ALL BLOG POSTS AND COMMENTS COPYRIGHT (C) 2003-2016 VOX DAY. ALL RIGHTS RESERVED. REPRODUCTION WITHOUT WRITTEN PERMISSION IS EXPRESSLY PROHIBITED.

Thursday, October 20, 2016

The national polls: stretch run

Notice something very informative about the wildly divergent national polls?
  • Trump vs. Clinton vs. Johnson vs. Stein Quinnipiac Clinton 47, Trump 40, Johnson 7, Stein 1 Clinton +7
  • Trump vs. Clinton vs. Johnson vs. Stein IBD/TIPP Clinton 40, Trump 41, Johnson 8, Stein 6 Trump +1
  • Trump vs. Clinton vs. Johnson vs. Stein Bloomberg Clinton 47, Trump 38, Johnson 8, Stein 3 Clinton +9
  • Trump vs. Clinton vs. Johnson vs. Stein Economist/YouGov Clinton 42, Trump 38, Johnson 6, Stein 1 Clinton +4
  • Trump vs. Clinton vs. Johnson vs. Stein Reuters/Ipsos Clinton 42, Trump 38, Johnson 6, Stein 2 Clinton +4
  • Trump vs. Clinton vs. Johnson vs. Stein Rasmussen Reports Clinton 42, Trump 42, Johnson 7, Stein 1 Tie
The number of percentage points being allocated to Johnson and Stein is between 7 and 14. In 2012, the combined Libertarian/Green vote was 1.35 percent. In 2008, it was 0.96 percent.

Now, let's be generous and pretend that the combined Libertarian/Green vote will be 2 percent, which would be a 48 percent increase from 2012 and 108 percent from 2008. That means that the national polls are, at a minimum, off by between 5 and 12 points.

I can't say that there is evidence of a Trumpslide at this point. On the other hand, I can't take seriously the evidence that suggests Hillary Clinton is going to win by Mondalean proportions either. The trick, I think, will be to watch what happens as the polls go into the final week. If they tighten dramatically, that means the pollsters have been playing games and are attempting to cover themselves, which suggests Trump will win. That's what happened with Brexit.

If, on the other hand, the polls continue to indicate significant leads for Hillary, that means they are not concerned about their accuracy and will tend to suggest a Hillary win, albeit a little closer than they're predicting.

UPDATE: Trump appears to have revived his momentum again. Remember, he pushes, then coasts, pushes, then coasts. It looks like he's gearing up for the stretch run.
It’s too early to measure the impact of last night’s final presidential debate, but Republican Donald Trump now has a three-point lead nationally on Democrat Hillary Rodham Clinton. The latest Rasmussen Reports national telephone and online White House Watch survey finds Trump with 43% support among Likely U.S. Voters to Clinton’s 40%. 

Labels: ,

224 Comments:

1 – 200 of 224 Newer› Newest»
Anonymous BGKB October 20, 2016 7:40 AM  

Good Christian woman Donna Brazil say Russia faked the emails, and if you read the fake emails you a thief.

https://twitter.com/FoxNews/status/788950903545958400

Blogger residentMoron October 20, 2016 7:45 AM  

Nice bit of doublethink there. Would that then be. DoubleThoughtCrime?

Anonymous Cantostop October 20, 2016 7:45 AM  

What I don't understand (and would like someone to explain please) is why, with all the media bias and corruption, well known for months if not decades, why hasn't some enterprising individual set up a genuinely unbiased poll (ie not "playing games") business and explained in detail the methodology? Can't imagine yuge overhead... No profit perhaps? That doesn't seem to stop O Keefe and co. though.

Blogger Josh (the gayest thing here) October 20, 2016 7:46 AM  

What I don't understand (and would like someone to explain please) is why, with all the media bias and corruption, well known for months if not decades, why hasn't some enterprising individual set up a genuinely unbiased poll (ie not "playing games") business and explained in detail the methodology? Can't imagine yuge overhead... No profit perhaps?

Polling is very expensive

Anonymous CloseHauled October 20, 2016 7:49 AM  

@1 The emails were stolen and then faked? Damn Russians

Blogger Chent October 20, 2016 7:50 AM  

Polling is very expensive

Maybe a crowdfounding campaign could help...for the next election

Anonymous Pereira October 20, 2016 7:51 AM  

>>Now, let's be generous and pretend that the combined Libertarian/Green vote will be 2 percent, which would be a 48 percent increase from 2012 and 108 percent from 2008.

Why shouldn't it be closer to what the polls are actually showing? The last election is a poor predictor of this one as far as third party candidates are concerned, for the basic reason that Obama and Romney didn't have nowhere near the same negative ratings as Trump and Clinton. Never have two candidates been so disliked, meaning it's reasonable to expect third party candidate voting to be much higher than in 2008 and 2012.

Blogger Josh (the gayest thing here) October 20, 2016 7:52 AM  

The number of percentage points being allocated to Johnson and Stein is between 7 and 14. In 2012, the combined Libertarian/Green vote was 1.35 percent. In 2008, it was 0.96 percent.

Now, let's be generous and pretend that the combined Libertarian/Green vote will be 2 percent, which would be a 48 percent increase from 2012 and 108 percent from 2008. That means that the national polls are, at a minimum, off by between 5 and 12 points.


Another possibility is that unfavorability drives up the third party vote (which this year appears to be more of a "none of the above" place holder).

In 2012, Obama was +4.2 and Romney was +4.8

Clinton is at -10 and Trump is at -26

So in addition to a huge uptick in the third party polling, you have a huge downtick in the favorable ratings.

Anonymous #3254 October 20, 2016 7:54 AM  

Yeah, the amount of votes supposedly going to 3rd party candidates is laughable.

Then again, could be that people are answering the polls that way. I can't believe the amount of "a pox on both their houses" I'm seeing on Facecrook. It makes people feel better to fancy themselves above such a thing as voting for a major party candidate. But they do when the curtain is drawn.

This is the first time in a long time I've enthusiastically voted for a major party candidate. Hope the MAGA mindset puts Trump over the top.

Blogger Basil Makedon October 20, 2016 7:55 AM  

I think this is a good point. There is no way that third parties (and fourth parties) will do this well on Nov. 8. So those points will necessarily be redistributed elsewhere. Where they go, is unknowable. However, "shy" Trump voters -- people who lie to pollsters, etc. seem more reasonable and likely to me than "shy" Clintonistas.

Blogger OGRE October 20, 2016 7:57 AM  

If you go in and look at the sample data for pretty much every single poll, they are taking a very large sample of democrats over republicans and independents.

For example, the Reuters poll has a sample size of 1750 with 723(41.3%) Dem, 643 (36.7%) Rep, and 210 (12.0%) Ind. The poll has Clinton up 43 to 39. But it has Indies breaking for Trump 39% to 24%. How in the world can she be up 4% overall if he has a 15 pt lead in independents?? According to Gallup's most recent data on party affiliation (Sept. 2016), the electorate breaks down as 32% Dem, 27% Rep, and 40% Indie. If you weight the Reuters poll according to the Gallup info, then Trump wins big.

Other people have shown this in greater depth than I have. Theres also tendency for polls to give more weight to other factors that favor Clinton, such as sampling more female than male voters, or higher percentages of college educated voters.

Anonymous Gedrin October 20, 2016 7:58 AM  

I could certainly set something like that up. The problem is where is the profit coming from?

As folks have said, it's enormously expensive. You have to hire telemarketing firms to make calls, then there's software to perform statistical analysis, etc.

I just don't see how you make money doing it.

Blogger Robert What? October 20, 2016 7:58 AM  

I'm not saying anything that most of you don't already know, but most of the media polls are for the specific purpose of making Trump voters think there is no point to getting out and voting.

Blogger VD October 20, 2016 7:59 AM  

Why shouldn't it be closer to what the polls are actually showing?

Because that has never happened before sans a figure of national standing like Wallace or Perot. Also, because Johnson was the Libertarian candidate in the last election. I expect he'll actually do worse than he did in 2012, and will be below one percent. I've actually voted Libertarian several times in the past and I don't know a single libertarian who is enthusiastic about him.

Blogger Basil Makedon October 20, 2016 8:00 AM  

@7 ans @8 The reason is that regardless of unpopularity when you have a geographic based winner-take-all system (as we do) a two-party dynamic always results. Voting for third parties results in wasted votes.

In US history, you start getting third party votes when one of the major parties (basically the right leaning party) starts to break-down -- see Federalists, Whigs. Perhaps the GOP is breaking down, but that seems strange considering it has not held as many state, local and federal offices since the 1920s as it does right now.

Anonymous BGKB October 20, 2016 8:00 AM  

The emails were stolen and then faked? Damn Russians

The Russians are so stupid they are hacking to help TRUMP when TRUMP wants to nuke Russia.

So in addition to a huge uptick in the third party polling, you have a huge downtick in the favorable ratings.

I will have to stop by the hippie places and tell them all of the voting is going to be done in strip clubs this year. Will the stoners go out to the poles for the free pot candidate?

Blogger Stilicho October 20, 2016 8:05 AM  

@3 there is a market for accurate polls and they do exist, but they are the internal polls commissioned by the candidates.

Another issue is the relative sample sizes. Every poll with Clinton ahead that I've seen has a ten point or greater bias in sample size in favor of democrats which is typically a 25 to 35 percent advantage in the final result. In other words, they will ask 45 democrats, 35 republicans, and 20 independents. That simply doesn't reflect the voter breakdown in America, nor does it reflect the dynamics of this"change" election. If anything, I would expect more independents this year as well as rough parity between dems and republicans on the order of 35 dems, 35 repubs, and 30 independents. I would like to see a poll with those sample ratios.

Anonymous CloseHauled October 20, 2016 8:05 AM  

They admit emails were stolen. But then call them fake? WAT?

I know a couple of NeverTrumpers that say they are voting for Johnson ... but guaranteed when they get to the poll station they vote trump when no one is looking because they HATE Hillary more than they do Trump.

Blogger Paul October 20, 2016 8:06 AM  

There will be no Trumpslide, but the Shy Trump Vote + the disaffected Bernie Bros + the air in the third party polling + last minute undecideds could easily exceed current poll margins for Hillary.

This is going to be an election the old fashioned way, without ultrasound. We're just going to have to wait and see whether it's a boy or a girl.

Blogger pyrrhus October 20, 2016 8:08 AM  

https://theconservativetreehouse.com/2016/10/19/anti-gaslighting-raw-and-current-state-vote-data-not-remotely-close-to-media-polling/

Anonymous verryfunnyminion October 20, 2016 8:08 AM  

O/T, BGKB do you Gab? If so, what's your handle?

Blogger Timmy3 October 20, 2016 8:10 AM  

Looks like a narrow win for either candidate. A nail biter finish.

Blogger Whisker biscuit October 20, 2016 8:13 AM  

Interesting that Trump is so unfavorable, but he's packing arenas, and set a record for GOP turnout in the primaries.

People are missing YUGE chunks of registered voters like myself who sat out previous elections due to shitty candidates that will be voting this year.

Nothing about this election is black and white; no amount analyzation can predict what will happen.

Anonymous EH October 20, 2016 8:21 AM  

BGKB wrote:[....]
I will have to stop by the hippie places and tell them all of the voting is going to be done in strip clubs this year. Will the stoners go out to the poles for the free pot candidate?


Very punny. A nude low, in f*ckt.

Blogger dh October 20, 2016 8:21 AM  

If they tighten dramatically, that means the pollsters have been playing games and are attempting to cover themselves, which suggests Trump will win. That's what happened with Brexit.

This has to be the dumbest take on polling in the entire season, for at least two reasons. Polling tightens closer to the election because undecideds make up their mind and pick a candidate.

Secondly, national polling is totally irrelevant at this point. State polling is the only thing that matters, for at least the last 30 days, and certainly until the election. Trump has a large problem in state-based polling, just like McCain and Romney did. It's more likely that Trump will lose Texas than it is he will win the election.

Notice something very informative about the wildly divergent national polls?

You do have a good point here, which is it would be earth shattering of the Stein/Johnson votes are multiples of what they were in the past. No one polling really knows what will happen. There have been cases with them polling this high before, but it always evaporates on election day, usually because of the vastness of the "most important election evar" talk from both sides. There has been no shortage of that here, yet the numbers are still much higher than past years. So most large pollsters are still polling 2-way and 4-way match ups.

