Moon of Alabama Brecht quote
November 13, 2016

Why Polls Fail

Today I discussed the U.S. election with a friend who studied and practices statistics. I asked about the failure of the polls in this years presidential election. Her explanation: The polls are looking at future events but are biased by the past. The various companies and institutions adjust the polls they do by looking at their past prognoses and the real results of the past event. They then develop correcting factors, measured from the past, and apply it to new polls. If that correcting factor is wrong, possibly because of structural changes in the electorate, then the new polls will be corrected with a wrong factor and thus miss the real results.

Polls predicting the last presidential election were probably off by 3 or 5 points towards the Republican side. The pollsters then corrected the new polls for the Clinton-Trump race in favor of the Democratic side by giving that side an additional 3-5 points. They thereby corrected the new polls by the bias that was poll inherent during the last race.

But structural changes, which we seem to have had during this election, messed up the result. Many people who usually vote for the Democratic ticket did not vote for Clinton. The "not Clinton" progressives, the "bernie bros" and "deplorables" who voted Obama in the last election stayed home, voted for a third party candidate or even for Trump. The pollsters did not anticipate such a deep change. Thus their correction factor was wrong. Thus the Clinton side turned out to be favored in polls but not in the relevant votes.

Real polling, which requires in depth-in person interviews with the participants, does not really happen anymore. It is simply to expensive. Polling today is largely done by telephone with participants selected by some database algorithm. It is skewed by many factors which require many corrections. All these corrections have some biases that do miss structural changes in the underlying population.

The Clinton camp, the media and the pollsters missed what we had anticipated as "not Clinton". A basic setting in a part of the "left" electorate that remember who she is and what she has done and would under no circumstances vote for her. Clinton herself pushed the "bernie bros" and "deplorables" into that camp. This was a structural change that was solely based in the personality of the candidate.

If Sanders would have been the candidate the now wrong poll correction factor in favor of Democrats would likely have been a correct one. The deep antipathy against Hillary Clinton in a decisive part of the electorate was a factor that the pseudo-science of cheap telephone polls could not catch. More expensive in depth interviews of the base population used by a pollster would probably have caught this factor and adjusted appropriately.

There were some twenty to thirty different entities doing polls during this election cycle. Five to ten polling entities, with better budgets and preparations, would probably have led to better prognoses. Some media companies could probably join their poll budgets, split over multiple companies today, to have a common one with a better analysis of its base population.One that would have anticipated "not Hillary".

Unless that happens all polls will have to be read with a lot of doubt. What past bias is captured in these predictions of the future? What are their structural assumptions and are these still correct? What structural change might have happened?

Even then polls and their interpretation will always only capture a part of the story. Often a sound grasp of human and cultural behavior will allow for better prediction as all polls. As my friend the statistician say: "The best prognostic instrument I have even today is my gut."

Posted by b on November 13, 2016 at 03:17 PM | Permalink

Comments
next page »

An equally interesting question about polls: what about the exit polls? If Greg Palast and others are right, exit polls indicate that the voting was rigged. What does your statistics friend think about that?

Posted by: Oscar Romero | Nov 13, 2016 3:23:53 PM | 1

After the 1948 election, statisticians started to get rid of the quota sampling for electoral polls. After this election, it's time to reassess Statistics.

https://www.math.upenn.edu/~deturck/m170/wk4/lecture/case2.html

Posted by: Andrea | Nov 13, 2016 3:28:21 PM | 2

NeverHillary turned out to be bigger than NeverTrump. Hillary got less than 6 million votes compared to Obama. Trump got nearly as much as Romney.

Posted by: ab initio | Nov 13, 2016 3:30:01 PM | 3

A good indicator was the size of the crowds each candidate drew to their rallies. Clinton tended to show more "bought" TV-ready extras. Bernie blew the walls out at his rallies, as did Trump. You can't look at that and say the polls are even close to accurate.

Posted by: stumpy | Nov 13, 2016 3:45:38 PM | 4

I suspect that the future of polling isn't as dire as you're painting it, b. There was huge anti-Trump bias in the Jew-controlled Christian-West Media from the beginning of the campaign. You drew attention to negative MSM bias yourself in the post which pointed out how consistently wrong the Punditocracy had been in predicting the imminent failure of the Trump campaign - thereby rubbing their noses in their own ineptitude and tomfoolery.

One factor which seemed important to me was occasionally hilighted at regular intervals by commenters here at MoA... The (apparent) fact that Trump addressed more, and bigger, crowds than Mrs Clinton. I accepted those claims as fact, and didn't bother to check their veracity. But nevertheless crowd size and frequency seems to have played a pivotal role in the outcome (as one would expect in a political campaign).

Posted by: Hoarsewhisperer | Nov 13, 2016 4:00:50 PM | 5

Exit polls have provided checks on the accuracy of the vote count -- but are liable to the same problem as the opinion pols, people who don't admit to their real position.

Posted by: Mudduck | Nov 13, 2016 4:01:15 PM | 6

I'm not surprised that the polls fail badly in this presidential election. When the Democrats unleashed thugs on Trump supporters while the media studiously looked away, it was not sensible to openly identify with Trump. Even Trump was saying so through out the campaign.The Democrats together with their media partners truly believed that Donald Trump's alleged character flaws would be enough to win the election. Despite the fact that it was obvious to anyone without a blinker on that the momentum was on the side of Trump all along. Obama's phenomenon of 08 was nothing compared to Trump's phenomenon of this year, but because neither the MSM nor the Pollsters liked him they transferred their biases to their jobs. In any case I'm sure happy that the result of the election turned out different from the skewed prognosis.

Posted by: Steve | Nov 13, 2016 4:03:18 PM | 7

On Wednesday after the election, I heard an interview with a woman reporter who worked with the 538 polling group. She said that it was impossible for most reporters to really investigate how voters in certain areas of the country were feeling about the election bcz newspapers and other news organizations, including the Big Broadcasters, did not have the ability to pay for enough reporters to actually talk to people.

Since statistics had worked so well, and were cheaper to deal with, they won the day. And lost the battle.

Now, most people at this site seemed to base their decisions of whom to vote for based on stands on issues and known actions of the various candidates. But, even so, we probably paid attention to the polling results. I know I took into consideration that Hillary would win big in NJ, leaving me free to vote for Jill Stein. Based on known actions of Trump I could not vote for him, even tho' I hoped he would kill TPP and have better relations with Russia. I feared and still do fear his nominations to the Supreme Court. (I am not religious, but if I were I would pray daily, perhaps hourly, for the continued good health of the Justices Kennedy, GInsburg, and Breyer. I would hope the other Dem appointed justices would take care to avoid, oh, small airplanes....

Would Hillary have adjusted her campaign if she could have seen the rising disappointment of the working class Dems (even middle class to higher income Dems)? I don't know. I do know that her husband ran his first campaign on the famous "It's the economy, stupid" reminder.

Somehow, I don't think it would have registered enough.

And Obama ran on Hope and Change, but was always the Corporatist Dem Wall Street wanted. What a waste. And now we have four more years of doing essentially nothing aboug climate change. It was have been a strategy to put off even regulatory actions to lessen CO2 emissions until near the end of his second term, but, dang, it makes it easier for Trump to negate those efforts.

Again, what a waste. But I didn't vote for Obama for either term bcz I saw that his actions as IL state senator and as US senator were always looking out for the Big Money, Big Corporations, and seldom worked for anyone below the middle class, more the top of the middle class.


Posted by: jawbone | Nov 13, 2016 4:08:45 PM | 8

This Wasn’t A Vote, It Was An Uprising
Paul Craig Roberts • November 12, 2016 >

Polls mean nothing when there is an uprising

Posted by: virgile | Nov 13, 2016 4:12:32 PM | 9

No need of polls...say PBS

How to (accurately) predict a presidential election

Posted by: virgile | Nov 13, 2016 4:15:15 PM | 10

A long explanatory report which signifies nothing critical. "The polls were wrong??" No.
The polls reported by MSM were wrong.

