patrick.net

Seriously?

They still have the audacity to even mention the word "poll" after the election of 2016?

Simply unbelievable. The polling industry's funeral was held late on Nov 8th. Gone, dead, forgotten.

I know you conspiracy nuts won't believe this, but the polling error was about average for Presidential elections since 1968. The problem was that most people don't understand statistics, uncertainty, and dependent variables so they made these ridiculous models showing Clinton with some ridiculous probability of winning. 538 is the gold standard and they had it about 70% for Clinton and 30% for Trump, IIRC.

Gallup showing similar things too. Good thing they have no history of polling at all, so no problem.

http://www.gallup.com/poll/201977/trump-pre-inauguration-favorables-remain-historically-low.aspx?g_source=Politics&g_medium=newsfeed&g_campaign=tiles
http://www.gallup.com/poll/201833/approval-trump-transition-low-inauguration-nears.aspx?g_source=Politics&g_medium=newsfeed&g_campaign=tiles
http://www.gallup.com/poll/193763/trump-image-last-year-stable-negative.aspx?g_source=Trump&g_medium=search&g_campaign=tiles
http://www.gallup.com/poll/201158/skeptical-trump-handle-presidential-duties.aspx?g_source=Trump&g_medium=search&g_campaign=tiles

Trump knows it is meaningless, so he won't comment on popularity, to bring it to people's attention ... oh ... wait ...

The MSM are > 90% democrat, and this last election we saw that they are just part of the DNC effectively offer 1000s of hours of free advertizing and polling for democrats.

They dont work outside of a 30 mile radius of major cities.

Sadly, the MSM where they overlooked major national security gaffs of democrats (Russia not a threat Romney, 80s wants their foreign policy back), but will with imaginary microscope beat Trump on a daily basis, even if it results in even more division than has been created over the last 8 years.

Trumps only hope is to reach out to people who may be moderate, to reach out to those in the inner city whos friend have been getting harrassed by police, shot at, and jailed. If in a few years those stuck in bad situations actually see their inner cities, drugs, and crime situation getting better, then and only then will those who believe in liberals primary drop their blinders.

Part of the immense pushback we see against Trump trying to keep jobs here, questioning the validity of all these security leaks, and appointing non political insiders, is that this is what has to be done regardless of one political beliefs. Got bless the democrats and independants who give Trump a chance, even if they dont have the same political beliefs.

Obama divided this country, and was slicker than Willie - for as these troubles brewed during his presidency, he cast blame and deflected the issues onto others. We not longer need to wonder why Chicagos south side is a bastion of crime - no one has done anything to even attempt to rectify this situation.

"You mean like WaPo, ABC, NYT and others OVERSAMPLING Dem voters by 14% to keep Hillary in the lead??"

Nope--they certainly didn't oversample by 14%. The national polls were off by 1%ish. If they oversampled by 14%, the error would have been much, much larger.

"Sure he was, remember HIS prediction of the EC count???"

Yep--and what's your point? With all or nothing EC, and several states VERY close, it causes a big swing in the EC even though it was a very small error.

If Bill Clinton would have come out and said that NAFTA had it drawbacks, and was not a panacea, Hillary would have won the election.

If Obama have even a mediocum of effort expended on bringing back low/high technology jobs, and would have fought to avoid one major firm outsourcing to Mexico/Asia with words like "keep american jobs here" and push firms to do this, Hillary would have one.

Hillary might have chosen a different running mate - AKA Bernard Sanders - she might have won.

One might question whether she wanted the job...........

Great--so they oversampled by 14% but somehow were only off by about 1% nationally? And performed at about the same accuracy as they average since 1968?

How did that happen??

joeyjojojunior says

most people don't understand statistics, uncertainty,

Ironman's understanding of these things is lower than most people. Sadly, this explains how terrible his opinions are.

Assuming that Trump is right, and the polls are rigged just like before, then he outperforms the polls by 1 to 2%. Oh, that's right. His poll numbers are still tragically terrible.

Ironman says

Sadly, I was the ONLY one right here about the election predictions.