There is very little discussion, with people who are relying on facts and the mathematical look of the race, to suggest that Clinton is looking at a triumph. She will be the least popular day #1 President since Ford, at least, and potentially ever. She will very likely be the first President in memory to be sworn in at near or below 50% approval (in contrast, Pres. Obama in his first term was in the 80's). There is no doubt that she thanks her lucky stars to be running against Trump in this election. As her campaign managers leaked email's show, they were always more worried about Rubio than anyone.

Blogger Brian H October 20, 2016 8:22 AM  

1) It's such an unconventional election year, and also a change year. One candidate was genuinely loathed, and the media has managed to paint the other as loathsome. I'm willing to believe that the 3rd parties will legitimately pull more than usual, just don't know how much

2) it's scarier to me if 3rd party support DOES collapse, because I sense it is mostly my dimwit millenial peers wanting to posture as sophisticated and "above" it all. Hard to see them veering to Trump over the media narratives. But maybe I am wrong.

Blogger James Dixon October 20, 2016 8:25 AM  

From Trump +1 to Clinton +9. The normal margin of error is plus or minus 5%, so those are the extreme ends for both of them to be right if Clinton actually wins by 4%.

> The number of percentage points being allocated to Johnson and Stein is between 7 and 14. In 2012, the combined Libertarian/Green vote was 1.35 percent. In 2008, it was 0.96 percent.

But Anderson took something like 5% and Perot took over 19%. A large third party vote is not unprecedented.

> If, on the other hand, the polls continue to indicate significant leads for Hillary, that means they are not concerned about their accuracy and will tend to suggest a Hillary win,

They're not concerned about their accuracy. As I said in the earlier thread, they know this is the last election that will matter. The media has completely blown their cover and no longer cares if the public knows if they're tools of the democratic party or not. The polls are almost entirely propaganda tools at this point.

> why hasn't some enterprising individual set up a genuinely unbiased poll

One, there's no such thing as an unbiased poll. Two, polling is hard (and as others have noted, expensive).

Anonymous MEC October 20, 2016 8:26 AM  

The determining factor could be the success or failure of the Dem's fraud works

Anonymous Pennywise October 20, 2016 8:27 AM  

Stilicho..

"Another issue is the relative sample sizes. Every poll with Clinton ahead that I've seen has a ten point or greater bias in sample size in favor of democrats...'

Which ones in particular? Please cite. Thanks.

Anonymous #3254 October 20, 2016 8:28 AM  

dh: This has to be the dumbest take on polling in the entire season, for at least two reasons. Polling tightens closer to the election because undecideds make up their mind and pick a candidate.

Why would this always result in tightening of the race? If undecideds make up their minds in proportion to the folks who have already made up their minds, wouldn't the absolute difference between the candidates remain the same?

As far as 3rd party candidates go, presumably Johnson poseurs will tend to break for Trump, while Stein poseurs will break for Clinton. That would tend to favor Trump in terms of votes gained.

Blogger dh October 20, 2016 8:29 AM  

Interesting that Trump is so unfavorable, but he's packing arenas, and set a record for GOP turnout in the primaries.

People are missing YUGE chunks of registered voters like myself who sat out previous elections due to shitty candidates that will be voting this year.


Trump set two records in the primary: the most votes for him, and the most votes against him.

https://www.washingtonpost.com/news/the-fix/wp/2016/06/08/donald-trump-got-the-most-votes-in-gop-primary-history-a-historic-number-of-people-voted-against-him-too/

This isn't a good thing. Yes, he won, but he was the most voted against candidate. There is no spin that makes this good.

As far as you new Trump voters, you are being swamped. The Democrats brown machine hoped to register huge new numbers of voters, but fell very far short. Here is a good article:

https://www.washingtonpost.com/politics/did-donald-trump-cause-an-outsized-surge-in-hispanic-voters-probably-not/2016/10/13/5a197cba-909b-11e6-a6a3-d50061aa9fae_story.html

Instead of registered 3 million new brown voters, they "only" hit hundreds of thousands, La Raza only registered 50,000; they only registered 25,000 new Dominican voters in New York.

These numbers, which are disappointing to the Democrats, are still huge numbers. They swamp you. And they are often in swing states like Florida and Pennsylvania.

There has been no huge surge in new registrations for Republicans. There is no new silent majority for Trump.

Blogger dh October 20, 2016 8:31 AM  

Why would this always result in tightening of the race? If undecideds make up their minds in proportion to the folks who have already made up their minds, wouldn't the absolute difference between the candidates remain the same?

That's point, they don't make up their mind in equal proportion, they make up their mind in proportion to the candidate who is going to win. For Trump to win, undecideds at this point, who are relatively few, would have to break in unprecedented levels, hidden from polling that try's to push you to announce a preference, towards Trump. It's not happening yet.

Polls always tighten the closer to an election. I would challenge any to find any case of it ever getting more undecided the closer to an election. I can't think of a single race I've ever followed, at any level, ever that has gotten looser the closer to election day. It. Always. Tightens. There isn't even a plausible reason to explain why it wouldn't.

Anonymous The OASF October 20, 2016 8:34 AM  

@22

"Looks like a narrow win for either candidate. A nail biter finish."

It's very simple: if Trump gets a mammoth amount of votes like he did in the primaries, they won't be able to rig it and he'll win. If he doesn't get quite enough then they can and will rig it.

The only way Killary wins is via a rigged election. Period.

Internal scientifically-driven polls show Trump ahead by 15, 20, 30 points. Internet polls show Trump ahead by even more. I laugh hilariously when trendy millennials point out that internet polls aren't scientific. HA! As if the MSM polls are!

The problem is the outdated and always-was-pointless electoral college system. The DNC just has to rig a few Counties in OH, PA, & FL and maybe NC to trigger an electoral landslide for Killary.

It will be quite interesting.

Anonymous WinstonWebb October 20, 2016 8:35 AM  

There has been no huge surge in new registrations for Republicans.

False

Blogger Student in Blue October 20, 2016 8:35 AM  

There isn't even a plausible reason to explain why it wouldn't.

Here's a plausible reason. I've never seen so many people on the left triggered, literally triggered, so hard by Trump. He activates their amygdala in a way that makes a lot of them physically ill, so they think he is literally worse than the devil and Hitler combined.

So when you have emotional diarrhea like that happening, it's far more than plausible that the side that produces the SJWs will double down.

Is that what's happening? I have no solid proof, but it is entirely plausible.

Blogger VD October 20, 2016 8:37 AM  

This has to be the dumbest take on polling in the entire season, for at least two reasons. Polling tightens closer to the election because undecideds make up their mind and pick a candidate.


That's cute. You're so innocent, dh. As Scott Adams says, no one is "undecided" at this point, except perhaps on whether to vote or not. They're not oversampling Democrats by accident.

It's more likely that Trump will lose Texas than it is he will win the election.

As always, we'll see. I don't trust any narrative that relies upon a media institution's honesty. That being said, I'd feel a lot more confident about Trump winning if he was doing better in the betting markets. I don't discount the possibility that they, too, are corrupted, but we'll see.

After all, the odds against Leave were 3/1 the day before. Trump currently stands at 4/1.

Blogger Michael Thompson October 20, 2016 8:37 AM  

Almost every poll that features all 4 candidates that has a companion poll featuring a head-to-head matchup featuring only Clinton and Trump sees her lead over him widen by a couple percentage points in that companion poll. I think there's pretty good supporting evidence to suggest that the third party candidates are pulling more from her than from him at this point and that if the third parties do collapse down to historically normal numbers, it will be more helpful to her than to him.

Anonymous Undocumented Civilizationalist October 20, 2016 8:38 AM  

Trump will win Ohio. That is just based on the vibe and feel I get from my state. But, I think he wins it.

Absentee ballots seem to indicate he is looking good in NC and FL--although FL can be dicey with all the ballot shenanigans and extended voting registration things going on there right now.

I think this election comes down to Colorado and Nevada. If trump gets both while holding FL and NC and he wins. But Colorado is a tough one...

Anonymous #3254 October 20, 2016 8:40 AM  

That's point, they don't make up their mind in equal proportion, they make up their mind in proportion to the candidate who is going to win.

If the undecideds break toward the candidate who they think is going to win - that is, the one ahead in the polls, that would suggest a widening of the lead. But I don't want to belabor this, so that's all I'll say on that.

Your Washington Post article trying to say that Trump also got the most votes AGAINST him sounds scary, but we need to assume that these Republican primary voters who voted against him will not vote for him in November. I doubt it, but we'll see. Point is that Trump inspired lots of folks to vote one way or the other. There'll be a few Bill Kristols who will vote for Clinton, but most of these #NeverTrumpers will fold like the cheap card tables they are.

Blogger Brian H October 20, 2016 8:40 AM  

The OASF wrote:The DNC just has to rig a few Counties in OH, PA, & FL and maybe NC

With all of the "rigged" talk coming from DJT, I wonder if anyone in his camp thought about amassing an army of watchers in these regions or organizing something to counter. Not sure whether there is much that can be done, but if you could affect the outcome in just a few of those key spots, it sure could go a long way.

Blogger Ransom Smith October 20, 2016 8:41 AM  

Has anyone seen the internal polls that apparently Gen. Flynn has been involved with? Seems to indicate a Trump win assuming they're accurate.

Blogger Michael Thompson October 20, 2016 8:41 AM  

"Trump will win Ohio. That is just based on the vibe and feel I get from my state. But, I think he wins it."

I'm pretty sure he's going to win Ohio as well. The polling in the state certainly leans that way. However, I doubt that Ohio is going to be a bellwether state this time. It looks to be the outlier among all the states that have historically been swing states. In fact, I think there is a perception that this election is actually closer than it is because Trump might win Ohio. In the end, winning Ohio isn't going to matter much because it doesn't look like he's going to win any of the other states he needed to get to 270 and is probably going to lose a couple of ones that he was expected to hold serve in.

Anonymous The OASF October 20, 2016 8:42 AM  

@29

"Which ones in particular? Please cite. Thanks."

Reuters was caught red-handed doing it. This is important because Reuters and the AP set the standard for the rest...

Anonymous Anonymous October 20, 2016 8:44 AM  

the thing i've been thinking about lately, is what would a candidate that wasn't bought and beholden to TPTB look like? One that could get onto the national stage that is.He/she would need a fair degree of wealth, be famous/notorious, very self-confident, top notch at PR/getting attention.Who else besides Trump fits the bill? Rubio may have been more likely to beat Clinton than Trump, but that's because other than the 2nd amendment and an R beside his name, there isn't any real difference between them.

Blogger dh October 20, 2016 8:45 AM  

There has been no huge surge in new registrations for Republicans.

False


That article cites switches in registration, not "new" registrations. And it's from August. Those voters are already captured in the existing polling which shows Trump losing handily.

Blogger dh October 20, 2016 8:48 AM  

After all, the odds against Leave were 3/1 the day before. Trump currently stands at 4/1.

The polling was accurate on Brexit, the pundits were wrong on Brexit, which was one election. The current issue is 50 elections.

We will see. I agree there are really very few undecideds left, which is why I say the paint is already dry on this election.

Trump is such a bad candidate he has put reliably red states in play.

Blogger Whisker biscuit October 20, 2016 8:54 AM  

Are you a retard? A bad candidate doesn't set records in the primaries. He brought out more voters than anyone before, you prick.

Rigging polls, letting floods of illegals poll and vote and flat out fraud and cohersion by the media is something no one can overcome, let alone a candidate that has hardly any support from his rep party.

Anonymous reg October 20, 2016 8:55 AM  

As disagreeable as Trump is , what would a candidate that wasn't bought and controlled by TPTB look like? He would have to be wealthy, famous/notorious, very self-confident and top notch at PR/getting attention.There aren't many that fit the bill. Rubio may have been more likely to beat Clinton than Trump but that's because other than the 2nd amendment and the R after his name , there isn't any real difference between them.

Blogger VD October 20, 2016 9:08 AM  

The polling was accurate on Brexit, the pundits were wrong on Brexit, which was one election. The current issue is 50 elections.

Were they? From CNBC: Here's why the majority of Brexit polls were wrong. Leading up to voting day, the vast majority of polls predicted the remain side would prevail.

See also: 1992, 2015 Parliamentary elections.

Blogger OGRE October 20, 2016 9:09 AM  

Yeah the Pope's hands are tied on that one.

I think the biggest thing from the debate last night is Trump's straight up pro-life stance. Has any conservative in recent memory been that boldly pro-life? He should have a lock on any evangelical/christian/catholic vote now.

Blogger VD October 20, 2016 9:10 AM  

Are you a retard?