Big time, including from those from Clinton loving CBC here in Canada, which for an extended time was reporting Hillary with an 11% lead. That number was far beyond any minor adjustments, for sure.
There were polls, such as Rasmussen, itself suspected of fiddling, which were reporting ups and downs of 2%, and ended up tied election day.

So, please schemers, please do not try to cover up the MSM's deliberate attempt to influence results by using garbage numbers. Figures can lie, and liars can sure figure.

Posted by: joey | Nov 13, 2016 4:19:53 PM | 11

the only way to predict is to use a crystal ball (Scrying)

Posted by: brian | Nov 13, 2016 4:20:19 PM | 12

the Los Angeles Times polls were correct (although the paper was pro-Clinton); can't get the link now, but they explained how they weighted their polls on the basis of the enthusiasm displayed for the preferred candidate, and Trump supporters were more "charged"

Posted by: claudio | Nov 13, 2016 4:23:05 PM | 13

I disagree with your friend, b. I read many stories about how the polls were fixed for Clinton for months before the election. The pollsters took the % of voters from the Obama election but they also added more Democrats than were representative in the 2012 election, thereby skewing the polls for Clinton. Many believed that the reason they did this was to try to manipulate the voting machines in Clinton's favour and have the polls match the result. I think that Trump crying foul so early got them worried that they might be caught. Remember, voting machines in 14 states are run by companies affiliated with Soros.

Posted by: mischi | Nov 13, 2016 4:25:01 PM | 14

i go back to what my sociology of the media instructor said.. polls are for massaging people's brains.. unless one knows who pays for them and what goes into them, they are just another propaganda tool for use..

Posted by: james | Nov 13, 2016 4:26:58 PM | 15

...
Polls mean nothing when there is an uprising
Posted by: virgile | Nov 13, 2016 4:12:32 PM | 9

Well, the Clinton-ista's and Soro-fuls certainly wasted no time when they switched from Anticipated Gloat to Full Spectrum Panic Mode, did they?

Posted by: Hoarsewhisperer | Nov 13, 2016 4:34:08 PM | 16

It has been known for a long time in the polling world that polling numbers are getting more and more unreliable because fewer and fewer people are willing to complete polls.

Posted by: BraveNewWorld | Nov 13, 2016 4:35:07 PM | 17

I have a weird conspiracy hypothesis that I mainly made up on my own;

The last FBI "reopening" and the quick subsequent "close-down" felt all too counter-intuitive and silly, when examined solely based on their face value.

However, what if there was more to this? What if this was a final threat from FBI to the Soros-Clinton mafia to "quickly unrig the voting machines" OR we will arrest the lot of you? Which, once the promises were made by "allow fair play", required FBI to pull back as their part of the deal?

Just an idea...

Posted by: Quadriad | Nov 13, 2016 4:41:10 PM | 18

This - admittedly conspiracy - theory would also explain the newspaper polls largely rigged to correspond to the planned vote theft, as well as the idiotic magnitude of overconfidence seen in the Pol-Est/MS Media/Wall Street complex.

Sorry on the split-think and double-post.

Posted by: Quadriad | Nov 13, 2016 4:43:54 PM | 19

I find it interesting b that you and your friend didn't seem to talk at all about the polling questions....at least that you shared with us. It is my experience and education that even with a "beauty contest" that we just had, that the structure of the polling questions make all the difference in how people being polled respond.

Polls are funded by parties with agendas and the questions, assumptions and biases are baked in to the result......IMO, they are all worthless or worse than that because folks see them, like the media as being something of an authority figure and therefore believable which we know is total BS.

Polls are just another propaganda tool of those rich enough to use them in their quiver of control.

Posted by: psychohistorian | Nov 13, 2016 5:02:02 PM | 20

Timid Trumpists is the major factor, I would think. A factor already well known in UK. People who are going to vote for a non-PC solution hesitate to admit it to poll questions.

Posted by: Laguerre | Nov 13, 2016 5:15:19 PM | 21

All of the above is true, but - in addition - polls are used to manipulate campaigns.

People sympathize with someone who is considered a winner and when someone is considered likely to lose people lose interest.

To get the vote out polls have to be tight. In addition to that polls are used to motivate donors. In the end there has to be a reason pollsters get paid.

But even if polls would be done for purely scientific reasons, this election was impossible to poll. The correct question would have been "Do you hate/fear candidate x enough to motivate you to queue for voting for canditate y, or are you too disgusted to bother at all"

In the end, it was not the wrong polls that sank Clinton but the strategy to leave the anti-elitist populist stuff to Trump and - unsuccessfully concentrate on winning the elitist Republican anti Trump vote. That way she lost more of the Democrat Sanders vote than she could gain right wing.

The other factor was her reliance on television ads and media ties (they all backed her), a reluctance to talk to large audiences and an inability to communicate via social media.

It is possible though she never had a chance against a well established reality show brand.

The good news is that after this election campaigns will be done mainly low cost social media. The bad news is that these campaigns will be more fact free than ever and that the age of independent quality newspapers is over.

Posted by: somebody | Nov 13, 2016 5:15:50 PM | 22

#22 somebody

So, you're saying that the age of independent quality newspapers has just ended, like about now. Interesting pov...

Somehow, the last few years of the MSM coverage of the NATO-Salafist War on Syria have had me convinced that the "independent quality newspapers" have become a*rse-wipe material a long time ago. Instead, we get the Sorosoid ZioTakfirism.

But, yeah, maybe it's all Trump's fault. Hey I also blame Hezbollah for kicking Yisrael's arse north of Litani in 2006. If they didn't piss of the Yivrim this much, maybe they wouldn't have punitively collapsed the faith in the Western Society from the inside.

Ultimately, it's all Putin's fault. He started it all by beating the pro-Saudi Chechens into a pulp back in 1999, and started the NATOQAEDA self-destruction.

Posted by: Quadriad | Nov 13, 2016 5:29:03 PM | 23

In this election, Pollsters got it wrong.

Two Exceptions:

1. IBD/TIPP (A collaboration of Investors Business Daily and TechnoMetrica Market Intelligence). TechnoMetrica was consistent throughout – final poll for election day had Trump leading by 2%. Also predicted the last presidential elections back to 2004.

Methodology
“Traditional Telephone method” includes cell –live interviews by
Region; Age; Gender; Race; Income; Education; Party; Ideology; Investor; Area Type; Parental Status; White – men, women; Black/Hispanic; Women-single, married; Household description –Upper/Middle-Middle, Working, Lower; Religion; Union Household; Intensity of Support.

http://www.investors.com/politics/ibd-tipp-presidential-election-poll/

and
2. LATimes


This election candidates’ crowd draw was a good indicator. It was very difficult to pre-program the Diebold machines. MSM polls were in the bag for Hillary, had her ahead. It backfired.

Is Newsweek embarrassed yet? They forgot some history. Truman-Dewey.
Madam President! How appropriate.

Posted by: likklemore | Nov 13, 2016 5:35:21 PM | 24

Some of b's posts regarding US politics seems naive but I chalk that up to his not being American. But this technocratic excuse for the polling is just wrong. b, what happened to your skeptical view of Western media????

Posted by: Jackrabbit | Nov 13, 2016 5:44:28 PM | 25

virgile @ 9: An excerpt: " It was about the union men who refused to sell out their futures and vote for a Democrat who is an agent of the One Percent."

And now, I fear, they still have no future.

James @ 15 said.." polls are for massaging people's brains.. unless one knows who pays for them and what goes into them, they are just another propaganda tool for use..

How true..

Trumps choices for his cabinet don't leave much room for positive change, for the millions of disaffected voters who put him in office. We'll see!

Posted by: ben | Nov 13, 2016 5:46:27 PM | 26

A bit about polling methodology explains the bias we've seen this election cycle. Typically, the polling samples are not big enough to be representative, so the results are corrected (weighted) based on the participant responses. The polls assume certain turnout percentages for different groups (Democrats, Republicans, Independents, rural, urban, ethnicity, gender etc.). A lot of the polls were weighting the polls with turnouts similar to 2012, corrected for the expected demographic changes over the last 4 years.

