. The article didn't weight anything. They simply pointed out what the pollster did.

Furthermore, almost ALL pollsters nowadays weight by Party ID. When they do it honestly and accurately, you get a good poll. When they do it dishonestly, you get this Georgia poll.

The AJC did not weight by party ID. You're seeing the numbers change, buy you aren't understanding why they're changing. They weight by age and race and gender and that changes the Party ID numbers. And you are patently wrong that almost all posters weight by Party ID.

State Level Party ID - Public Policy Polling PPP

Margin of Error: Quinnipiac believable? Quinnipiac

Are Pollsters Oversampling Republicans? | FiveThirtyEight Gallup and Survey USA

Are Polls Skewed Too Heavily Against Republicans? - The Atlantic Marist
Why & How Pollsters Weight (excellent read for those interested in polls) CBS/CNN/ABC/Newsweek/Time/Pew

 

[Redress]

Here we go again...

The liberal lemmings see a doctored poll and think it reflects reality.

I don't know which is more pathetic...that people blindly believe this stuff...or that the crooked pollsters believe they can get away with it.

Busted! Crooked Media Plays With Polling Numbers To Pretend Georgia Is Hillary Country

Holy ****, some one actually used Gateway Pundit for a source....

Hint: their reputation for accuracy is so bad, the Enquirer looks good by comparison.

 

[roughdraft274]

How do you know it's close?

I mean, it's shown that the poll is bogus. Who knows...it could be that Trump has a 10 point lead.

But, you see, that's the whole purpose of a bogus poll...to sow doubt. To cause people to think things are troublesome for Trump.

Are you one of those people who blindly believe this stuff?

Because there's more than one poll. It's much closer than it should be. Complain about polls all you want but they are currently the best indicators we have and have very good track records when averaged etc.


Sent from my iPhone using Tapatalk

 

[Mycroft]

The AJC did not weight by party ID. You're seeing the numbers change, buy you aren't understanding why they're changing. They weight by age and race and gender and that changes the Party ID numbers. And you are patently wrong that almost all posters weight by Party ID.

I don't believe you actually looked at the numbers in that article. They were clearly labeled "Republican" and "Democrat". Party ID's.




Now...if you happen to know the pollster's criteria, please link it.

 

[Mycroft]

Holy ****, some one actually used Gateway Pundit for a source....

Hint: their reputation for accuracy is so bad, the Enquirer looks good by comparison.

shrug...

If you have something credible that's disputes their data, let's see it.

 

[Kobie]

I mean, it's shown that the poll is bogus. ?

No, it wasn't. The Gateway Pundit is simply making stuff up, using its' readers fundamental misunderstanding of polling methodology to score partisan points.

But please, do continue to repeat 2012 and insist the polls are all "skewed" and that a Trump landslide is imminent. Worked so well for you guys last time.

 

[Anagram]

I don't believe you actually looked at the numbers in that article. They were clearly labeled "Republican" and "Democrat". Party ID's.

View attachment 67205274

They ask them their Party ID but they don't weight by it. If you notice the proportions of men to women, whites to blacks, and the age groups are almost exactly the same in the two polls. That's because they weighted by them. The Party ID proportions changed, because it's literally changed in Georgia since the other poll. That's not AJC weighting.

Party ID is self reported and it fluctuates wildly throughout Presidential elections as voters feelings about the candidates and the parties change. This is exactly why good pollsters, and by now almost all pollsters don't weight for it.

And it's been vindicated. The only prolific pollster who weighted by party ID the last few cycles was Rasmussen. They missed badly and now barely poll at all anymore, and when they do they no longer weight by party id. Unskewed Polls and the Daily Kos tried to retroactively weight polls by Party ID and they weren't even close.

 

[Mycroft]

Because there's more than one poll.

Well, cool. Present these other Georgia polls. The OP only presented that bogus poll.

I'd be quite interested in seeing others. Maybe THEY won't be bogus.

 

[Mycroft]

The Party ID proportions changed, because it's literally changed in Georgia since the other poll.

LOL!!

In THREE FREAKING MONTHS???

It doesn't fluctuate THAT wildly.

Dude...find someone else to sell your bridge to.

 

[Anagram]

LOL!!

In THREE FREAKING MONTHS???

It doesn't fluctuate THAT wildly.

Dude...find someone else to sell your bridge to.

It really, really does especially early on. This has been demonstrated for years, and is why almost no pollsters weight it anymore. It's something that doesn't necessarily seem like it's true at first glance, but has been verified with MOUNTAINS of data over years and years.

 

[roughdraft274]

Well, cool. Present these other Georgia polls. The OP only presented that bogus poll.

I'd be quite interested in seeing others. Maybe THEY won't be bogus.

He mentioned them in the post of I'm not mistaken. Pretty much all polling is readily available at rcp for anyone that cares. Or go to 538 for their predictions.

Regardless the poll isn't bogus just cause your ****ty link says so.


Sent from my iPhone using Tapatalk

 

[Mycroft]

It really, really does especially early on. This has been demonstrated for years, and is why almost no pollsters weight it anymore. It's something that doesn't necessarily seem like it's true at first glance, but has been verified with MOUNTAINS of data over years and years.

Okay. Then show me the data that proves it happened in Georgia this year.

 

[Mycroft]

He mentioned them in the post of I'm not mistaken. Pretty much all polling is readily available at rcp for anyone that cares. Or go to 538 for their predictions.

Regardless the poll isn't bogus just cause your ****ty link says so.


Sent from my iPhone using Tapatalk

The 538 predictions that JANFU presented give Trump the edge. The other link he gave was about Florida.

If you dispute the link I gave, make your case.