You were wrong about a lot of things, and you were too chicken shit to put a number on the national popular vote. Here's what I said prior to the election https://patrick.net/?p=1298427&c=1343785#comment-1343785

Silver is predicting she'll win by 3.5%. I agree with him that it will probably be 4%. I'd be surprised if it wasn't between 1% and 7%. You posted facts without any interpretation. You want to be able to imply something without offering an opinion, so that you can hedge your bets.
I think that the polls are pretty accurate and the results usually fall within +-2%. I don't believe that there is any giant conspiracy of Jewish overlords controlling the banks and media and skewing these polls for Clinton. If you do (like Trump and Giuliani) then go ahead and tell us what you think the popular vote will be.

The result did fall within 2%, which is pretty typical. Clearly, I was not surprised by the result given my prediction of the popular vote.

In contrast, you were too much of a pussy to make any concrete predictions on the popular vote. Your many threads suggested that the results would be hugely different from the polls. In fact, the results were pretty similar to the polls. So in that respect you were very wrong. You are just too stupid or willingly dishonest to see reality for what it is.

Entitlemented says

The MSM are > 90% democrat ...

... and because I put a faux statistic here, that makes it totally credible.

YesYNot says

joeyjojojunior says

most people don't understand statistics, uncertainty,

Ironman's understanding of these things is lower than most people. Sadly, this explains how terrible his opinions are.

Assuming that Trump is right, and the polls are rigged just like before, then he outperforms the polls by 1 to 2%. Oh, that's right. His poll numbers are still tragically terrible.

Ironman understanding of most things is less than most people.

He's a slobbering idiot, evidinced by his inability to ever delve into specifics of any issues.

His realm is

Personal Attacks
Silly pictures
Memes

Sounds like a pre-teen

Uh oh, Ironman must have pissed and shit in his adult diapers. He's all miserable lately.

Why so #triggered, bro?

errc says

His realm is

Personal Attacks
Silly pictures
Memes

AKA : not worth listening to ... at ... all. Zero value add.

BOYCOTT ALL REPUBLICANS!

Ironman says

Here's MY concrete prediction:

https://patrick.net/?p=1298469&c=1343469#comment-1343469

Ever think of getting a job as a pollster???

joeyjojojunior says

I know you conspiracy nuts won't believe this, but the polling error was about average for Presidential elections since 1968.

Everyone, including the pollsters were shocked by the results. If the polls were within the margins of error, you would not have had this extreme level of shock.

Strategist says

Everyone, including the pollsters were shocked by the results. If the polls were within the margins of error, you would not have had this extreme level of shock.

They weren't shocked at 538. Nate was tweeting before the election that the margin of error was greater than normal because there were undecided voters right down to the wire than was typical. He gave Trump roughly a one in three chance because of it. He knew that if Clinton won nationally by 2%, she'd probably lose the electoral college.

The margin of error depends on the poll and how many participants--what is a fact is that the polling error was pretty typical for Presidential elections over the last 40 years.

joeyjojojunior says

Strategist says

Everyone, including the pollsters were shocked by the results. If the polls were within the margins of error, you would not have had this extreme level of shock.

They weren't shocked at 538. Nate was tweeting before the election that the margin of error was greater than normal because there were undecided voters right down to the wire than was typical. He gave Trump roughly a one in three chance because of it. He knew that if Clinton won nationally by 2%, she'd probably lose the electoral college.

The margin of error depends on the poll and how many participants--what is a fact is that the polling error was pretty typical for Presidential elections over the last 40 years.

One person, Nate is not everyone. The stock market had discounted a clear Clinton victory. That can be construed as everyone. When the news of Trump's victory started coming out, the futures initially collapsed due to the uncertainty, indicating no one had factored in Trump winning. It was only after they analyzed the affects of the Trump victory, did the stocks journey on a sharp rise.

"Oh really???"

Yes, of course.

Here's an article written BEFORE the election:

https://fivethirtyeight.com/features/trump-is-just-a-normal-polling-error-behind-clinton/

From the same article--again, written BEFORE the election

More for you:

https://fivethirtyeight.com/features/can-you-trust-polling-in-the-age-of-trump/

"why the polls were so WRONG"

That's the point--they weren't that wrong. Did you read any of the material I posted?