Poll weighing is a tricky business. This is why most polling has a 4% error margin, so it does not produce as accurate picture as is typically presented by the media. The error is not randomly distributed, it is closely related to the poll weighting. The weighting error was favouring Clinton in the polls as it assumed higher Democratic turnout, which ended up not being the case, she underperformed 2012 significantly and lost the election.

It is important to stress that the election results ended up within the margin of error (+-4%). The polls were not wrong, it is the media and the analyst who over-interpreted the data and gave Clinton the win where she did not have a statistically significant (<4%) lead. This is why if Nate Silver at 538 was consistently writing that the polls in many of the swing states were within the error margin, although favouring Clinton, and their election prediction still gave Trump a ~30% chance of victory. Other analysts were more careless (hello Huffington Post) and even made fun of 538 for giving Trump any chance of victory.

There is no way to make more accurate polling for the future elections as the accuracy of the poll is tied in to poll weighing, which is guesswork (although somewhat educated by the historical data). Short of forcing everyone to vote, election-to-election turnout will change and affect the accuracy of the polls.

Posted by: voislav | Nov 13, 2016 6:13:07 PM | 27

Some fun but sadly true.

https://www.youtube.com/watch?v=Fk8DDFE8v3I

https://www.youtube.com/watch?v=GLG9g7BcjKs


#8 this for you
https://www.youtube.com/watch?v=P-5Y74FrDCc&index=25&list=WL

#25 Yep

Posted by: jo6pac | Nov 13, 2016 6:18:04 PM | 28

#27 voio

Instead of interpreting every single of those Polls as plausibly biased on one side, why don't you take the entire population of Western MSM Polls, and see if their median predicted outcome vs actual final outcome difference is statistically significant?

I'd say you'd find their entire population to be likely biased at least to six-sigma level.

(I have no time to show this myself, just proposing someone's hypothesis, as a research idea for someone's M Sci thesis for example)

Posted by: Quadriad | Nov 13, 2016 6:18:19 PM | 29

I have lived in the D.C. area for the past 22 years with a land line phone and am listed in the White Pages. I have never been called by a pollster, although I am often called by political campaigns. I do not know anyone who has been called by a pollster.

Posted by: lysias | Nov 13, 2016 6:18:32 PM | 30

Whocares about the pollsters?
What about Trump now?
'rinse Prybus'? Sessioms? Giuliani? Newt?

Posted by: bbbb | Nov 13, 2016 6:29:18 PM | 31

Palast puts up good information that difference was good 'ole GOP voter purges.

Posted by: jdmckay | Nov 13, 2016 6:35:22 PM | 32

Are the polls done to discover "what's up", or are they done to project the view that one side is winning?

I go with the second view. That's what the 'corrections' are all about. The 'corrections' need to be dropped completely.


Unless that happens all polls will have to be read with a lot of doubt.

Mike Whitney posted a link to a guy who got it right ... Patrick Caddell; The Pollster Who ‘Got it Right’. His methods were not those of the captive pollsters.


More expensive in depth interviews of the base population used by a pollster would probably have caught this factor and adjusted appropriately.

No more 'adjustments' allowed. A desire to actually discover the lay of the land and to publish it is what's required. Good luck on getting that from the political class and/or their captive msm. Everything they do is a lie, calculated to keep themselves in power.

Posted by: jfl | Nov 13, 2016 6:40:08 PM | 33

The polls were obviously blatantly skewed towards urban Blue zones, and did not include working adults in Red zones, then were 'massaged' by reporting media in clearly a Rodham-paid PAC marketing campaign to brand the sheeples 'Wear Rodham!'
Only Nate Silver's FiveThirtyEight even came close, but he had to rely on those same skewed polls. After all, since 1990, you can buy a CD set of American voting records by street address, it's not rocket science to be able to 'algo' that into a 'poll' that skews whichever way the highest bidder's (Rodham) quants tell you to.
https://www.facebook.com/viralthread/videos/598130190359668/

Posted by: chipnik | Nov 13, 2016 6:42:19 PM | 34

jo6pak @28: Thanks for the videos.

On Tuesday a democratic site was taken down. This video was put up in it's place.

Strange and troubling. Seig heil anyone?

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

Posted by: ben | Nov 13, 2016 6:42:20 PM | 35

@likklemore #24:

Glad you said that, and much better than I would have.

@somebody #22:

polls are used to manipulate campaigns.

This. There was a Wikiliks Podesta email in whdich Clinton operatives discussed oversampling certain groups to inflate the poll in her favor.

Demian is now known as Adalbrand.

Posted by: Adalbrand | Nov 13, 2016 6:53:27 PM | 36

Oh Lookie – “Media Polls” Show Trump Back On Top, Go Figure…

As if on cue, or something. All of a sudden, S.U.R.P.R.I.S.E,… a litany of polls released today show Donald Trump ahead in key battleground states (Ohio and Florida), and tied –or closer than the margin of error– in new national polls…. […]

Remember what we stated on October 20th: […]

The real battle is the battle for your mind. The peak U.S. media false polling cycle is thankfully in the rear-view mirror.

It was because I followed that right-wing blog that I ignored all polls other than the LA Times tracking poll. (I didn't know about the IBD/TIPP poll until after the election.)

Posted by: Adalbrand | Nov 13, 2016 7:08:31 PM | 37

Hmm ... what can I say that no-one else has already said except to observe that the polling and the corporate media reporting the polling statistics were in another parallel universe and the people supposedly being polled (and not some over-sampled group in Peoria, Iowa, who could predict exactly what questions would be asked and knew what answers to give) live on planet Earth?

Posted by: Jen | Nov 13, 2016 7:12:20 PM | 38

I most certainly did not predict Trump would win. But I did question the polls.
What I questioned a few weeks ago was the margin of victory for Hillary. There were two big variables that the pollsters had to guess at. One was the voter turnout numbers for those precincts that had many working class people with a high school or less education level. As it turns out those people came out in higher numbers than they have in elections over the past two decades. The other was voter turnout for many precincts that supported Obama in 2008 and 2012. What happened here was many of those voters who did turn out voted for Trump, instead of the Democrat. There was a third uncertainty here that no on has yet figured out. That was those people who would never admit to a stranger that they were going to vote for Trump and simply lied to the pollster.

In any case those three uncertainties worked in directions that none of the pollsters really picked up on.

Posted by: ToivoS | Nov 13, 2016 7:18:03 PM | 39

#29 Quadriad

This is because most of the polls were weighting more Democratic (based on the 2012 election), which overestimated Clinton's support. For example, the Rasmussen poll, which traditionally weights more Republican, gave Clinton 1.7% lead, 44.8% to 43.1% (3% margin of error), so fairly close to the election results (47.3% to 47.8%). So the difference between the poll and the actual result is 1.2% in favour of Trump (1.7% lead to Clinton in poll vs. 0.5% in the election). All are well within the error of the poll, so 1.2% difference between the election and the poll is well within the stated 3% error margin of the poll.

When you mention 6 sigma, you really don't really know what you are talking about. Typical polling error is 3 - 4% and the election result was within this error for most polls in all of the states. Standard deviation (sigma) that you mention is a random uncertainty associated with a measurement and it does not apply here. As I tried to convey, the errors in polling tend to be systematic, not random, because they are tied to weighting of the polls, not to the sample of the population as this is mostly corrected by the weighting. So because most of the MSM polls use similar weighting methodology based on the same historical data, they will all be off, there will be no random distribution of some for Trump, some for Clinton. Weighing based on different historical data skews the whole picture one way, it's not a random error. This is why pollster slap a relatively large 3 - 4% error on their polls, it is meant to cover any systematic bias of the weighting as well as random errors.

Posted by: voislav | Nov 13, 2016 7:23:32 PM | 40

You assume public polls are conducted by impartial actors who wish to inform and illuminate.....

......your assumption is incorrect.