DH isn't a retard, he's a process guy. On the plus side, that tends to work well, right up until the process becomes obsolete and it doesn't. If the process holds, dh will be correct. If it does not, he won't.

I think - I think - this election is sufficiently abnormal that the process will not hold. But we'll see.

Anonymous That Would Be Telling October 20, 2016 9:12 AM  

@14 VD:

Why shouldn't it be closer to what the polls are actually showing?

Because that has never happened before sans a figure of national standing like Wallace or Perot.


@27 James Dixon makes a relevant point that according to Wikipedia Anderson got 6.6% of the vote. And one similarity was the denigrating attacks on Reagan by the establishment, although the GOPe wasn't as much against him.

However:

Also, because Johnson was the Libertarian candidate in the last election. I expect he'll actually do worse than he did in 2012, and will be below one percent. I've actually voted Libertarian several times in the past and I don't know a single libertarian who is enthusiastic about him.

Johnson is no Anderson, and it can't help him or the Libertarian brand that he picked a historically notorious and now on the campaign trail anti-gun VP. And the Green Party candidate all but endorsed Trump when she said she feared Hillary's warmonging, saying, for example, ""But on the issue of war and nuclear weapons, and the potential for nuclear war, it's actually Hillary's policies which are much scarier than Donald Trump, who does not want to go to war with Russia." Which I've found in debates to be a point no one can effectively reply to.

Anonymous WinstonWebb October 20, 2016 9:13 AM  

That article cites switches in registration, not "new" registrations.

Which bodes BETTER for Trump. Not only is he gaining a new voter, but Hillary is also losing one.

Anonymous Elipe October 20, 2016 9:14 AM  

Why would I tell (((polls))) I'm voting for Trump?

Seems like a pretty good way to lose your job. Why do that when I could have my cake and eat it too by surprising the pollsters on Election Day and getting to keep my job?

People are smart enough to realize that in the age of identity politics, you really shouldn't reveal your power level until you're casting your vote.

Anonymous Bobby Farr October 20, 2016 9:14 AM  

Isn't Trump still up or tied in all the Romney states plus Iowa and Ohio? I keep reviewing the polls but haven't seen the impending disaster that is the current media narrative. Even if he were to lose, he seems on track to do better than Romney or McCain despite worsening demographics.

Anonymous Sherrif T October 20, 2016 9:15 AM  

Talking about national polls at this point is dumb. It ignores the fact that the Electoral College determines the winner. National polls don't indicate anything with regard to who will win the elections.

Look at state polls.

Anonymous Sherrif T October 20, 2016 9:18 AM  

"Why would I tell (((polls))) I'm voting for Trump?

Seems like a pretty good way to lose your job. Why do that when I could have my cake and eat it too by surprising the pollsters on Election Day and getting to keep my job?"

More dumbness. People don't lose their job for talking to pollsters.

Anonymous Ominous Cowherd October 20, 2016 9:19 AM  

dh wrote:There has been no huge surge in new registrations for Republicans. There is no new silent majority for Trump.

Republicans aren't for Trump. If you are registering for the first time just so you can vote for Trump, why would you register as a Republican?

Blogger Johnny October 20, 2016 9:21 AM  

When they lie over and over, at some point it is best to simply assume you are being lied to because it is so likely. Don't worry it any more.

The polls tighten every election because they are cheaters. Tighter allows the pollsters to retain some reputation.

Even if there is a third party upsurge this year, I would imagine many will not bother to vote. None of the candidates are viable.

Trump will outdo his true numbers, whatever they are, owing to enthusiasm. I suspect a tight race or even a win in the popular vote, and I am not into it enough for any Electoral College prediction. No doubt if her margin is bigger there they will focus on it.

It seems no matter how cynical I am it is not cynical enough. What a bunch of dirt balls we have in Washington.

Blogger G-S. October 20, 2016 9:23 AM  

List of polls on Wikipedia: https://en.wikipedia.org/wiki/Nationwide_opinion_polling_for_the_United_States_presidential_election,_2016

I will agree that the 3rd and 4th party numbers are high this year. The media has attributed this to republicans going libertarian because of Trump and Democrats going Green because of Sanders. I'm doubtful that it will end up as low as 2% but can't imagine it goes any higher than 6% nationally. Since they are wasted votes the spread between Clinton and Trump is more important.

Blogger Alexander October 20, 2016 9:24 AM  

@51

My facebook is an excellent source of conservatives who want to be seen as nice non-racists and liberals who trend Bernie.

- Following Trump's nomination, lot of Republican signalling to Gary Johnson. None of these people have come put pro-Trump (and some of the women are happy to attach their 'like' to strong brave women who take stands against lewd comments, but none of them have come out for Hillary and none of them mention Gary Johnson any longer.

They either won't vote, or they'll go into the booth and throw one final private tantrum before voting for Trump.

On the left, all the women who were Berners are now going for Hillary, with various levels of enthusiasm but no real apparent angst about it. But Jill is a literal godsend - looks like the Bernie boyz can go to her and be shielded from claims of misogyny. Personally, my own efforts have been spent shilling for Jill w/o being anti-Trump and trying to shore up Berners from running back to Hillary.

Anecdotes aren't data, etc. etc. but if there's any surge among 3rd parties, it'll be entirely among the Greens. And I agree it won't break the 2% barrier.

Blogger dh October 20, 2016 9:24 AM  

Republicans aren't for Trump. If you are registering for the first time just so you can vote for Trump, why would you register as a Republican?

Gee, I guess you got me. All those new brown voters must be registering to vote for Trump!

Blogger Alexander October 20, 2016 9:26 AM  

@55

I'm sure Ken Bone thought that asking a bland question at a townhall debate wasn't going to get him in any trouble, either.

Blogger Daniel October 20, 2016 9:28 AM  

Scott Adams predicted a rise in polls to third parties. He says those are trump shy votes

Blogger Teri October 20, 2016 9:28 AM  

Of the media aligned against Trump, the NYT and Post are the worst. I trust nothing they put out.
You are also ignoring record small donor contributions.

Anonymous Ominous Cowherd October 20, 2016 9:28 AM  

Sherrif T wrote:More dumbness. People don't lose their job for talking to pollsters.

Do the people the pollsters are calling believe they might lose their jobs?

Blogger Michael Thompson October 20, 2016 9:29 AM  

"I will agree that the 3rd and 4th party numbers are high this year."

Not particularly. I mean, sure, they are closer than they've been in about 4 election cycles, but Perot eclipsed them easily in 1992 and probably even in 1996 and John Anderson polled as high as 26% in 1980, before crashing down to lower than 7% by election day. Anderson and Perot easily met the threshold to be included in the debates and Johnson didn't even come close.

Blogger dh October 20, 2016 9:29 AM  

Are you a retard? A bad candidate doesn't set records in the primaries. He brought out more voters than anyone before, you prick.

They are BAD RECORDS. It's not a good record it's like setting a new long-time record in the 100m dash. It's a fail.

From CNBC: Here's why the majority of Brexit polls were wrong. Leading up to voting day, the vast majority of polls predicted the remain side would prevail.

This is a pundit fail. The poll they talked about shows 48-45 with a 4.5% MoE at 95% CI. Meaning, too close to call. But they still called it for remain. And so they were wrong.

Blogger Johnny October 20, 2016 9:30 AM  

Republicans aren't for Trump. If you are registering

You speak for the party leadership, not for the rank and file. Plus Trump is running as a Republican. So of course, even if crossover votes are allowed, the Trump people will register as Republicans.

Anonymous That Would Be Telling October 20, 2016 9:31 AM  

@56 Sherrif T:

"Why would I tell (((polls))) I'm voting for Trump?

Seems like a pretty good way to lose your job. Why do that when I could have my cake and eat it too by surprising the pollsters on Election Day and getting to keep my job?"


More dumbness. People don't lose their job for talking to pollsters.


Yet.

That we know of.

And how do you know if the nice voice on the other end of the line is really working for an ostensibly honest pollster, vs. someone making lists for that sort of thing? Our cold civil war is heating up enough that you certainly can't rule it out, although if it's ever done, whoever does it will become very unpopular because it'll poison the well.

Anonymous Sherrif T October 20, 2016 9:33 AM  

"I'm sure Ken Bone thought that asking a bland question at a townhall debate wasn't going to get him in any trouble, either."

There is a difference between pushing a button on your phone and standing up in front of a national audience of millions. Right?

Blogger Cail Corishev October 20, 2016 9:36 AM  

Yes, the idea that third-party candidates most people have never seen are threatening to get Perot-style numbers seems like wishful thinking. Perot bought TV time to get his message out; Johnson/Stein aren't even in the game. Johnson's apparently an authoritarian "bake the cake" leftist running for the wrong party.

The same people who post "Why can't we get rid of these two and start over?" memes on social media are probably telling pollsters (when prompted?) they're going to vote third-party, but does that mean they will? There was plenty of "can't we do better than this?" angst in previous elections too, and nearly everyone still found a way to choose between Romney/McCain and Obama (or stayed home). The differences are much starker this time, so when that sense of duty that drives the people who vote every time kicks in, I think the same thing will happen.

Which way those votes go then, who knows. Considering we're talking about the most wishy-washy sector of the populace, it could depend on whatever narrative is loudest on November 7.

Anonymous Ominous Cowherd October 20, 2016 9:37 AM  

dh wrote:Gee, I guess you got me. All those new brown voters must be registering to vote for Trump!

My point is that it's unrealistic to expect most - let alone all - newly registered Trump voters to show up in the (R) column. Independent seems much more likely.

I could believe essentially all new Hillary voters are showing up as (D)s.

Blogger pyrrhus October 20, 2016 9:37 AM  

@69 Exactly. As the mobster in 'Casino' says, "why take a chance?" I stopped answering polls long ago when I realized there was no upside and significant downside to providing any personal information to anyone....

Anonymous That Would Be Telling October 20, 2016 9:39 AM  

@55 Sherrif T:

Talking about national polls at this point is dumb. It ignores the fact that the Electoral College determines the winner. National polls don't indicate anything with regard to who will win the elections.

Look at state polls.


Point, but correct me if I'm wrong about the public ones being both infrequent and typically only sampling 500 people instead of the normal national minimum of 1,000, significantly increasing their error margin even if done as honestly as possible.

Me, I'm wondering just how much polling is getting ever harder to do accurately as more and more people become unreachable by the old methods like POTS landlines. Perhaps one reason Scott Rasmussen sold out and departed from the firm he started?

Blogger Stilicho October 20, 2016 9:41 AM  

dh isn't retarded, not bright, but not retarded. What he is is a leftist despair troll. He's described himself as evil in the past and has since claimed to convert to Christianity, choosing the Roman Catholic church because he claims it allows him to vote for the abortion party because its social justice theology outweighs other concerns. YMMV.

Blogger Johnny October 20, 2016 9:43 AM  

This is a pundit fail. The poll they talked about shows 48-45 with a 4.5% MoE at 95% CI. Meaning, too close to call. But they still called it for remain. And so they were wrong.

Using your numbers there is only one chance in twenty that the predicted Brexit loss would be wrong, or 5%. Which is to say there is a ninety five percent chance (near certainty) that the polls were wrong or rigged. Probably a little of both.

This stuff happens over and over. Hey the stuff is rigged, plus not predictable in new situation.

It has largely been shown that people either lie to pollsters or are fuzzy headed enough to be unclear as to their own thinking. As for why one might lie, a person involved in any way with science as a source of employment would be (I think) very unwise to admit to not buying into the global warming hoax.

Blogger Whisker biscuit October 20, 2016 9:45 AM  

They're only bad because they challenge your narrative. That's not a defense as so much it is a reaction.

Anonymous Ominous Cowherd October 20, 2016 9:47 AM  

Johnny wrote:You speak for the party leadership, not for the rank and file. Plus Trump is running as a Republican. So of course, even if crossover votes are allowed, the Trump people will register as Republicans.

If you know a NeverTrumper, he's an (R), because if he were a (D), he'd be a Hillbot. If you don't know a NeverTrumper, you know that the GOPe is solidly NeverTrump. We keep hearing that Trump isn't really a conservative, and is a closet (D).

Separate ballots are only for the primaries, not the general election, so unless you registered back in the primary season, that's not a consideration.

If you were a discouraged voter who hated the GOPe because they are just like the Dems, or vice-versa, and you registered just to vote for Trump, why would you register as a (R)?

Anonymous Napoleon 12pdr October 20, 2016 9:49 AM  

What I'm seeing in all those polls is that Hillary isn't cracking 50%. She's hitting 47-48% and stalling. Which means Trump has a fighting chance.