"when REALITY tells every one else your BLUE team put up a totally shitty candidate."

No shit. Have I ever said Clinton wasn't a bad candidate? In fact, I've said she was a horrible candidate on many occasions here.

Rew says

Gallup showing similar things too. Good thing they have no history of polling at all, so no problem.

http://www.gallup.com/poll/201977/trump-pre-inauguration-favorables-remain-historically-low.aspx?g_source=Politics&g_medium=newsfeed&g_campaign=tiles

http://www.gallup.com/poll/201833/approval-trump-transition-low-inauguration-nears.aspx?g_source=Politics&g_medium=newsfeed&g_campaign=tiles

http://www.gallup.com/poll/193763/tr...

1. Do you know if they followed the same methodology that they always do?

2. Do you understand the limitations of telephone and internet surveys

3. They polled spanish speaking residents of 50 states and DC; they did not mention how many though
--------------

Gallup polls are prone to screw-ups no matter their experience, just look at these articles from huffington post

http://www.huffingtonpost.com/2013/03/08/gallup-presidential-poll_n_2806361.html (Obama election)

http://www.huffingtonpost.com/2013/05/18/gallup-review-polls_n_3299895.html (Obama election)

http://www.huffingtonpost.com/2013/06/03/gallup-2012_n_3379264.html(Obama job approval rating)

----------

Results are based on telephone interviews conducted January 4-8, 2017 with a random sample of –1,032—
adults, aged 18+, living in all 50 U.S. states and the District of Columbia. For results based on this sample of
national adults, the margin of sampling error is ±4 percentage points at the 95% confidence level.
Interviews are conducted with respondents on landline telephones and cellular phones, with interviews
conducted in Spanish for respondents who are primarily Spanish-speaking. Each sample of national adults
includes a minimum quota of 70% cell phone respondents and 30% landline respondents, with additional
minimum quotas by region. Landline and cell phone telephone numbers are selected using random digit dial
methods. Gallup obtained sample for this study from Survey Sampling International. Landline respondents are
chosen at random within each household on the basis of which member has the next birthday.
Samples are weighted to correct for unequal selection probability, non-response, and double coverage of
landline and cell users in the two sampling frames. They are also weighted to match the national demographics
of gender, age, race, Hispanic ethnicity, education, region, population density, and phone status (cell phoneonly/landline
only/both and cell phone mostly). Demographic weighting targets are based on the March 2016
Current Population Survey figures for the aged 18 and older U.S. population. Phone status targets are based on
the January-June 2016 National Health Interview Survey. Population density targets are based on the 2010
census. All reported margins of sampling error include the computed design effects for weighting.
In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce
error or bias into the findings of public opinion polls. For questions about how this survey was conducted,
please contact galluphelp@gallup.com

"Nate Silver was WRONG every step of the way with THIS election. Anyone who uses him as a point of reference is a total idiot. Did you read the articles I posted referencing AFTER the election from your favorite Libbie news sources?"

Nope--he and his 538 colleagues were dead on actually. As the only ones who could understand statistics and develop a proper model for the probability of a Trump win based on polling numbers. As the article I posted clearly shows--he was banging the drum claiming that most of the media was vastly underestimating the chances of a Trump win. That all Trump needed was a normal polling error in his favor and he would be the next President.

Guess what--we got a normal polling error in Trump's favor.

You, on the other hand, are clearly and irrefutably incorrect. Claiming that pollsters were oversampling Dems by 14% is obviously BS. The elections results were within the usual polling error. If they had oversampled by 14%, the results would have been WAY outside normal polling error.

Yes, and the election results showed you were obviously and completely wrong. The polling company, not surprisingly, knew much better than you what their party ID mix should be in order to get an accurate poll.

Polling companies know that what people tell you for their party ID isn't really reality. Lots of Republican voters like to claim they're Independents even though they haven't voted for anyone without an R next to their name in 40 years. Pollsters know this and account for it.