Posted by: bigmango | Nov 13, 2016 7:23:48 PM | 41

@ToivoS #39:

those three uncertainties worked in directions that none of the pollsters really picked up on.

Have a loook at the LA Times tracking poll. It had Trump ahead by 3.2% on election day, which is close to the margin of error. The graph there is interesting, because dates of various events, such as the debates are marked. The poll figures moved in response to those events as one would expect.

Before the election, the people who do that poll said that they did best at predicting the 2012 election. Oh, in a post about the election's outcome, Alexander Dugin singled out that poll for praise.

Posted by: Adalbrand | Nov 13, 2016 7:31:27 PM | 42

I have a better idea--how about we stop the stupid polling altogether since there is only one poll that really matters? Then the media would have to focus on the issues rather than the horserace. Oh, the humanity!

Posted by: Bill Hicks | Nov 13, 2016 7:44:37 PM | 43

I know exactly what I am talking about.

Hypothesis A - that it's all explainable by random distribution of their samples.

If you use Hypotethesis A, and then disprove it in it's own game (be it 3, or 6 sigma), then you have to suggest an alternative.

I don't know what the alternative is. I don't even claim I do. But you can more easily disprove the veracity that the polls could have mostly been non-biased by showing that hypothesis is unlikely to be RIGHT. That's where sigmas make absolute sense.

Nice try though, Voislave.

Posted by: Quadriad | Nov 13, 2016 8:08:16 PM | 44

Furthermore, what you are proving here is that the POPULATION of ALL COMBINED polls has a mean that must be different from the POPULATION of all actual voters, not of disproving the polls one by one.

I think you've totally ignored my point, you keep looking at individual polls as trees, I am looking at the poll forest and saying the entire forest is buggered if almost all polls erred on one side, regardless of their individual margins of error.

Posted by: Quadriad | Nov 13, 2016 8:12:56 PM | 45

The New York Times recent admission that it writes the narrative first, then builds the story to suit says about everything for me regarding polls. 'Hey, my editor needs someone to come out and say something, can you say this...?' <-- Now, if that is standard practice in journalism at 'the paper of record', then skewing polls to suit a common agenda is a given, again in my opinion. This of course is great news for sites like MofA.

Also impossible to capture The Don's campaign playing the electoral college system like an old mandolin, as it turns out. 306 Trump bts 232 Hillary it looks like in the wash up. That's old school work rate doing the job. Fair play. Great to see all the student debt laden brainwashed libtards out there doing there nut. They don't even know what a bullet they dodged + shite like the TPP is now dead. Some gratitude.

Hopefully in 2020 there are some more scientific polls like the USC Dornslife/LA Times poll, each having their own differing methodologies preferably. This should give the punters a better 'feel' for the electorate.

In other news...

Assange is being interviewed tomorrow by Swedush police (for the 2nd time I should add). There are and were no charges laid. I suspect their will be no charges brought tomorrow.

...so what happened...? Did The Rule of Law just...magically appear...?

Posted by: MadMax2 | Nov 13, 2016 8:14:16 PM | 46

The most extraordinary thing I learned about polls is that exit polls are altered as soon as the official election or primary vote is in-- to match it.

https://richardcharnin.wordpress.com/tag/mathematical-proof-of-election-fraud/

Posted by: Penelope | Nov 13, 2016 8:16:37 PM | 47

Polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls,polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls, polls and POLLS!

Since when did Polish people become so important in U.S. presidential elections? I mean, I enjoy kielbasa as much as the next guy but c'mon, nobody takes a plate of encased mince meat into the voting booth. Nobody says, "I'm voting 'cause the polls know how to make great sausage". Nobody holds up a sign that says, "Vote the Polls", especially since nobody running for office was Polish. Let's get a grip, people and keep things in perspective.

Now Spam, that could sway my vote. Spam, spam, spam, spam, ...

Posted by: PokeTheTruth | Nov 13, 2016 8:24:03 PM | 48

2 heartstopping items:
-- http://phibetaiota.net/2016/11/robert-steele-the-accidental-president-will-he-resign-the-closed-system-is-still-rigged-and-likely-to-remain-so/ Challenging Trumps legitimacy.
-- http://usdefensewatch.com/2016/11/putin-issues-international-arrest-warrant-for-george-soros-dead-or-alive/ This last-- like most overly dramatic news-- appears to be a scm but is widely dispersed across the web. Kind of curious. Of course I guess everybody knows that he's behind the protests in the US.

Posted by: Penelope | Nov 13, 2016 8:42:14 PM | 49

Who is Trump speaking to?

According to reports, the first leader Trump spoke to on the phone after his election victory was the Egyptian president, Abdel Fatah al-Sisi. Sisi congratulated him on the election victory, a spokesman for the Egyptian leader said.

Ireland’s government said the taoiseach, Enda Kenny, had a 10-minute call with Trump, and was invited to visit the White House on St Patrick’s Day.

Mexico’s president, Enrique Peña Nieto, has said he and Trump agreed in their call to meet before Trump takes office, while Israel’s prime minister, Benjamin Netanyahu, was invited to the White House.

Other leaders to have a chat with Trump so far include the Turkish president, Recep Tayyip Erdoğan, India’s prime minister, Narendra Modi, the Japanese prime minister, Shinzo Abe – they reportedly talked for 20 minutes and agreed to meet soon in New York – and South Korea’s president, Park Geun-hye.

Australia’s prime minister, Malcolm Turnbull, was reported to have chatted with Trump about security and trade in their call.

No surprises there.

It may be unfortunate, but I can see Trump & Erdogan getting along very well. Although, if they bring Putin into that triumvirate that could actually be very beneficial for the Middle East.

Notably absent

Francois Hollande (GONE)
Angela Merkel (GONE)
Mark Rutte (GONE)
Matteo Renzi (GOING?)

Posted by: Julian | Nov 13, 2016 8:54:34 PM | 50

@MadMax2 #46:

Concur with all your points. And yes, the timing of the Swedes finally deciding to interview Assange is funny.

I never thought that Hillary would become president, btw., from the moment she declared for 2016. Which is not to say that I was not concerned that the demonization of Trump might throw the election. We'll never know, but it is possible that Trump wouldn't have won without Wikileaks. And the two sets of leaks were very well timed.

To return to polls. It's not just most media polls that were off. The Clinton campaign's internal polls were off, too. They didn't have much doubt that they would win. (The same thing happened with Romney of course, but in their case, their internal polls differed from the media polls.) Apparently, they really did believe they have a firewall, with redundancies no less.

Clinton staffers: Arrogance from the DNC leadership cost Clinton the election

Posted by: Adalbrand | Nov 13, 2016 8:55:25 PM | 51

#45 Quadriad

Your Hypothesis A is wrong because the polls are not random. As I tried to explain, the poll results are highly dependent on the turnout assumptions built into the weighting model, it's not just the random distribution of people you talk to. You profile every person in the poll and you slot them into bins based on age, ethnicity, affiliation etc. Each bin is assigned a certain number of votes in the election, which changes depending on the assumed election turnout.

For this election the turnout rate was low, 51%, compared with 55% in 2012 and 57% in 2008. Most polls assumed that the turnout will be similar as in 2012, not 4% lower. That's 10 million fewer voters than expected. The turnout rates change with age and the changes in the turnout rates are mostly on the lower age range, which votes Democratic more than Republican. So the same exact poll of 1000 people will produce a different result based on whether I apply 51% turnout or 55% turnout rate for the election, because this changes the % of votes I assign to the 18-35 age group and how much I assign to the <65 age group.

If I have 20 pollsters do a poll of 1000 people each and apply the same 55% turnout rate weighting model, they will all skew Democratic because they will have a larger percentage of young voters which vote more Democratic. This is why a pollster runs the same polling data through multiple turnout models (typically 50%, 55% and 60% as the US turnout rates are almost always in this range), this is where the 3 - 4% error comes from, which covers all likely outcomes. But the headline number that is reported is the one for the 55% turnout and MSM ignores the caveat that comes with the error margin.