Everything will depend on turnout. Personally, I think all the polls are junk at this point, simply because nobody knows how the turnout will look on Election Day. Neither candidate has much appeal, but Trump seems to have more hard-core support than Hillary Clinton.

Blogger Whisker biscuit October 20, 2016 9:50 AM  

Dh or "d*ck head", is the typical, obtuse lefty comment thread troll. He shifts the conversation to emotional hyperbole by saying "fail" in place of an explanation, and keeps needling at anyone who retorts him for the fun of derailing the conversation.

Anonymous Undocumented Civilizationalist October 20, 2016 9:50 AM  

So of course, even if crossover votes are allowed, the Trump people will register as Republicans.

We didn't. Registered for the first time in our current state. We both registered just to vote for Trump in the general, but we left party affiliation blank.

Just 2 voters, but...

Anonymous Sherrif T October 20, 2016 9:53 AM  

"
Point, but correct me if I'm wrong about the public ones being both infrequent and typically only sampling 500 people instead of the normal national minimum of 1,000, significantly increasing their error margin even if done as honestly as possible."

The number of people polled in order to make the poll reliable depends on the number of the whole. The whole is much smaller on the state levels. Also, there have been a number of polls from multiple polling firms in the battleground states.

"Me, I'm wondering just how much polling is getting ever harder to do accurately as more and more people become unreachable by the old methods like POTS landlines. Perhaps one reason Scott Rasmussen sold out and departed from the firm he started?"

It's an issue. It's important to reach cell phone users in surveys for better and more reliable results. However, doing so is much more expensive than hitting landlines. It's why polling results by better funded firms who make the effort to contact cell users tend to be better. Pew announced its surveys would now include 75% cell phones.

Blogger kennymac October 20, 2016 9:54 AM  

Here's an example of those state polls dh puts so much stock in.
https://theconservativetreehouse.com/2016/10/19/about-that-pro-clinton-arizona-polling-narrative/#more-123379

Blogger Whisker biscuit October 20, 2016 9:55 AM  

Robert Creamer: “It doesn’t matter what the legal and ethics people say, we have to win this motherfucker.”

Creamer: “…there are a couple of different organizations that’s their big trick [getting around voter ID laws to register illegal aliens].”

Gotta love the DNC Machine.

Anonymous Ominous Cowherd October 20, 2016 9:57 AM  

Whisker biscuit wrote:Dh or "d*ck head", is the typical, obtuse lefty comment thread troll.

VD wrote:DH isn't a retard, he's a process guy.

When the process is leftist trolling, being a process guy makes you a leftist troll.

Blogger Stilicho October 20, 2016 9:58 AM  

@pennywise: here's a start. You can Google individual polls.

http://www.americanthinker.com/blog/2016/06/no_the_other_polls_are_not_good.html

http://wwwmthegatewaypundit.com/2016/08/new-polls/

Anonymous Broken Arrow October 20, 2016 9:59 AM  

Rubio would be winning at this point, BUT it's because Hillary is sick and wikileaks. Nobody predicted the impact of these two things during the primaries and given the massive war chest of funding Clinton had created and the Clinton's political savvy she would have crushed Rubio, and made a laughing stock of Jeb.

Trump had a real chance until the release of the tapes and then the trend has been against him.

Rubio would have won the nomination in the first place if he hadn't broken his senate campaign pledge and joined the gang of eight. The GOP establishment has nobody to blame but themselves for the rise of Trump given their massive misunderstanding for their base, inability to communicate an effective narrative about their goals, and being pro-immigration.

If there's a Hiilaryslide then any pro-Trump people will be ruthlessly purged from the GOP, it will take a large step left and try to be a center-right, pro-immigrant, free trade party.

Blogger swiftfoxmark2 October 20, 2016 9:59 AM  

This just proves that the Libertarian and Green vote are primarily welfare recipients who are more concerned with gay wedding cakes than anything else.

Blogger Cail Corishev October 20, 2016 10:00 AM  

In other words, they will ask 45 democrats, 35 republicans, and 20 independents. That simply doesn't reflect the voter breakdown in America, nor does it reflect the dynamics of this"change" election.

What I keep wondering is why are they doing that. It's such an obvious way to skew the results; there are far more subtle ways to do it in the calculations that they don't release.

OpenID basementhomebrewer October 20, 2016 10:01 AM  

Whisker biscuit wrote:Robert Creamer: “It doesn’t matter what the legal and ethics people say, we have to win this motherfucker.”

Creamer: “…there are a couple of different organizations that’s their big trick [getting around voter ID laws to register illegal aliens].”

Gotta love the DNC Machine.



All the Faux outrage by the people who know it is rigged is getting rich. It's like telling a morbidly obese person who is eating a cake that they are a fat slob that should go on a diet.

They act with outrage because they know it's true.

Blogger Johnny October 20, 2016 10:01 AM  

If you were a discouraged voter who hated the GOPe because they are just like the Dems, or vice-versa, and you registered just to vote for Trump, why would you register as a (R)?

I suspect you are over thinking this stuff and that people would go Republican reflexively because Trump is running as a Republican.

GOPe is solidly NeverTrump

What is this GOPe you speak of?? Really, who are they? Do you think the ordinary (non leadership) self declaring Republican is more opposed to Trump than the population at large. Lacking some kind of proof this strikes me as very unlikely.

Blogger Natalie October 20, 2016 10:05 AM  

Given how hard they've doubled down I'm not even sure that polls not tightening would mean anything. They have staked their entire world on Hilary. They're toast if Trump wins anyway, so they don't have much incentive to do anything other than what they're doing IMO

Blogger kaflick October 20, 2016 10:06 AM  

One thing that has me a little more optimistic here in Toledo is that there are several homes with signs for local and state democrat candidates and trump signs. For a heavy union town like this that is unprecedented.

Blogger Johnny October 20, 2016 10:09 AM  

Natalie wrote:Given how hard they've doubled down I'm not even sure that polls not tightening would mean anything. They have staked their entire world on Hilary. They're toast if Trump wins anyway, so they don't have much incentive to do anything other than what they're doing IMO

The usual explanation for the polls tightening is that it is the polsters protecting their personal reputation. It allows them to appear to be less wrong. I think they will still tighten, but you could be right about this. More intense than usual. Clearly they see Trump as having a good shot at it, if not better.

Anonymous Sam the Man October 20, 2016 10:11 AM  

Dh,

You are very sure of yourself and the Hillary win and I can understand that by a straight up look at the polls. However there is a point to what Vox said, that prediction algorithms only work when they are in the piecewise continuous realm. During periods of disturbance such algebraic prediction modes do not work, as the polynomial equations no longer reflect the function they are modeling. Examples in US history of such elections are 1860, 1932 and 1980. The last is within living history and the polling got that very wrong.

In the case of 1932 and 1948 there were issues with the polls, because the method used to collect the data missed populations. Now it seems today that a vast number of folks with land lines never actually answer them unless they know the person calling, and a similar population never actually respond to polls of any nature. I have seen no articles that indicate they know how to factor this in. I certainly do not know.

To that you have to add the systemic push polls where they assume 40 to 46 % democrat turn out, when the 2012 turn out is 35%. It is hard to find any way in which this over polling of the left is based on reality rather than mendacity. All the evidence seems to show that the Democratic turn out will be less, specifically among the white blue collar population(based on cross over registrations and small local polls)

The evidence for this I see in the state of PA, where I live and work is: 1) the polling organizations are under polling blue color white votes. The local county articles show Trump doing very well in any white blue collar population, yet the polls of Dems seem to show Hillary has more support than Obama did back in 2008. Not realistic.
2) cross over registrations, The Dems lost 100,000 registrations since 2012, the republicans have gained 132,000. A shift of almost 250,000 in republican favor, hardly the swamping by minorities you spoke about. Of course PA, OH, MI all represent rust belt where Trump my well be a disruption candidate (i.e. the old assumptions do not apply)

3) One way to get the polling answer you want is to poll by zip codes you know will produce the vote you want, which is the only way I see correlation of the percentages they say Hillary has here in PA, they are polling regions they all ready know favor her by a large percentage. If that were the case you would expect a widely divergent number of college educated folks...and sure enough that shows up with the college grads polled being around 50 % of the polled, instead of the 25% that represents the actual electorate.

Not saying trump is winning, just suggesting that this year indicates some really unprecedented push polling and the possibility that the old models are really off. Said another way, a lot of mathematical folks tend to confuse the polls for reality, when in fact all such effort only an attempt to model reality. If certain assumptions are wrong, then they are useless.

Again not making a prediction, but I could see this being very much like Brexit, or 1948.

Blogger Johnny October 20, 2016 10:12 AM  

>>All the Faux outrage by the people who know it is rigged is getting rich.

Ain't it something! What a bunch.

Anonymous Sensei October 20, 2016 10:13 AM  

@14

Also, because Johnson was the Libertarian candidate in the last election. I expect he'll actually do worse than he did in 2012, and will be below one percent. I've actually voted Libertarian several times in the past and I don't know a single libertarian who is enthusiastic about him. -VD

It's not libertarians driving that vote, it's evangelicals who won't vote for Trump because he says unseemly things and they are trying to reconcile conservative theology with the progressive worldview they've swallowed more or less whole without noticing.
My FB feed is chocked full of people claiming that voting the lesser of two evils is impermissible, and therefore our only hope to vote our conscience and keep both people from being elected is an alliance between Johnson and someone named Evan McMullin who is apparently popular in Utah.
No that is not feasible nor does it make a shred of logical sense, either pragmatically or morally, but any naive faith I once possessed in the ability of average Christians to make morally coherent tough decisions has been shattered by this election. Give them an out, even an absurd out that only aids a declared enemy of the Church, and they'll bolt in that direction, and manage to elect Hillary while believing all the time they've behaved in the only morally defensible manner possible.
There is definitely a sizable silent majority, however, that wouldn't even say on a poll that they'll vote Trump, but will. Whether that's enough to overcome what will surely be the most comprehensive attempt at vote fraud in national history remains to be seen. (How they are going to blame the Russians when all the dead votes are for Hillary I don't know, but they'll find a way)

Blogger Cail Corishev October 20, 2016 10:13 AM  

With all of the "rigged" talk coming from DJT, I wonder if anyone in his camp thought about amassing an army of watchers in these regions or organizing something to counter.

I saw a call from him on social media for Republican volunteers to do just that sort of thing.

Anonymous That Would Be Telling October 20, 2016 10:20 AM  

@95 Sam the Man:

All the evidence seems to show that the [Pennsylvania] Democratic turn out will be less, specifically among the white blue collar population(based on cross over registrations and small local polls)

There's also the identity issue, blacks aren't going to vote for Hillary like they did for Obama, both in turnout and ballots cast, and of course she's a woman, although many have made convincing cases that's not going to work hardly so well for her.

Blogger Student in Blue October 20, 2016 10:22 AM  

It's not libertarians driving that vote, it's evangelicals who won't vote for Trump because he says unseemly things and they are trying to reconcile conservative theology with the progressive worldview they've swallowed more or less whole without noticing.

With how Trump affirmed Pro-Life and attacked Hillary for her abortion stances in this last debate, there might not even be that many evangelicals protest-voting Johnson.

I'd say VD's guess is still spot-on regarding Johnson.

Blogger CarpeOro October 20, 2016 10:23 AM  

@91
YMMV. I registered independent so I could vote Trump. Trump support is not reflexively Republican support - the GOPe have done an excellent job of separating themselves from him, and I think a greater number than normal show it. The GOPe hated Reagan also, but didn't separate themselves so publicly. People registered R to vote for him, but the scenario isn't playing out that way this time.

Blogger lowell houser October 20, 2016 10:31 AM  

IDB/TIPP was the most accurate poll in 2012 of the ones listed. We're not looking at a Trumpslide. We're looking a another Brexit where the professional pundit class sits there stupefied at the end of the night wondering what the hell just happened.

Blogger Whisker biscuit October 20, 2016 10:32 AM  

Evangelicals set back and let abortion clinics sell baby parts; men "marry" men; and grown men tuck their ween between their legs and use the potty alongside little girls.

They're one to object to a little vulgar boy talk.

Anonymous Amir Larijani October 20, 2016 10:32 AM  

I find it hilarious that the MSM has their collective panties in a twist over Trump calling Hitlery a "nasty woman".

Allah damn...a Republican finally had the balls to tell the truth about her.