There is no conspiracy here, it's just that majority of polls use very similar weighting methods that are based on the turnouts for recent elections. This is why most polls overestimated the number of Clinton voters, they all used turnout models that were too high. It's a systematic problem, so it will skew all the polls that use that weighting model. To use your tree/forest analogy, the trees are not randomly planted, but they are planted in rows and each row is a different weighting model. The number of trees in each row will depend on how many polls use that particular weighting model. So if the model uses a higher turnout rate, it will be more to the Clinton side of the forest. In this election everybody overshot the turnout rate, so the forest was skewed to the Clinton side.

Posted by: voislav | Nov 13, 2016 9:09:23 PM | 52

@Penelope #49:

Thank you very much for that Robert Steele article. I will add that blog to my RSS application.

Posted by: Adalbrand | Nov 13, 2016 9:11:25 PM | 53

I don't think polling bias was meant to sway the voters.

I think polling bias was meant to sway the pundits and talking heads. These bozos were thus incentivized to be dismissive of Trump. At times media discussions were an echo chamber of Trump-hate.

What mattered was not 'margin of error' but that Hillary was in the lead, week after week. Hillary & Co. hoped to create an aura of inevitability that would lower the turnout for Trump.

Posted by: Jackrabbit | Nov 13, 2016 9:11:37 PM | 54

b,

I'm old enough to remember when scientific polling results were fairly accurate. During the reign of Bu$h II, it became apparent that polls were being skewed and that the results always reinforced and supported the neocon junta's latest objective like weapons of mass destruction or 911 and Osama bin laden is his gold lame' flowing robes. The Muslim Elvis.

Money talks bullsh*t walks.

Just my opinion

Posted by: ALberto | Nov 13, 2016 9:14:38 PM | 55

b
Your point about the adjustments and corrections made sound like what the government says about unemployment, or inflation, or the economy. You wonder what the numbers really are because they're massaged and adjusted so much.

Yes Oscar #1.
When the polls read differently in Ohio 2004, many of us pointed to election problems and there were. In the end we have a system that is better than some others (like Yemen 2012 w/Hadi) but has many problems that we as Americans should be able to fix. Unfortunately too many corrupt fingers in that pie to ever fix it.

Posted by: Curtis | Nov 13, 2016 9:25:51 PM | 56

Posted by: voislav | Nov 13, 2016 9:09:23 PM | 52:

There is no conspiracy here ...

What about this: ZH: New Podesta Email Exposes Playbook For Rigging Polls Through "Oversamples"
Earlier this morning we wrote about the obvious sampling bias in the latest ABC / Washington Post poll that showed a 12-point national advantage for Hillary. Like many of the recent polls from Reuters, ABC and The Washington Post, this latest poll included a 9-point sampling bias toward registered democrats.

Posted by: Jackrabbit | Nov 13, 2016 9:25:54 PM | 57

I found this article to be of little or no value because it was based on two assumptions, both of which I regard as completely bogus. The first is that any of the organizations that conducted these polls had any intention of conducting them in an honest and scientific manner. They were conducted to disseminate pure propaganda for the Hildabeast's campaign for two basic reasons. One is the assumption that a substantial percentage of the American voters wish to be on the winning side in order to identify themselves as “winners.” But more importantly was the intention of making the theft of the election more plausible and not on-the-face-of-it, totally fraudulent manipulation. The Clinton camp had their real clandestine pollsters who informed Podesta & Co. that Trump was headed for a landslide victory in an honest balloting. This came out in the Podesta Wikileaks.

The second flaw in the article is that the official ballot numbers have any resemblance to the actual numbers. I regard the official vote count supporting Trump’s victory as absurdly low. The question then arises, why did Trump win the “official” election at all? The Deep State had the tools to throw it to the Hildabeast without arousing disruptive suspicion from the bulk of the citizens, thanks to the bogus polls. Of all the various frauds favoring the Dems, and I will not enumerate them here, electronic fraud was the most powerful - rigging the numbers where they were centrally collected, the most notable of which is the GEM software backdoors via fractional votes. Bev Harris has laid this out brilliantly. So the question arises again, why did Clinton lose? One possibility is that the Trump vote was such a landslide that even all the Queen’s men couldn’t put it back together. This is possible, but in my opinion unlikely. I would weight as far more likely that a “rogue” branch of one or more alphabet soup agencies, and the NSA comes to mind first, hacked the Hildabeast’s hackers in a masterful manner. The entire controlled TV media and the Clinton campaign people stared on in total disbelief aa the results were tabulated despite the fact that they knew what the “real” ballot results were “supposed to be." The best explanation would be that the Dem’s hackers were outhacked and overridden. Not to give Trump the landslide he actually won, but just enough to give him the White House. Why? Well that could be the subject of endless, if fruitful, speculation.

But returning to the point of this comment, criticizing the polling results as if their errors where caused by honest technical misunderstandings in procedure is a waste of electrons.

Posted by: el Gallinazo | Nov 13, 2016 9:36:42 PM | 58

If you watch Trump's interviews, it was almost 30 years ago when he was asked for the first time about his presidential ambitions. And since then, one thing was absolutely consistent in all of his replies: "I think America is being ripped off". It comes without saying that he was referring to the USA's bloodsucking, freeloading "allies", as well as its parasitic trading "partners". He is most certainly correct about that (Lithuania spending less than one percent of its GDP on defense while faking mortal fear of Russia is scandalous, if not fraudulent). I fully expect Trump's administration to place NATO, South Koreans, Japanese and Australians on a horns of a dilemma - either they fork out $200B (annually) or so for the US military protection, or America's commitments will be considered null and void. Same goes for the trade predators like Germany and China. They'll have to stimulate domestic consumption and make sure that their trade surpluses dissipate, or America will tax their goods at the border.
If Donald has guts - and there is every reason to believe that he does - to follow through on his promises, he will become one of the greatest Presidents in US history, hands down.

Posted by: telescope | Nov 13, 2016 9:43:13 PM | 59

b

Maybe it's a lot simpler than your statistician friend describes it here, and it's just a matter of the media intentionally falsifying the poles, so as to demoralize and discourage the opposition to Hillary from coming out and actually voting. I understand that some private and in-house poles (even from among inside of Hillary's camp) overwhelmingly favored Trump to win.

Posted by: RayB | Nov 13, 2016 9:55:23 PM | 60

@RayB #60:

I understand that some private and in-house poles (even from among inside of Hillary's camp) overwhelmingly favored Trump to win.

What's that based on? Because I hadn't seen that mentioned anywhere. If it's true, then that that would mean that they expected to be able to steal the election, because apparently the Clinton camp was pretty sure she would win. And that brings up el Gallinazo's suggestion – that the deep state didn't allow her to rig the vote.

From the link that Penelope gave:

Hillary Clinton lost this race twice over after first over-coming deep documentation of her lack of character and deep personal corruption.[14] Her attempts to blame the Russians for hacking her emails (they did not, that was disgruntled NSA employees according to NSA whistleblower William Binney[15]) and for tampering with the ballot boxes (that was all her) backfired. She drew so much attention to electronic voting fraud that honest elements of the government closed those doors and she could not do to Trump what she did to Sanders. She probably did not actually win the popular vote; at least some of those votes were almost certainly residual fraud missed by the Department of Homeland Security (DHS) monitors.

Posted by: Adalbrand | Nov 13, 2016 10:21:58 PM | 61

b

Maybe it's all a lot simpler than what your Statistician friend thinks. And the distortions in the poles were intentionally done by the media so as to demoralize and discourage Hillary's opposition from going out and actually bothering to vote at the poles. (It's my understanding that privately conducted in-house poles, (even from among inside Hillary's own camp) strongly favored a Trump win.

Posted by: RayB | Nov 13, 2016 10:22:27 PM | 62

#57 Jackrabbit

There are and have always been polling firms that will do their party's bidding. Washington Post is now owned by Jeff Bezos (Amazon), who is a big Clinton supporter and has skewed the Post coverage heavily towards Clinton in this election, so I am not surprised they have skewed their polling too. But that has no bearing on the polls as a whole.