Anonymous Gertie Gasche October 20, 2016 10:34 AM  

There are 5 voting age adults in our family. Husband and I have not voted since 2000. Adult daughter was Republican but switched to no affiliation. Last time she voted was McStain. Son and his wife have never registered to vote; are in their 20's. We are all now registered to vote, with no party affiliation, and plan on filling in the bubble for Trump. There's an awful lot of Trump signs here in Oregon in and near Eugene. Thousands.

Blogger James Dixon October 20, 2016 10:35 AM  

> Polls always tighten the closer to an election.

And yet in this election year we're supposed to believe that Hillary is pulling away. Or at least that's what the MSM is reporting the polls as saying.

> The poll they talked about shows 48-45 with a 4.5% MoE at 95% CI.

And most of the current polls are also with in the margin of error.

> There is a difference between pushing a button on your phone and standing up in front of a national audience of millions. Right?

Not to the democrats. Do you really trust that polling data to remain private?

Blogger Cail Corishev October 20, 2016 10:35 AM  

I'll be interested in the Catholic response. Not only do we have a clearer contrast than ever between the two sides on abortion, but combine that with the leaks showing the Dems playing liberal Catholics for chumps. Liberal bishops are irked, partly from the embarrassment of being shown to be the fat girl at a pig party, but also because it shows what a sham their "seamless garment" arguments were all this time.

You wouldn't think that at this stage of the game, it would come as a surprise to anyone that Hillary Clinton is full-bore pro-abortion, yet I'm seeing just that in social media. And just this morning I talked to a traditional Catholic woman whose #1 issue is abortion, and she didn't know where any of the candidates stood on it. But she knew both candidates were horrible (though she couldn't say why exactly) and the whole thing is depressing, right in line with the Narrative.

Anonymous 11B October 20, 2016 10:35 AM  

Trump made a definitive statement last night that the female accusers were lying and were probably put up to it by the DNC machine.

Let's see if we hear about these women again. You know if there was any legitimacy to their claims, the MSM would be hanging Trump with it today. But if these gals disappear down the memory hole, then it will confirm Trump was right and it was all a big lie.

Blogger Stilicho October 20, 2016 10:36 AM  

Gallup's historical tracking of voter party affiliation shows the following:
11/08 28r 37i 33d
11/10 26r 41i 31d
11/12 30r 33i 35d
11/14 28r 41i 28d
09/16 27r 40i 32d (most recent)

The dem advantage swings between 0 and 5 percent. The voting patterns of independents should be predictive. Although turnout is a huge issue this year with Hillary receiving 1.5 million fewer primary votes this year than she did when she lost the primary in 08.

Blogger Natalie October 20, 2016 10:37 AM  

After being told I MUST vote for Romney or McCain I have no sympathy for the moral posturing of the neverTrump evangelicals (ahem - Matt Walsh).

McCain is a notorious philander who ditched his first wife after she stuck with him through Vietnam, and he is on record telling rape jokes.

Ryan is a open borders fool.

Romney is a lib-lite creepy cult member.

Egghead is a virtue signalling CIA numpty who panders to skittles while prosing about social justice and racism.

Any neverTrumper who voted for or approves of any of the above is an absolute hypocrite who deserves scorn and ostracism. Look, I'm a nice SAHM, and I want to start troll accounts on twitter just to spam people like Walsh with gay pepe porn. He is doing his damnedest to sell out my family's future because his poor ittle feels are hurt because apparently Trump didn't consult the secret kings' feelings when running his campaign.

And I can't even work this off at the range because no more indoor ranges for me until I pop out this kid....

Blogger VFM #7634 October 20, 2016 10:39 AM  

This morning...

LATimes/USC Dornsife: Trump +1
Rasmussen: Trump +3

Florida Vote-by-Mail results so far:
Republican: 369,528
Democrat: 356,661
Other: 23,421
Independent: 136,863

Blogger Cail Corishev October 20, 2016 10:41 AM  

I find it hilarious that the MSM has their collective panties in a twist over Trump calling Hitlery a "nasty woman".

As soon as I heard that, I knew that's what they'd fixate on today. They literally have nothing else. There's no issue they can hammer on that helps her, because unlike her husband, she doesn't even make a pretense of being a moderate. Abortion, guns, war, taxes, immigration -- she's on the extreme globalist left on all of them, and it's too late for her to pretend otherwise (and as a SJW, she's probably incapable of it anyway).

So all they have is character, and the hope that they can depress Trump's vote by yelling about stuff like this all day, allowing their 47% or whatever of government dependents to carry the day.

Anonymous JI October 20, 2016 10:42 AM  

I wonder how the pollsters are handling non-response bias. Don't the pollsters call landlines, or go door-to-door? If so, I'm thinking of the bias caused by the fact that hard-working people are not at home to answer the phone or the door. The people who are at home are often those on public assistance who would vote Democrat because that party has a history of paying better. Or they get housewives, women, and we know how women tend to vote. I just wonder if they have good, reliable methods for dealing with this source of bias.

Blogger Johnny October 20, 2016 10:42 AM  


@100

I am not against religion so I don't get into it, but where I live a lot of the the "intensely religious" are bunch of brain turned off knuckleheads. Often but not necessarily older. What makes me a little crazy is that Trump's perceived flaws in this area have little to do with governance. Despite being seriously full of himself he is willing to delegate, making vanity a minor issue in terms of outcomes, and a lot of his sexuality is simply his willingness to act out more than most men do. A lot of it can be attributed to a higher level of opportunity. You know, like the Hollywood folk, they sleep around quite a bit because they can.

Blogger Cail Corishev October 20, 2016 10:49 AM  

I just wonder if they have good, reliable methods for dealing with this source of bias.

They have very good methods of filtering out all sorts of bias -- as long as things go pretty much as they expect, following the patterns of previous elections and polls. If things don't follow the status quo and there are voting block shifts, that makes it a lot harder, and their attempts to correct for bias could even multiply the error.

Anonymous Casey October 20, 2016 10:52 AM  

https://www.youtube.com/watch?v=eOkAyUmyQko

Anonymous The OASF October 20, 2016 10:55 AM  

@92

"They have staked their entire world on Hilary. They're toast if Trump wins anyway, so they don't have much incentive to do anything other than what they're doing IMO"

It's a little more of a far fetched conspiracy than usual, but they may being pulling the "Boogeyman Rusky" routine so that Obama has a pseudo-plausible rationale for overturning or suspending the election outcome if Trump (probably) wins. Read the security agency dossiers that Killary referred to last night regarding the rigged-Rusky elections. It's badly penned Hollywood scriptwriter stuff - no wonder Trump was so bold in telling her and her agencies that they have no clue if and who, may be rigging what!

Anyway, the legal authority is now in place for POTUS to assume emperor status. And Obama has expressed enthusiasm for installing himself as a USA Muslim dictator. So the "blame it on the Rusky" campaign along with an outright war deceleration on Russia immediately following the election cannot be dismissed out of hand.

Blogger SemiSpook37 October 20, 2016 10:57 AM  

Natalie wrote:After being told I MUST vote for Romney or McCain I have no sympathy for the moral posturing of the neverTrump evangelicals (ahem - Matt Walsh).

Except that Walsh is a cucked Catholic. A garden variety Maryland CINO, if you will.

I know hardcore Traditionalists that have been on the Trump Train since Day One, and they've NEVER faltered at any point of this process. They've only gotten stronger in their support and winning others to his side.

Blogger Student in Blue October 20, 2016 11:00 AM  

@VFM #7634
LATimes/USC Dornsife: Trump +1
Rasmussen: Trump +3


To be fair, I believe those are still pre-debate sampling. Polls that reflect the debate should come out early next week I believe.

Please correct me if I'm wrong on that though, I'm only something like 60% sure of that.

Anonymous Will October 20, 2016 11:00 AM  

@21 and @83 both posted the link but didn't explain it. The point was the latest Arizona Republic poll for the state of AZ showing HRC "up 5 pts". But in order to get her +5 they used a D +34 sample, i.e. Dems were 58% of the poll, R's were 24% of the poll, and I's were 19%.

Tell me how that reflects any sort of reality.

Blogger Student in Blue October 20, 2016 11:02 AM  

And Obama has expressed enthusiasm for installing himself as a USA Muslim dictator.

At this point, he just wants to quietly play golf without being bothered.

Cankles on the other hand, there's a person who would love to have as much power as long as possible.

Anonymous Sherrif T October 20, 2016 11:05 AM  

"Not to the democrats. Do you really trust that polling data to remain private?"

It hasn't been released ever before. Releasing that information would be a guarantee of the polling organization going out of business. Yoiur claim is absurd.

Blogger Stilicho October 20, 2016 11:05 AM  

@will see my post at #111 for gallup's history of voter affiliation.

Blogger James Dixon October 20, 2016 11:06 AM  

> Anyway, the legal authority is now in place for POTUS to assume emperor status.

The justification is in place, not the legal authority. Not that that has ever stopped Obama.

Blogger Natalie October 20, 2016 11:07 AM  

Semispook - point taken. He's just shared and emulated by a lot of my actually evangelical FB circles, so somehow I guess I missed the Catholic part.

Blogger dh October 20, 2016 11:08 AM  

And most of the current polls are also with in the margin of error.

The national polls. Which is why you can't use them at this point. National polls have small sample sizes from a huge population, and require lots of statistical games to get them to the 95% CI.

State-based polls, which have similar samples but much smaller populations, have much lower MoE, which is why you should use them by now, not national polls.

For example, North Carolina was sampled by SurveryUSA, 700 voters in a smaller state. They had a MoE of 3.9%.
http://www.surveyusa.com/client/PollReport.aspx?g=ce17282d-d070-493a-8bcd-ee02ca9016e6

Qunnipac surveyed Pennsylvania recently, showing Sec. Clinton up 6%, with a 3.2% MoE.
https://www.qu.edu/images/polling/ps/ps10172016_Sub62fgv.pdf

Arizona Republic polled Arizona, found Clinton up +5%, with a 4% MoE.
http://www.realclearpolitics.com/epolls/2016/president/az/arizona_trump_vs_clinton_vs_johnson_vs_stein-6087.html

Larger relative sample sizes, better MoE/CI or rates of sampling error. It matters at this point. This is literally the reason that modern pollsters are using the aggregated state models to predict winners - many statistically superior surveys, average together, reduce outliers, produce more stable results, and are more predictive.

Anonymous That Would Be Telling October 20, 2016 11:11 AM  

@122 Sherrif T:

"Not to the democrats. Do you really trust that polling data to remain private?"

It hasn't been released ever before. Releasing that information would be a guarantee of the polling organization going out of business. Yoiur claim is absurd.


He didn't say "release", he said "remain private." In a election season where the InfoSec practices of one party have been shown to be abysmal. Are you so very sure all these polling companies have their raw data sufficiently locked down?

And that doesn't address the issue of false polls, which we know exist for other reasons, especially push polling, which Gore started in Florida many hours before the polls closed, so he could then try his recount gambit.

Blogger James Dixon October 20, 2016 11:11 AM  

> It hasn't been released ever before.

And how often has a CEO been fired for contributing to a political campaign before? How often have people who deliberately mishandle classified documents been allowed to walk away without any action taken?

> Releasing that information would be a guarantee of the polling organization going out of business.

Who said it would be "released". Their computers could be hacked, or it could be mysteriously leaked by "persons unknown". You seem to think the left cares. They don't. They would cheerfully put a bullet in the head of every Trump voter if they could get away with it.

> Yoiur claim is absurd.

Your naivety is absurd.

Anonymous A Deplorable Paradigm Is More Than Twenty Cents October 20, 2016 11:12 AM  

nd just this morning I talked to a traditional Catholic woman whose #1 issue is abortion, and she didn't know where any of the candidates stood on it

More like "she didn't want to admit". The DNCe and NARAL have been besties for 25 years, it's not a secret. Plenty of ethnic Catholics just look at something else when that part of their party platform shows up.

Or she was a "Christmas & Easter" Catholic. Lot of those.

Blogger SemiSpook37 October 20, 2016 11:14 AM  

Natalie wrote:Semispook - point taken. He's just shared and emulated by a lot of my actually evangelical FB circles, so somehow I guess I missed the Catholic part.

No worries. I used to read him, and then the whole primary election happened, and he just wouldn't get off of his high horse.

Blogger dh October 20, 2016 11:18 AM  

@21 and @83 both posted the link but didn't explain it. The point was the latest Arizona Republic poll for the state of AZ showing HRC "up 5 pts". But in order to get her +5 they used a D +34 sample, i.e. Dems were 58% of the poll, R's were 24% of the poll, and I's were 19%.