There was an industry wide skew towards Clinton in this election, even typically Republican-heavy Rasmussen overshot to Clinton by 1 point. This can be attributed either to a conspiracy or a systematic overestimate of the election turnout (as explained in my previous comment). My view is that the latter is far more likely as the turnout was historically low (lowest in this century).

Posted by: voislav | Nov 13, 2016 10:26:34 PM | 63

36

The correct spelling is addle-brained.

Posted by: chipnik | Nov 13, 2016 10:43:14 PM | 64

The Podesta email about "oversampling" certain groups like black americans or democrats to push poll numbers for media effect was probably true of some poll companies, that may have skewed poll aggregates used by Silver, Wang, etc.

Another problem is tuning the sampled groups to the last Prez election. Repubs:2008 59million,2012 60million,2016 Trump 59million. In 2008 Obama had 69 million voters!, 2012 = 65million, 2016 Clinton got only 59 million votes. Obama and Hillary drove out 10 million past democratic voters from 2008 to 2016, about 14%. So much for the vaunted Clinton/DNC big expensive "ground game".

It was hard to sell Hillary, like selling a shit sandwich, a lot of people just won't eat it. Pollsters either "oversampled" for media effect or used 2012 groupings and didn't believe their own lying eyes when no one showed up at Clinton rallies. Also most primaries in the South had fewer Clinton voters showing up to vote. The clues were there but the pollsters must have believed Hillary's shit didn't stink.

Voter suppression tactics such as described by Palast have an effect but it has been a problem at least since the 2000 elections. I am not sure it was more effective this year than in 2008 and 2012. But if you suppressed a million votes from 70 million it is a bigger effect if you suppress a million votes from 60 million (1.4% vs 1.7%) so it is a factor.

Posted by: Joe Sixpack | Nov 13, 2016 11:21:54 PM | 65

Adalbrand @ 61

>> From the link that Penelope gave:

No evidence in that article whatsoever, backing up claims.

I followed some of author's other "references"... whacko.

For example, among certs backing up claims both Clintons are pedophiles was this email:

> From: H
>
> Subject: re: Lunch Order
>
> a hot dog with no bun
>
> Thanks,
>
> H
>
>> On Friday, May 6, 2016, Goffey, Derek (..)
>>
>> Hi everybody,
>>
>> Please send my your lunch orders and I'll
>> pick everything up. Thanks!

This is evidence of pedophilia? Maybe poor lunch nutrition, but...

Followed 10+ of author's links, all the same... innuendo, downright psychotic... no evidence, pure garbage.

This reminds me of all the crap on Free Republic after W' "liberated" Iraq, creating subculture certain anyone not supporting that (or criticizing Jr.) was traitorous.

Posted by: jdmckay | Nov 13, 2016 11:52:58 PM | 66

@jdmckay #66:

Followed 10+ of author's links, all the same... innuendo, downright psychotic... no evidence, pure garbage.

Well, thank you for following those links, so I don't have to. I guess I won't add that blog to my RSS feeds after all. And I won't finish reading the article, because the rest is speculation about what will happen.

Maybe it's because I'm a cis hetero white male, but I tend to be skeptical of all claims of pedophilia, even about people I don't like.

Also, the economics he touted struck me as pretty fringe, and I'm no fan of mainstream economics.

Posted by: Adalbrand | Nov 14, 2016 12:15:05 AM | 67

This is just too weird:

Clinton’s data-driven campaign relied heavily on an algorithm named Ada.

Ada is a complex computer algorithm that the campaign was prepared to publicly unveil after the election as its invisible guiding hand. Named for a female 19th-century mathematician — Ada, Countess of Lovelace — the algorithm was said to play a role in virtually every strategic decision Clinton aides made, including where and when to deploy the candidate and her battalion of surrogates and where to air television ads — as well as when it was safe to stay dark.

The campaign's deployment of other resources — including county-level campaign offices and the staging of high-profile concerts with stars like Jay Z and Beyoncé — was largely dependent on Ada's work, as well. …

According to aides, a raft of polling numbers, public and private, were fed into the algorithm, as well as ground-level voter data meticulously collected by the campaign. Once early voting began, those numbers were factored in, too.

What Ada did, based on all that data, aides said, was run 400,000 simulations a day of what the race against Trump might look like. A report that was spit out would give campaign manager Robby Mook and others a detailed picture of which battleground states were most likely to tip the race in one direction or another — and guide decisions about where to spend time and deploy resources. …

Like much of the political establishment Ada appeared to underestimate the power of rural voters in Rust Belt states

LOL. Her Silicon Valley buddies must have cooked Ada up for her and told her it would give her unbeatable tech magic.

Posted by: Adalbrand | Nov 14, 2016 12:31:06 AM | 68

More on the LA Times tracking poll

The poll was set up differently than other major polls. Roughly 3,000 respondents were recruited into a panel that used an unusual method of "micro-weighting" to reflect the overall voter population. The poll was conducted by dipping back into this same pool of people each time. This may have created a more stable baseline from which to detect shifts in voter preference.

The poll design allowed respondents to assign themselves a probability, from zero to 100, of their voting for either candidate. This approach, rather than simply asking for a concrete voting preference, may have allowed the poll to be more precise in detecting shifts in sentiment. The USC poll’s results also were weighted based on how people said they voted in 2012—an approach that experts criticized on the basis that many people misstate or misremember how they voted in the past.

Everyone who has participated in this thread should read that article.

Note: I have changed my name from Demian to Adelbrand, because Adelbrand is my Twitter name. I am making this comment, even though I just tweeted a reference to this Bloomberg article, as a courtesy. ;-)

Posted by: Adalbrand | Nov 14, 2016 1:25:33 AM | 69

The entire surmise is based on the premise that voting is actually measured correctly, and the true results given to "we the people"... I am not naive enough to believe that in America that is the case. It is way too vulnerable to fraud to qualify as science. Fuck the polls, the entire thing is a farce. Convincing, to the naked eye, but upon closer scrutiny, is a joke.

Posted by: dan | Nov 14, 2016 3:05:41 AM | 70

Adalbrand | 42

If understand that LA poll the +3% for Trump was the popular vote which Hillary won so that can't be more accurate than the other polls.

Posted by: ToivoS | Nov 14, 2016 3:21:00 AM | 71

@69 D

Is it Adalbrand? or Adelbrand? chipnik had some unsolicited help with the 'correct' spelling :)

Posted by: jfl | Nov 14, 2016 3:49:19 AM | 72

Dan @ 70, I think you're right that the whole thing is a fraud, theatre for the masses. The whole purpose of computerized voting is to count the votes in a black box (in secret).Changing Reason's most recent youtube explains why he thinks the election isn't merely rigged, but theatre.
---
As Psychohistorian is always reminding us, the Banking system is at the heart of our stolen economic & political welfare. I've been spending time researching the Financial CHOICE Act, which has passed the House. Trump is said to approve of it and that is consistent with various remarks he's made about wanting to deregulate banking.

I'm very sorry to report that the Financial CHOICE Act is terrible. If you read only the first two pages of this link you'll see why. http://ourfinancialsecurity.org/wp-content/uploads/2016/07/AFR-Hensarling-Choice-Act-7.12.16.pdf

Posted by: Penelope | Nov 14, 2016 4:57:27 AM | 73

@ToivoS #71:

If understand that LA poll the +3% for Trump was the popular vote which Hillary won so that can't be more accurate than the other polls.

Based on the LA poll, I believe Trump won the popular vote. If the official tally stands at Hillary winning the popular vote, I attribute that to voter fraud. It is hard not to accept the Republican narrative about how the Democrats engage in massive voter fraud after Obama said that it is all right for non-citizens to vote.

@jfl #72:

Is it Adalbrand? or Adelbrand? chipnik had some unsolicited help with the 'correct' spelling :)

Thanks for that question, because I had seen chipnik's comment, but had not realized that it referred to me.

The correct spelling is "Adalbrand", which I found on a list of German names. The reason why I misspelled it as "Adelbrand" is that that spelling comes more naturally to me, since "adel" is a word in German but "adal" is not.