This is basic mathematics. You correct for sample misdistribution.

Here is what they wrote about it in the actual poll report:

The sampling frame used only registered voters. The weighting procedure involved a poststratification technique that adjusted the sampling weights of the individuals who answered
the questions analyzed to match the proportions in the registered voter database of county
(determined from sampling frame), party (determined by sampling frame), age (determined by
a question at the end of the survey), and gender (determined by interviewer observation at the
beginning of the interview). Before using the weights, the total counts of voters were adjusted
to compensate for the number of likely voters (based on the survey data proportions) and
survey non-response.


Anonymous Sensei October 20, 2016 11:21 AM  

@100
With how Trump affirmed Pro-Life and attacked Hillary for her abortion stances in this last debate, there might not even be that many evangelicals protest-voting Johnson.

The immediate reaction I got upon pointing that out was "but Trump is just saying that, he wasn't pro-life before."

However, I agree that it's going to bring back some evangelicals. There had been little to nothing on the abortion issue thus far, which of course has been a big deciding issue for conservatives in the past, but now as you say Trump came out so strongly pro-life that the cognitive dissonance for a lot of Christians is going to flip flop. The "Hillary: kill babies up until birth but ban guns to protect toddlers" memes are already popping up on FB.

Blogger Frank Brady October 20, 2016 11:21 AM  

What a shock! The polls sponsored by the corrupt establishment media uniformly report that the corrupt establishment candidate and criminal Hillary Clinton won last night's third and final debate. Those polls NOT sponsored by the corrupt establishment media show that Trump won, overwhelmingly so in most cases. By the way, we should probably cut CNN a bit of slack since they apparently included a lot of ORCS in their sample of voters.

Blogger kennymac October 20, 2016 11:36 AM  

Bullshit dh. No way a poll that samples Dems at almost 60 PERCENT of the total is anywhere near accurate.

Anonymous That Would Be Telling October 20, 2016 11:37 AM  

@132 Sensei:

but ban guns to protect toddlers" memes are already popping up on FB.

And how she specifically pushed that made another opening for the pro-gun crowd, for the details in the D.C. ban included those who had registration grandfathering were required to keep their guns disassembled, and if your home was invaded, unloaded until the criminal entered the room you were in. Given how the country's gun culture is strongly shifting from "1.0" hunting and target shooting to "2.0" self-defense, it's a big opening, although of course not as big as calling for an Australian style actual confiscation.

Blogger SouthRon October 20, 2016 11:42 AM  

Here in North Carolina I've noticed a change over the past week.

All of a sudden there are Trump signs popping up all over the place. Before, most people here have considered it unsafe to post a Trump sign in your yard or put a sticker on your car. Sheesh, I've even seen trucks with big Gonzalez lettering on the back window starting to sport Trump stickers.

I think the reason for this visual surge is twofold.

Last week Franklin Graham had a rally with 14,000+ people on the capitol steps. Now most places Franklin Graham may not carry any weight, but the Graham's are North Carolinians, and we are talking about Evangelicals and Churchians, so here it means something.

There he basically gave evangelicals very solid reasons for voting in this election, and voting for Trump. 4-5 potential Supreme Court appointments. We live in an evil and fallen world. So, it has always been an issue of voting for the lesser of 2 evils.

This resonated. I popped into a little country church this weekend and this was the message given to the life-long locals. Many seemed to breathe a sigh of relief that they could vote for Trump without violating their Christian conscience or endorsing his behavior.

The other thing is what happened after Hurricane Matthew. While it didn't make landfall here, it made an over-sized mess of things due to how the storms tore things up inland.

What we saw afterward was Obama showed up, hardly mentioned the storm or the devastation instead taking it as an opportunity to campaign for Hillary. The obvious lack of care and compassion coupled with raw opportunism spoke volumes.

At the same time, the Trump women including Lara Trump, who is from NC, suspended their campaigning for a couple days and pitched in to help in Fayetteville, Lumberton and the surrounding areas. They showed up with a semi full of food and supplies "daddy" had paid for and helped distribute it and clean up a bit. When the press wanted to talk about her father-in-law she answered briefly and then said, excuse me, that's not what we came here for, let us get back to helping those affected by the storm.

That contrast in behavior between the two sides couldn't have been more obvious. I have heard more than one previously undecided state that that alone made up their mind to vote for Trump, even one or two that have voted D their entire life.

Blogger dh October 20, 2016 11:47 AM  

Bullshit dh. No way a poll that samples Dems at almost 60 PERCENT of the total is anywhere near accurate.

It's like you are not paying attention. The sampling rate is irrelevant; they discount the samples that are over sampled on a county by county basis.

So if the Registered Voter break down is 33R, 33D, and 33I, but the response rate is 33R, 66D, 18I they will count ever Republican response 1 time, every Democratic response 1/2 times, and every Independent response 2 times.

This is basic statistics. Literally, entry level probability and statistics.

Last cycle, when all the morons decided to unskew polls, they adjusted the factor of R voters to compensate for what they believed was under-counting of the Republican vote and over predicting of the brown vote.

It was so bad that Mitt Romney was the last person in America to know that Obama was kicking his ass all over the country.

http://www.theatlantic.com/politics/archive/2012/11/whole-romney-ticket-believed-unskewed-polls/321618/

Blogger dh October 20, 2016 11:48 AM  

I love this thread. Bookmarked for Nov 9th.

Blogger kennymac October 20, 2016 11:54 AM  

dh, thanks for the explanation. so what you're saying is that I was basically right, a skewed sample rate of D+34 would be junk, so that's why they weight it as you said.

Blogger Johnny October 20, 2016 11:57 AM  

A near universal in science and the science of polling is to overestimate the merits of this, shall we say, math stuff. I suspect part of the reason is that it is their product. What they have for sale is their outcomes and naturally they don't want to get into the flaws involved.

Typically in these error terms is sample size, and because in almost everything there is other stuff going on, the utility of the numbers is much less than is supposed.

Maybe the easiest example of this is in nutrition and food. Every couple of years what we are supposed to eat changes. And I bet ya with near certainty the new proof is a statistical analysis of eating patterns.

Anonymous Deplorable S E Delenda October 20, 2016 11:59 AM  

@137 @138

"It was so bad that Mitt Romney was the last person in America to know that Obama was kicking his ass all over the country."

You are amusing SJW.

http://www-personal.umich.edu/~mejn/election/2012/countymaprb1024.png

Anonymous Sam the Man October 20, 2016 12:04 PM  

Dh,

I know you put a lot of credit on state polls, but lets look at the one you just posted:

https://www.qu.edu/images/polling/ps/ps10172016_Sub62fgv.pdf

Now this is straight from that poll:

Pennsylvania independent likely voters back Clinton 43 – 36 percent, compared to a too-close-to-call 38 – 35 percent Clinton lead among this group October 3. Republicans go to Trump 85 – 7 percent, as Democrats back Clinton 83 – 10 percent.
She leads 52 – 37 percent among women likely voters and 72 – 17 percent among non-white voters. Trump leads 47 – 40 percent among men and 47 – 41 percent among white voters.

If in fact the above is true, and whites make up 81.5 % of the electorate, that the 47/41 ratio shows Trump has an absolute lower end, assuming he wins not one Hispanic/black voter of 38 % Hillary gets 33% of the vote from whites. That means of the remaining 18.5 % of the electorate, She is getting 14% and trump is getting 3 %, or only 16 % of the non white electorate. Very possible but kid of extreme if the ethnic polls that show Trump with 11% of blacks, 35% of Hispanic and 48 % of Asians hold any validity. But there is one more rub: Will blacks you make up around 11 percent of the population show up in 66% ration like the did in 2012? If there participation drops to something close to traditional norms Hillary is in trouble. They do not post the expected turn out, but most of the models I have seen assume a 2012 like turn out. What the poll does not seem to gauge is the enthusiasm gap, the folks for Hillary are not that wild about her compared to 2012 or 2008.

Second the percentage of democrats supporting Hillary: only a 10 % drop off voting for Trump. That does not match what I see in the working class of eastern PA, and it most certainly does not match what folks tell me in the central/western part of the state. What gives? I can only conclude the poll is missing something.

The number of folks who were polled in PA was 660. There are 67 counties in PA so it was around 10 per county. There are 2,176 zip codes in PA so it was around 1 person polled per 3.3 zip codes. If the polling was not distributed evenly across the state it is very possible to miss a movement in one part of the state away from predicted norms. I am not saying they did that, just telling you that the % of Dems who are not supporting Hillary does not reflect what I am seeing on the ground here in PA or hearing in discussions with folks from other parts of the state.

The issue is this: in any kind of a phase change elections the polls do not reflect the underlying reality they are trying to measure, because basic assumptions about group responses that have been valid, no longer are.

Come November 9th I will not be surprised if Hillary wins. Not saying she will not. But based on what I can see, I would not be surprised if the entire polling this election has made some great big boners, simply because the base assumptions underlying the polling data does not reflect reality,

Anonymous #8601 October 20, 2016 12:05 PM  

The Democratic turnout is gonna be very, very low.

The Republican turnout will be Yuge.

Blogger BassmanCO October 20, 2016 12:17 PM  

Here in Colorado, I am a registered (I), voting Trump. I know of one registered (D) who always tells pollsters he is voting for Hillary when he has been on the Trump train from day one. Another LIBERAL Democrat who confided she will be voting for Trump.

Purely anecdotal, but combine enough other such stories and a pattern emerges.

Anonymous FP October 20, 2016 12:21 PM  

Gertie Gasche wrote:There are 5 voting age adults in our family. Husband and I have not voted since 2000. Adult daughter was Republican but switched to no affiliation. Last time she voted was McStain. Son and his wife have never registered to vote; are in their 20's. We are all now registered to vote, with no party affiliation, and plan on filling in the bubble for Trump. There's an awful lot of Trump signs here in Oregon in and near Eugene. Thousands.

Good to hear Trump has visible support down in Eugene. It has been mostly quiet in Marion county, very little signs of any sort in repub or the dem parts I've seen. I figure Trump will win by at least 3% here though as that what Romney got.

As for third parties, I don't have an inkling that they will hurt Trump here in Oregon but Clinton. Lots of Bernie Bros may break for Stein or Johnson. Both of them were on the ballot last time and didn't break more than 2-4% in any county then. Comparing either of them to Perot is nonsense even with Trump and Clinton having such high "dislikes".

As for polling. I stopped answering most calls for a couple of reasons. One, with the advent of VOIP calling, most of those polls come across as just random numbers (usually cell phone based area codes) and I will just let that go to voice mail as it could just as much be a telemarketer. Second, most of the polls I participated in as a non-affiliated voter from 2008-2012 or so were so obviously biased in their questions it just made it seem pointless.

Anonymous Yay Trump October 20, 2016 12:23 PM  

What early voting and absentee data we have doesn't show a rising brown tide drowning out white votes. And there was a very recent article that noted there didn't appear to be any Hispanic turnout surge, but I can't recall the link offhand.

When we have a week of in-person early voting (so, sometime next week), we can make more solid guesses about how screwed Trump or Hillary might be.

Blogger dh October 20, 2016 12:23 PM  

Sam--

Very possible but kid of extreme if the ethnic polls that show Trump with 11% of blacks, 35% of Hispanic and 48 % of Asians hold any validity.

Right, so this poll is evidence that these other polls, which I have never seen, are probably wrong. 11% of blacks would be what Pres. GW Bush pulled in from blacks on his second term, when he beat Sen. Kerry handily. Does it seem like Mr. Trump will do better or worse than blacks with Pres Bush?

But there is one more rub: Will blacks you make up around 11 percent of the population show up in 66% ration like the did in 2012? If there participation drops to something close to traditional norms Hillary is in trouble. They do not post the expected turn out, but most of the models I have seen assume a 2012 like turn out. What the poll does not seem to gauge is the enthusiasm gap, the folks for Hillary are not that wild about her compared to 2012 or 2008.

In absolute numbers the vote total will be similar to 2012, but not as a percentage. All the models I have seen expect lower than 2008 turnout percentage wise, and equal or lower than 2012. Pres. Obama was a special case for black voters. What you have forgotten is that since 2012, all states including PA are browner and younger.

Come November 9th I will not be surprised if Hillary wins. Not saying she will not. But based on what I can see, I would not be surprised if the entire polling this election has made some great big boners, simply because the base assumptions underlying the polling data does not reflect reality,

There could always be some unseen reason why none of the polls capture this new reality on the ground. I agree, this is a possible outcome. It is made ever so unlikely because of the multiple polls.