The reason I changed from "Demian" to "Adalbrand" by the way is that "Adalbrand" is very Germanic, and the Alt Right likes that sort of thing. So Adalbrand seems more appropriate to the spirit of the times after Trump's victory. "Demian" was German but from Herman Hesse, and signified the predeliction by Europeans for the East over the West.

Posted by: Adalbrand | Nov 14, 2016 5:52:14 AM | 74

The maths behind polling is apparently well established. If it had been used to predict market share of Samsung versus Toshiba washing machines the polling would have been accurate within the limits of error. But since a difference of one or two percentage point makes all the difference between winning and losing one cannot use it to predict the outcome with certainty.

I had myself expected Trump to win the popular vote on the expectation that there was a large percentage of white voters, hitherto uninterested in voting would turn up in the polls. In the event there was not such a group that were unaccounted for in the polls. Trump very nearly did not become president. It was finally Clinton's arrogance in dismissing the white blue collar workers, that gave the 'Rust Belt' states over to Trump.

Posted by: Ivan | Nov 14, 2016 6:09:32 AM | 75

Polls have become a tool of propagandists, just like lies, half-truths, omissions and misdirections.

Posted by: Mick McNulty | Nov 14, 2016 6:29:04 AM | 76

@74 AD

Thanks for the explanation ... I had guessed the origin of Demian, and looked up Adalbrand in wikipedia. The German version wasn't so helpful ... what German I had is pretty rusty ... but it looks as though he lived in Hamburg - b is from Hamburg, maybe? - about 1000 years ago and was a priest ... I couldn't figure the connection to a lapsed orthodox/lutheran like yourself ... but if it's just for effect it's not immediately German looking, not to me anyway. I'd think you'd choose a Russian name? Or Horst or Fritz or something easy to type and remember, if you're stuck on a German persona.

But Russia is undoubtedly exotic eastern Europe. I admire the Germans but ... I once read the story of the Mayan hieroglyphics and their decipherment in modern times ... the University of Pennsylvania, I think it was, a high-powered US university, spent years and millions sending teams to the Yucatan working the project ... and then a Russian mathematician solved the problem in Russia, maybe in Leningrad, working in pencil on the back of old envelopes, in an afternoon, never having seen the Yucatan. If I wanted to adopt a persona, I'd go back and find out that guy's name and adopt it as my nom de plume. But I'm too lazy to do that, and ... once you start pretending, where do you stop? That'd be my problem, anyway. I have an active imagination. I'm better off sticking with the facts, like Sergeant Friday ...

Posted by: jfl | Nov 14, 2016 6:50:09 AM | 77

Someone's probably already mentioned this but there was one poll done by the LA Times that used a different technique and got it right. Instead of polling "random" new people every time it tracked the same 3000 people over time. That enabled it to accurately track shifting voting tendencies.

I don't have a link but the LA Times poligtical editor just was on Tavis Smiley describing the advantages.

Posted by: ritzl | Nov 14, 2016 7:39:48 AM | 78

trying to ascertain useful information from a manipulated inexact science seems a rather futile endeavor to me. a kinda Rorschach test for people with tunnel vision.

Posted by: john | Nov 14, 2016 8:01:24 AM | 79

Sure hes is already rich but if he keeps this promise, its pretty big.

Trump: I wont take any salary as a president.

https://twitter.com/MashableNews/status/797981624830824448

Posted by: Ripo | Nov 14, 2016 8:08:40 AM | 80

Balderdash! Only fools and children pay attention to polls.

Posted by: V. Arnold | Nov 14, 2016 8:45:44 AM | 81

@ jawbone 8.
Cheap and lazy (and biased), that's our media. I once read an article by two black sociologists from NYC who went to the Appalachians and said the face of the poor isn't black, it's white because it's so easy for the media and PC science to step out into any city and make claims without consideration for the whole country.
Unfortunately cheap and lazy is profitable. Some in our intelligence services have complained about how SIGINT trumps HUMINT because HUMINT is hard, costly, and requires effort to sift for the truth (along with dealing with some bad actors in order to get to worse ones).

Posted by: Curtis | Nov 14, 2016 9:23:32 AM | 82

1998 interview with Soros
https://www.youtube.com/watch?v=Jg2P4kcIrZ0

No conscience for bad things he does. But he "tries to be moral" and does a handful of philanthropic things. 60 Minutes did for Soros what they've done for others like the Clintons.

Posted by: Curtis | Nov 14, 2016 9:33:54 AM | 83

Nassim Taleb's criticism of 538's political predictions is relevant. 538's predictions were based on a variety of polls and statistical methods. Taleb used options pricing to show that 538's predictive capability was bogus. No rational investor would/should put money on 538's numbers.

The day before the election, 538 predicted an 85% chance that Hillary would win.

Posted by: Jackrabbit | Nov 14, 2016 9:53:41 AM | 84

artificial intelligence predicted a Trump win!

a million laughs!

Posted by: john | Nov 14, 2016 10:25:03 AM | 85

Posted several times re. other indicators and the polls… - sigh. But it’s the past… still, repeat:

Nobody shows up for Clinton rallies? Her likes on Facebook / subscribers on reddit / more / were about between 1/8 and to push it.. 1/5 of that for DT. Who switched to the Dem party this year? What did the early voter stats in Florida show? And so on.

Ppl believed that HRC would be a shoo-in. Why? Because of the relentless grip of the PTB - Media.

If ppl didn’t go to her rallies, or sign > Dem, or join her campaign, or contribute to it (FIRE and Corps set aside), or buy her book, or other, why would they go out and vote for her? They didn’t, as predicted.

Everything b wrote in the top post is OK, fine. But there is more to it, in the sense that the polls were supposed to signal that HRC was going to win in any case, the election was a ‘done deal’ so there was not much point in voting against. (Huh, dumbo ppls your vote doesn’t count..) Yet, even that can be attributed to unconscious bias and faulty conventional methods that shore up self-fulfilling prophesies. To play a little catch-up, some ‘admitted’ the ‘race is close’ so they could be ‘surprised.’

Insiders knew all of this at least 6 months ago (not the Dems caught in a bubble.) Internal ‘secret’ polls showed a huge advantage for DT.

Btw, French pollsters don’t even bother with ‘weighing’ or statiscal high jinks, they have no interest in justificatory BS, they just add between 3 and 5%, bang off, depending on type of election, to the National Front. They call it a ‘correction mechanism’, of course that is not made public.

Posted by: Noirette | Nov 14, 2016 10:56:41 AM | 86

What does the result say about media's power over people? How much power does media actually have today? I cant understand it really after their pro-clinton bias, still Trump won, how?

Posted by: Ripo | Nov 14, 2016 11:09:33 AM | 87

This election exposed the bare ass of the Mainstream Media with its shorts pulled down around its ankles.

Revelation: they're all a bunch of lying handmaidens for the PTB (and for Zion).

Posted by: fastfreddy | Nov 14, 2016 11:30:57 AM | 88

@Ripo, fastfreddy:

I think that the lies of the media have become so regular that even the average Joe caught on.

I think Americans are pretty traumatized by the Iraq war, given that it was made into such a big deal at the time, but we effectively lost. And I think that most Americans realize that the Bush administration and the media conspired in lying to fool the American people into backing the war.

Then there was Libya and Syria. Countries most Americans have never heard of, being constantly in the news, with the US wanting to bomb them. A pattern emerges.

Trump did not come out of nowhere. Wikipedia lists him as being on The Apprentice from 200.–2015, That is a well-known show. I've never watched it, but I know that in it Trump says "You're fired", because that's just part of American pop culture. So Trump had this persona with vast swathes of the American public as being a rich guy with a sense of humor. But then, when he runs for president, he starts getting smeared as a racist, misogynist, etc. These accusations about him really came out of nowhere. (The accusations were clearly a central part of Clinton's electoral strategy. This is a change election, and in a change election, voters vote for the change candidate, who is Trump in this election. So the Establishment had to convince voters that Trump is unacceptable, so that you can't vote for him even if you want change.)