What you are seeing is anecdotes, which could be a harbinger for something totally new. But it's probably not.

Blogger dh October 20, 2016 12:24 PM  

You are amusing SJW.

Tell it to Pres. Romney.

Blogger Richard Stone October 20, 2016 12:27 PM  

"I'll be interested in the Catholic response. Not only do we have a clearer contrast than ever between the two sides on abortion, but combine that with the leaks showing the Dems playing liberal Catholics for chumps. Liberal bishops are irked, partly from the embarrassment of being shown to be the fat girl at a pig party, but also because it shows what a sham their "seamless garment" arguments were all this time."

The left Catholics always had contempt for their Bishops. Very few traditional Catholics defend them now. Even the most ardent "we must follow the example of our Bishops" people are getting quiet. With the Pope's new appointments, expect the contempt to grow. It's a moot point anyways, as the Church lost all of it's political clout when it destroyed Catholic identity in VII.

Blogger Basil Makedon October 20, 2016 12:27 PM  

@dh

That is interesting analysis of the Arizona Republic poll. I understand what you are saying about how they weight to correct for their oversampling and agree with you to a certain degree. [Please note that I am not a "skewed" poll guy, been burned by that before. That being said, I'm not going to let myself be gaslighted either].

The only thing that I would say in response to this is the following: If one's sampling methods result in such oversampling -- and they admit to this -- then it seems to me that one should mistrust the entirety of their sample, not just attempt to correct it, regardless of whether it is a reasonable method of correction or not.

In other words, if they called people and got 60 D to 20R or whatever, they were doing something so fundamentally wrong that they should throw that data out rather than try to work with it.

Blogger Whisker biscuit October 20, 2016 12:29 PM  

dh loves him some Hilliary, but has never said why.

Anonymous Yay Trump October 20, 2016 12:31 PM  

Many state polls are using higher than 2008 or 2012 female and minority turnout, even where registration data doesn't reflect this for the state or locality. Many are also assuming 65% R for Trump as a ceiling, not a floor. Again, these are things the current (admittedly small) early voting, absentee and mail data isn't really showing.

Anonymous a_peraspera October 20, 2016 12:32 PM  

"Rubio would be winning at this point..."

No. If both candidates were for immigration amnesty, then voters would go back to the "not a dime's worth of difference between the two parties" mindset, and stay home like they did for McCain/Romney.

Blogger Whisker biscuit October 20, 2016 12:32 PM  

But it gives dh the results he wanted. It's narrative, not truth.

He's fun to smack around, but any "Catholic" that supports a candidate that's in favor of sucking a kid's brain out and hocking the spare parts for profit is not one to be honest. Only self-evidently.

Blogger Cail Corishev October 20, 2016 12:33 PM  

The immediate reaction I got upon pointing that out was "but Trump is just saying that, he wasn't pro-life before."

If Hillary even made a pretense at moderation like her husband did, it would be easier for them to hide behind that. But even if Trump were the crypto-liberal of NeverTrump fantasies, he would still be far better than Hillary on life, because he brings with him Mike Pence and a bunch of other pro-lifers, a list of good potential judges, and the expectations of his supporters. Hillary brings a determination to fully fund Planned Parenthood and do whatever she can to make sure no girl who ever decides one day that she wants an abortion has any reason to change her mind. I think that's going to be hard for a lot of people to brush off, once they've thought about it.

Anonymous Darth Dharmakīrti October 20, 2016 12:33 PM  

It is made ever so unlikely because of the multiple polls.

Not saying you're wrong, but this strikes me as missing the point. All of the pollsters seem to be making the same basic kinds of assumptions about their LV screen. This gets back to VD's point: if LVs in 2016 look like LVs in previous elections, the polls are right and Hillary wins easily.

If there's something wrong in the standard model of the LV, though, then the fact that "multiple polls" show similar results doesn't actually mean much of anything.

Blogger VFM #7634 October 20, 2016 12:34 PM  

@149 Richard Stone

Catholics who actually believe the religion's tenets are either for Trump, or are somewhat cucked and aren't supporting either candidate because of badfeels.

Catholics who would actually vote for thecunt are either actually atheists/agnostics of a Catholic background, or are heavily cucked Churchians. (This is why white Catholics who identify as Democrat are in steep decline, whereas those who identify as Republicans are holding steady.)

This is a bit confusing for non-Catholics because the Church hierarchy actually behaves more like the liberal pseudo-Catholics (viz., Francis, Blase Cupich, etc.), but that's because, as I've insisted all along, that it has been taken over by SJWs.

Anonymous Kevin October 20, 2016 12:47 PM  

Like the primary voters choosing Trump and Hillary the general voters are also insane faced with those options. Utah Mormons have gone off their rockers and are 30% for some guy no one has ever heard of just because he is a Mormon (A much smaller % of Mormons viters for Romney Than % of Blacks that voted for Obama - skin trumps culture and even religion). They voted for one Mormon and now they need to go for a Mormon every time - even one with no record, with a lesbian Mom.

My family leans heavy conservative. All voting 3rd party for random dumb reasons. One was practical Trump until "pussy" and then he lost that. I have him hell and I am the weakest Trump supporter ever but that crap is ridiculous.

What is my point? Only state polls matter. And I would not be surprised if the libertarian and Green Party pull 7-10%. It's a bizarre year and republicans are defecting faster from Trump than disgusted democrats from the criminal.

In the end it's not that Trump is boorish. That is a part of it but I think it's because he does not speak to their issues. He does not feel like their representative. But that will change as the decline continues and things get crazier. People are going to stop preening at the polls and want some change. But democrats are holding their ship together because a criminal they know is liberal is ok. A republican that is boorish and talks much of the time like a democrat unsettles them - even if he is the only one serious about lots of issues they care about. Strange times.

Blogger Salt October 20, 2016 12:48 PM  

Trump live NOW in Ohio

https://www.youtube.com/watch?v=Ki6lvK1_Hpg

Blogger Cail Corishev October 20, 2016 12:51 PM  

I realize liberal Catholics feel no shame about voting Democrat despite the Church's clear teaching that abortion is non-negotiable. But some of them, especially the bishops, might object to being embarrassed by the leaks, enough that they don't preach as enthusiastically the usual wink-wink "we must consider all aspects of life" dodge. I dunno, it's just a possibility I'm watching for.

Anonymous Yay Trump October 20, 2016 12:54 PM  

They don't have issues. It's pretty obvious the whole college-educated white collar pool is coalescing around a neocon type liberal worldview. College-educated Republicans would much rather lose and lose and lose the Presidency as long as the R candidate managed to pull a majority of them, as Romney did. Because that's their tribe, not whites or Americans or white Americans.

Anonymous #8601 October 20, 2016 1:06 PM  

@159 Salt

About 30,000 people watching on Youtube. I wonder if it's the same people that watch every time? I confess I've watched more than few. The side chat is usually even more entertaining than the speech itself!

Anonymous Dave October 20, 2016 1:07 PM  

The media tried to save Carter in 1980, and only made themselves look stupid. In 1984 and 1988, they removed their polling bias as the election approached and it looked obvious that the Dem was going to lose.

Those lessons have since been forgotten, and it doesn't matter as today's media is mostly non-profit anyway. Pleasing the billionaire who bought your newspaper for one dollar and signs your paychecks is more important than informing the ten people who still read your birdcage liner.

Blogger Salt October 20, 2016 1:08 PM  

The debates constrain Trump. The rallys are pure Trump.

Blogger John Williams October 20, 2016 1:09 PM  

Remember, he pushes, then coasts, pushes, then coasts. It looks like he's gearing up for the stretch run.
It's wave theory. You can't push all the time, you'll burn out, the people will burn out, and it'll crash & burn. Like any good book, movie, song, etc their will be lulls before the next action. DJT is controlling the peaks & valleys and looks to be timing the next peak for Nov 8th.

Blogger Sheila4g October 20, 2016 1:10 PM  

@110 Natalie: Congratulations and best wishes on the forthcoming child. Do you not go to indoor ranges while expecting because of personal concerns, or the volume level, or do ranges place restrictions? I've just started going and it is a good way to release stress, although I still find a hard workout to European metal works better for me (while I glare at all the mystery meat).

Blogger Salt October 20, 2016 1:11 PM  

People are at Trump's rallys because they want to be. People are at Hillary's rallys because their being paid to be.

Blogger Cail Corishev October 20, 2016 1:17 PM  

The debates constrain Trump. The rallys are pure Trump.

He's done very well so far with rallies and social media, but I wonder if he'll buy some TV time, Perot-style, to rally more people that way. Just broadcasting a rally live could be yuge, and it would put the networks on the spot in an interesting way if they tried to refuse him.

Anonymous Ominous Cowherd October 20, 2016 1:20 PM  

dh wrote:This is basic mathematics. You correct for sample misdistribution.

Here is what they wrote about it in the actual poll report: ...


Some of those adjustments require making assumptions. Some of them require putting large weights on small samples. Those are both ways one can go way wrong.

My understanding of what you are describing is that they might sample 1,000 people, find that only one was (R), and give that one (R) a weight equivalent to 300 respondents, if they were adjusting to get a 30% (R) response. That's a ridiculous example, but it illustrates the potential problem with giving large weights to small portions of the sample.

Experiment design and sampling is not my field. Can you recommend a decent introduction to how this sort of weighting is handled when doing the statistical analysis?

Also, I see confidence intervals and margins of error. Are these different? Does confidence interval have the usual frequentist meaning? What is the definition of margin of error?

Blogger Natalie October 20, 2016 1:22 PM  

Sheila4g wrote:@110

All of the above. Indoor ranges are not generally recommended during pregnancy because you're exposing your child to some unpleasant substances. Even with a good air filtration system most indoor ranges have signs posted warning pregnant woman against shooting. I'd be fine shooting outdoors - particularly in the winter when you're wearing layers and can just strip off your outer layer and wash up well afterwards.

Hard workouts are out for now too :/ I'm still waiting for my energy levels to kick back in - I walk and do some bodyweight stuff, but that's about it right now.

Blogger bosscauser October 20, 2016 1:27 PM  

You got it...we Independents hate both parties. Both parties hate Trump....
Indies (swing) voters always decide elections since we went Reagan..

Let's go#Trump and hurry!

Blogger bosscauser October 20, 2016 1:27 PM  

You got it...we Independents hate both parties. Both parties hate Trump....
Indies (swing) voters always decide elections since we went Reagan..

Let's go#Trump and hurry!

Anonymous A Deplorable Paradigm Is More Than Twenty Cents October 20, 2016 1:27 PM  

Sheila4g
Do you not go to indoor ranges while expecting because of personal concerns, or the volume level, or do ranges place restrictions?

Probably Natalie doesn't go to indoor ranges because of lead issues. Cartridge primers contain a lead compound that is vaporized at the moment of firing. Outdoors the vapor is dispersed quickly, indoors it hangs around no matter how good the ventilation. Breathing vapor with lead in it will get some lead into her bloodstream, and that's not good for the baby.

Washing hands in cold water after shooting to get any residue off of them, while keeping skin pores closed, is a good idea. Some people take wipes to the range and just wipe down their hands and face, that's not a bad idea.

Lead-free primers exist, they are not as reliable, expensive, and so not all that common.

She probably should restrict herself to dry fire only until the sprog is out, but that is up to her husband, not me.

Anonymous rienzi October 20, 2016 1:28 PM  

122. Sherrif T.: "Not to the democrats. Do you really trust that polling data to remain private?"

" It hasn't been released ever before. Releasing that information would be a guarantee of the polling organization going out of business. Yoiur claim is absurd."

It doesn't matter if your response is actually as secure as nuclear launch codes. What matters is a respondent's perception of their response's security. If they think it might cause them trouble they shut up.

Anonymous A Deplorable Paradigm Is More Than Twenty Cents October 20, 2016 1:30 PM  

@149

The left Catholics always had contempt for their Bishops.


Dunno why, the US Bishops have always been very solidly in favor of Social Justice, worked hand in hand with the Communists in the 1980's, think it's great to give communion to women who have had abortions or divorces. What's for an SJW not to like?

Anonymous A Most Deplorable Paradigm Is More Than Twenty Deplorable Cents October 20, 2016 1:37 PM  

@122

If some unknown person called you on the phone and asked how many guns you own, what would you tell them?

Blogger Natalie October 20, 2016 1:42 PM  

Deplorable Paradigm: I also feel that shooting in the back 40 is less of a problem than an outdoor range, but the back 40 is a ways down the road from us currently :/ You are reminding me that now would be a good time to clean my EDC and practice dry firing.

Anonymous Ominous Cowherd October 20, 2016 1:46 PM  

A Most Deplorable Paradigm Is More Than Twenty Deplorable Cents wrote:If some unknown person called you on the phone and asked how many guns you own, what would you tell them?

Since Sherrif T assures me my responses are confidential, I would tell the anonymous caller posing as a poll taker how many guns, and my work schedule, and the combination of the gun safe. What's to lose? And think of the potential upside!

Seriously, back when I had a landline I screened calls, and hung up on pollsters that got through. Why waste my time for no possible gain?

Anonymous That Would Be Telling October 20, 2016 1:51 PM  

@147 dh:

11% of blacks would be what Pres. GW Bush pulled in from blacks on his second term, when he beat Sen. Kerry handily.

By 16 electoral votes and, according to Wikipedia and a "margin of victory in the popular vote [that] was the smallest ever for a reelected incumbent president", does not say "handily" to me, especially against someone like Kerry.

The slim margins of 2000 and 2004 are one of the reasons I've never taken Karl Rove's reputation as a political genius seriously.

Blogger Frank Brady October 20, 2016 1:53 PM  

@138. dh

Tad, is that you?

Blogger James Dixon October 20, 2016 1:57 PM  

Folks, please remember a couple of things. One, from memory, dh isn't a native American (please correct me if I'm wrong, dh). Two, his experience with polling only goes back to the 1990's (again, correct if wrong). Both of these make him far more likely to take polls at face value than the people who have watched polling since the 1960's or so.

> State-based polls, which have similar samples but much smaller populations, have much lower MoE, which is why you should use them by now, not national polls.

And yet the state polls were all over the place for Obama-Romney, while the national figures were what the margin turned out to be. And yes, I remember that because we discussed it in detail during the election.

Blogger Snidely Whiplash October 20, 2016 2:23 PM  

@dh
Jimmy Carter would like a word with you.

Anonymous Sam the Man October 20, 2016 2:40 PM  

Dh,

On your observation that perhaps I am seeing anecdotal evidence: Possibly true but it is the overall direction of every indicator being in a republican direction, with no reflection of that showing up in the polls.

In 2012 Obama won PA by 5.2 %. The poll you reference shows Hillary winning with 6%. Hillary is not as popular as Obama was in 2012.

In my county Obama won by 1.2 % in 2012. The most recent in depth poll of the 3 counties outside of Philly showed Hillary winning by 11%. Need I say more?

Neither result reflects reality, especially with a +232,000 republican registration (Delta republican gains/Dem losses) since 2012. That is 4.6 % of the electorate, and those that changed their registration can be expected to vote if they made that effort, which makes it 8% of the expected turn out (58% of registered voters in 2012). That assumes blacks turn out in the 66% range, when in fact if they go back to their traditional norm in PA they will only come out at around 55%.

Further PA has not moved sharply browner. In fact the electorate was 80.8 % white in 2012, now it is 81.5%, which may reflect simply registrations, but it does not bode will for the left.

It is on that statistical basis I can say something s wrong with the local PA polls, and I have used no anecdotal evidence....so far.

Anecdotal evidence.

None of the above would matter if there was a ground swell of support for Hillary. There is not and I have seen such as in 2008 when there was one for Obama. When I look a the number of election signs, there is no way Hillary is more popular in my county in 2016 than Obama was in 2008 or even 2012. The same is true of folks I talk to, I can see folks who voted democratic for Kerry in 2004 who are not voting for Hillary this time. Same is true of folks that did not come out to vote in 2012, because though they did not like Obama, they though Romney was just in it for the country club set. They are coming out to vote for Trump. The ground game for Trump is much more impressive than for Hillary, and is reflected in the amount of activity I see at the headquarters in the county seat (I live close by)

The fact that the polling is so far off what I see and can statistically get my hands on tells me the polling organizations, even if they are sincere, are missing something.

Anonymous Yay Trump October 20, 2016 2:51 PM  

Also, Hillary's camping out on PA, NC, FL and OH. Not exactly the states you'd pick if you really believed Trump was "putting red states in play".

Anonymous BigGayKoranBurner October 20, 2016 3:00 PM  

Johnny 10:01 AM What is this GOPe you speak of?? Really, who are they? Do you think the ordinary (non leadership) self declaring Republican is more opposed

If you are not a SOROS TROLL & don't know how to google: The GOPe is the crony establishment like the leftist establishment that stole the nomination from Bernie. They are the cuckservatives who would rather lose than have crony handouts cut off & would rather lose than be called racist.

They are the people who got caught paying blacks to vote in a primary against an actual conservative candidate in favor of a cucked milk toast NYC jewish lawyer. If you ever wondered why Lindsey Garaham keeps getting to be the only choice for Rebupicans, despite common knowledge of his bathhouse activities (yes I know they Identify as gyms snopes, Hairy Reed's eye) it's because the GOPe would rather have corruption than keep mad moslem men out of bathrooms they are not welcome in(little girl's and gay bar's).

Anonymous BGKB October 20, 2016 3:03 PM  

Sorry I misspelled (((GOPe)))

Blogger Snidely Whiplash October 20, 2016 3:04 PM  

The basic problem with state-level polls is they want to capture "likely voters". That's a good thing, but how do you determine likely voters?
Some polls, like the LA Times survey, simply ask. Some ask question like "Did you vote in 2008 and 2012", or use voting records. Some don't bother and just use demographics to re-weight the sample.

Most of these methods are going to understate the support among people who don't normally vote. That is exactly what they are designed to do, cut out people who don't vote.

And that is why they are off. How much, neither you nor I not the polling companies nor dh knows, but they are off.

Anonymous BGKB October 20, 2016 3:12 PM  

Further PA has not moved sharply browner. In fact the electorate was 80.8 % white in 2012

Refugee Resettlement Watch reported that Lancaster PA, famous for it's Amish had so many 3rd world refusegees shitted on them that in a 3 year period the city became 4% refusegee dumped directly onto welfare to the point that the good cucks said they got to many for the "safety net" Voter fraud happens in cities with substantial non Asian minorities because everyone involved is OK with it, as seen in Project Veritas videos. Lancaster PA gets mentioned in RRW almost as often as Lewistown Maine.

http://lancasteronline.com/news/local/lancaster-county-ranks-rd-in-pennsylvania-for-refugee-resettlement-in/article_22003e76-5bbe-11e5-8c93-0bdbca2e5b6c.html

https://refugeeresettlementwatch.wordpress.com/2012/05/11/so-what-is-going-on-in-lancaster-pa-more-refugees-than-the-city-can-handle/

"May 16, 2012 at 8:36 am
“Pennsylvania has “welcomed” 63,142 refugees in the last 15 years putting it in the top ten states for “welcoming” refugees.”

"“CWS is dumping refugees on to the city”

Many Lancaster people that I talk with…who also try to help refugees here… use the word “dumping”… CWS is dumping refugees on to the city… taking a payment to settle the refugees and moving on to the next group of refugees who will bring in more money to pay for CWS salaries, office space and fund raising events. This is the impression of many compassionate, frustrated people here in Lancaster, who are then accused of being part of a racist or anti-immigration backlash. We are not anti-immigration. We are expending our own personal time and money and pressuring our churches to spend time and money, picking up the pieces that CWS drops into our community. We now want a say in how many refugees can be settle here."

Anonymous Sam the Man October 20, 2016 3:33 PM  

Maybe true, but they cannot vote...yet.

Anonymous BGKB October 20, 2016 3:36 PM  

I realize liberal Catholics feel no shame about voting Democrat despite the Church's clear teaching that abortion is non-negotiable.

Did you miss the WikiLeaks emails describing the plans and ongoing SJW entryism into the catholic church?

The immediate reaction I got upon pointing that out was "but Trump is just saying that, he wasn't pro-life before."

Tell them he is a grandfather now, & abortion could have robbed him of that.

Anonymous BGKB October 20, 2016 3:38 PM  

Maybe true, but they cannot legally vote...yet, just like the Turk moslem illegally did 3 times before killing 4 women in a Washington mall, tweeting he voted for HERpes in the primary.

FIFY

Blogger Unknown October 20, 2016 4:16 PM  

Any poll that uses the 2012 election as a base line for anything is immediately flawed. Why?

2014. What happened then? Oh yea, a historic 7 member Senate flip.

2012 presidential election data does not account for Trump's enthusiasm advantage, the new or long time lapsed voters coming out for Trump, the inevitable lower turnout for black voters, or the lack of enthusiasm for Clinton.

Dh can try to explain away that obviously flawed AZ Repub poll all he wants to, but no way in Arizona do you get a 58% to 24% Dem advantage for respondents. Not without deliberately choosing your poll respondent pool to produce that result. They can say they randomly called numbers all they want to, but randomly picking numbers means nothing if the pool of potential respondents was heavily biased to begin with.

Try to tell me that pollsters don't trade phone # lists amongst themselves, and I'll call you a liar to your face. Telemarketers have lists of good phone #s, so do pollsters. They know who to call to get a reliable poll result.

Blogger VFM #7634 October 20, 2016 4:28 PM  

Did you miss the WikiLeaks emails describing the plans and ongoing SJW entryism into the catholic church?


@BGKB
I don't understand why they don't realize they already did it in the 1950s and 1960s with the Second Vatican Council. But nobody ever said SJWs were all that bright...

Blogger Franz Lyonheart October 20, 2016 4:31 PM  

The reason is that regardless of unpopularity when you have a geographic based winner-take-all system (as we do) a two-party dynamic always results.

That's false. America has inherited the first past the post British voting system, but in Britain, there's many, specifically geographically focused, parties in Parliament.

There's different parties in England, Scotland, Wales, and northern Ireland, and different two horse races whether is country or city. Eg in affluent London suburbs, the two horse races often are Conservative vs Lib Dem, who alternatingly win, with UKIP as third place, and Labour not even in the running; whereas in most of the country it's Conservative versus Labour.

The Republocrat/ Democrican stranglehold on the USA is supposed by first past the post, but that's not the sufficient factor, there's much more.

Eg, the state run primaries. How can it be that individual states' governments are so closely intertwined that they a) register party affiliation (R, D, I; but not Libertarian or others?), that they b) organise state level primaries for these two parties? I'm not sure if they even use taxpayer funds to organise and administer the two-party primaries?

Let alone MSM and other factors.

The voting system may be a contributing factor, but it doesn't explain this quasi aristocratic dominance. The rot goes much deeper.

Anonymous A Paradigm Is More Than Twenty Deplorable Cents October 20, 2016 4:31 PM  

Madonna promises BJ's for Clinton voters

Blogger Franz Lyonheart October 20, 2016 4:32 PM  

* supposed by = supported by.

Blogger James Dixon October 20, 2016 4:32 PM  

> 2014. What happened then? Oh yea, a historic 7 member Senate flip.

2010 was equally historic. Obama still won re-election. Don't count your chickens yet. But Hillary isn't Obama. She's not going to get the turnout he did, so there is cause for hope.

Blogger Franz Lyonheart October 20, 2016 4:36 PM  

Madonna promises BJ's for Clinton voters

Yucky, way too late.

Maybe 35 years ago, when she was just like a virgin.

Then again, at that age I was a schoolboy and girls were still yucky.

So, yucky either way.

Anonymous Triggerman October 20, 2016 4:52 PM  

This guy created a model that has successfully predicted all but one presidential election result dating back to 1912:

"The prediction formula of the Primary Model, as shown in Table 4, leads to this forecast: In the match-up between the Republican and Democratic primary winners, Donald Trump will defeat Hillary Clinton with 52.5% of the two-party popular vote, with her getting 47.5%.[i] It is 87-percent certain that Trump, not Clinton, will be the next President. Trump benefits from a swing of the electoral pendulum to the Republican side in 2016 and his superior performance in early primaries. While Trump won the Republican primaries in both New Hampshire and South Carolina Hillary Clinton split the Democratic primaries in those states with Bernie Sanders."

http://primarymodel.com/2016-forecast-full/

Regardless, it likely pays not to get too caught up in the minutia. The incumbent party rarely wins after an 8-year run; the reason you don't hear that fact is that large swathes of the media rely on drumming up interest in day-to-day trivia that no one will remember in a week or two.

1 – 200 of 224 Newer› Newest»

Post a Comment

Rules of the blog
Please do not comment as "Anonymous". Comments by "Anonymous" will be spammed.

<< Home

Newer Posts Older Posts