When things started going bad for Hillary, she started blaming everything on Russia. And the media started attacking Russia, too. Trump made a point of saying that the US has no problems with Russia, and that Obama and Clinton in effect back our enemies in the Middle East. I think all this talk of war, combined with Trump being consistently against war, made people connect the media lying about countries because the US wants to bomb them with the media lying about Trump.

Posted by: Adalbrand | Nov 14, 2016 12:00:30 PM | 89


Excuse my second post.
Strange
With early reported election results, Clinton was winning lots of states, and the pop. vote got to Trump 50% with Clinton only at at 45%.
Then as things progressed, with Trump winning more and ever more states, the more states he won, the more the popular vote was reported decreased, to what it now is. Trump trailing.
So, please, someone, start looking for the bears hiding in the bushes.

Posted by: joey | Nov 14, 2016 12:12:02 PM | 90

Adalbrand @ 89

> But then, when he runs for president, he starts getting
> smeared as a racist, misogynist, etc. These accusations about
> him really came out of nowhere.

They didn't come out of nowhere. 2 main reasons in particular off top of my head:
a- Kansas Sec. of State Kolback has been involved in voter (minority) suppression efforts across the country, most of which were killed by the courts. Kobach is reportedly the "idea man" behind Trump's wall

Kobach is also behind this election's latest GOP caging scheme called: CrossCheck, which kind of makes it look like maybe the pollsters weren't wrong after all. From Palast:

The system, called Crosscheck, is detailed in my Rolling Stone report, “The GOP’s Stealth War on Voters,” 8/24/2016.

Crosscheck in action:
Trump victory margin in Michigan: 13,107
Michigan Crosscheck purge list: 449,922

Trump victory margin in Arizona: 85,257
Arizona Crosscheck purge list: 270,824

Trump victory margin in North Carolina: 177,008
North Carolina Crosscheck purge list: 589,393

On Tuesday, we saw Crosscheck elect a Republican Senate and as President, Donald Trump. The electoral putsch was aided by nine other methods of attacking the right to vote of Black, Latino and Asian-American voters, methods detailed in my book and film, including “Caging,” “purging,” blocking legitimate registrations, and wrongly shunting millions to “provisional” ballots that will never be counted.
>Trump signaled the use of “Crosscheck” when he claimed the election is “rigged” because “people are voting many, many times.” His operative Kobach, who also advised Trump on building a wall on the southern border, devised a list of 7.2 million “potential” double voters—1.1 million of which were removed from the voter rolls by Tuesday. The list is loaded overwhelmingly with voters of color and the poor.

- Bannon. Take your pick of references. psychohistorian @ 211 on "Why Polls Fail" brought up some of these with Mother Jone's link... and to avoid putting words in his mouth ph makes a distinction between "nationalist" and racist/white supremacist.

We know about Bannon's new job, but Kobach's been key member of his transition team as is Sen Sessions... a guy quite unashamed of advocacy for years many would consider racist. Not to mentions Sessions' (IMO) very troubling "Nationalist" (eg. treasonous) rhetoric vociferously backing W's run-up to invading Iraq. I've long thought he was just dumb as rocks. Whatever the case, not in line with "draining the swamp".

I guess people can draw their own conclusions.

Posted by: jdmckay | Nov 14, 2016 12:44:54 PM | 91

Adalbrand & Jdmckay, thank you for alerting me to the fact that I posted the wrong Robert Steele article.

HOW TO EMPOWER THIRD PARTIES - Much of interest here, but don't click on the "Accidental President" link. It's the others which are valuable.


https://memoryholeblog.com/ "Robt Steele's 3-Track. . ."

jfl of spec'l interest to you, I think. Your comment?

Posted by: Penelope | Nov 14, 2016 12:56:15 PM | 92

91;You have fondness for serial liars.Why?
The zionist MSM tried to manufacture dissent vs Trump by terrible undemocratic propaganda,and they did affect many,witness the idiots carrying peace signs for the Hell Bitch.
Trump and America first is the last thing they want,as it implies no more Israeli domination of American policy.I just saw today he wants a deal for I-P,Assange is interviewed,he won't tear up Iran deal,and the headchoppers are now a target.
What's not to like?
And Palast is a no good lying zionist provocateur who was mum throughout Obombas abysmal reign.Another dentist gone to waste.
And the only moron responsible for her defeat is the Hell bitch herself,who dissed half the country,and whose policies,all hidden by the serial liars,were nothing but more of the same garbage we already endure in our misery.
Trump 2020,and stop critiquing Trumps choices for his cabinet or staff,he is a loyalty guy,and un-politically correct,thank God.
Give him a chance,and don't fall for the MSMs divide and conquer.

Posted by: dahoit | Nov 14, 2016 1:20:50 PM | 93

91

While I totally respect Palast and it's a shame his research is being buried by the media, and have no doubt RNC was busy re-districting even before Clinton 1, together with blatant voter roll purging as seen in Florida in 2000 and Ohio in 2008, and 2016 became a race between DNC flying Puerto Ricans into FL and NC as fast as RNC was purging black voters, nevertheless, now we have the known-known under Trump-Wall Street-MIC, as opposed to the Great Unknown under Rodham's Pay-for-Play Clinton Foundation:

https://www.youtube.com/watch?v=2ayHf7qbZGA
https://www.youtube.com/watch?v=xubN6ctNNv8
https://www.youtube.com/watch?v=_NfFAaPZqs8

With the known-known, you can see the slaughterhouse corral in the distance, and plan your own strategic exit. With the unknown-unknown, we were surrounded by a pack of wolves on every quarter. A known-known is more Orderly and Beneficent.

Posted by: chipnik | Nov 14, 2016 1:46:55 PM | 94

When will people get real? The polls don't work anymore because the actual votes are meaningless. EVERY U.S. ELECTION IS RIGGED. IT HAS BEEN ADMITTED BY BRAGGING ELITISTS. When are wishful thinking but thick headed people going to believe them?
Note on the last election that Hillary was winning everything, and predicted to have a landslide, until the real fixers, the cabal in the City of London, told her the dirt coming out would expose their game if they put her in. She then told her people not to bother with the victory celebration but to go home and relax. Trump was given the word too and the phony vote count, at that moment, switched completely and began giving Trump the advantage.
It's all a fake. Trump and the Clintons are friends. It doesn't matter that much to the Rothschild cabal which prostitute they put in office, they will always get their way unless rooted out and destroyed.

Posted by: Tony B. | Nov 14, 2016 1:49:36 PM | 95

93

Members of Trump's family are either converted Jews, married to Jews or in 1:1 relationships with Jews. His grand-children are Jews, and his inner circle on the campaign were Jews, included his propaganda minister. He is rolling back banking restrictions on Wall Street for the Globalist Bankhim, he has always pushed the Zionist Exceptionalist meme, he is first entertaining Netanyahu and Al-Sisi at the White House, both Jews, and he proclaims that he, 'Stands With Israel'.

"Trump's 'America First' implies no more Israeli domination of American policy."

That all depends on Who's on First.

Posted by: chipnik | Nov 14, 2016 2:12:46 PM | 96

@Ritzl - 78

Back when I was managing campaigns for a living, I always designed my field operation so that it also functioned as a gigantic tracking poll.

As a result I managed to call the resulting vote count within one-half of one percent in almost all the campaigns I managed.

Pollsters who are interested in actual data will use tracking polls.

Posted by: AntiSpin | Nov 14, 2016 3:31:19 PM | 98

@97

How I predicted Hillary? I put 'H' one side of a coin and 'T' on the other, then flipped

Posted by: bbbb | Nov 14, 2016 3:41:56 PM | 99

Trump and Putin talked.
Relations to be normalized.

Iran, Hezbollah and the Palestinians should be very afraid right now in absence of quality information about Trump's policy ideas

Posted by: bbbb | Nov 14, 2016 3:46:11 PM | 100

next page »

The comments to this entry are closed.