Were polls way off in 2016?

[struggle4progress]

By Dan Balz
May 4 at 1:00 PM

... A new report examining what happened comes to a split conclusion: National surveys were generally accurate in projecting the popular vote but state polls had “a historically bad year” in forecasting the results in the electoral college.

... Trump won .. by rolling up 306 electoral votes to .. Clinton’s 232. But Clinton won the popular vote by almost 3 million votes and a margin of 2.1 percentage points ... the second time in .. five elections .. a Democrat won the popular vote and a Republican won the electoral college ...

... The AAPOR committee concludes that the national polls “were generally correct and accurate by historical standards,” and that they were more accurate than in 2012. The polls, on average, pointed to a Clinton victory in the popular vote by about three percentage points. Her eventual advantage was well within the margin of error of the national polls.

But where elections are decided, in state-by-state contests, things were not so rosy for pollsters. State polls were historically bad — the report calls it the largest error in state polling of elections starting in 2000 — and the key failure was the underestimation of Trump’s support. This was particularly true in the Upper Midwest, where the election was decided ...

https://www.washingtonpost.com/politics/were-the-polls-way-off-in-2016-a-new-report-offers-a-mixed-answer/2017/05/04/a80440a0-30d6-11e7-9534-00e4656c22aa_story.html

[JI7]

1. Fuck Comey

[joshdawg]

13. Got my vote!

[LenaBaby61]

2. Were polls way off in 2016?

Wasn't too difficult for the polls to be off, especially when you had a set of insane circumstances IE: ruskie meddling, sexism, misogyny, racism, voter suppression, voter-crosschecking, voter purging etc. done to Dems by thuglicans in certain states, and cherry on top: Interference by FBI head James "'Mildly Nauseous" Comey, who chimed in 11 days before November 8th with bogus news about Hillary Clinton's emails, meanwhile, sitting on scandalous and horrible information that he and other intelligence agencies were (And still ARE) investigating tRumputn/ruskie collaboration/ties.

Yep, not difficult for pollsters to have gotten some things about that tainted 2016 GE wrong.

[democratisphere]

3. HRC won the popular vote by a mere 3 million votes; in every other state and in every other country

THIS would have placed her in the oval office of the WH. Our ass backward electoral college skewing of THIS election allowed an insane idiot to take HRC's rightful position instead. The polls had it right, the f*cked up election system got it wrong.

[dchill]

4. No, they weren't. Especially the exit polls.

[llmart]

5. That would be my answer also.

It's just so much easier for them to use this an an excuse not to further investigate the real reasons we have an imbecile in the White House. That might be too much trouble for them.

[uponit7771]

19. +1

[joshdawg]

6. Isn't it about time to slow down this business about why or how Democrats lost the election?

We lost and we already know why and how. Why belabor the point?
2018 is around the corner and that is what our focus should be on.
We need to regain the House and the Senate.
Impeachment for the dog crap in the W.H. is viable, but will it happen? With a republican controlled Congress?
Personally, I would not vote for another republican ever! None has my best interests in mind.
Maybe I'm one of those yellow dog Democrats. Maybe.
I do know I am a Liberal AND Progressive.
Ok, I'm off my soap box.

[planetc]

7. Did we lose? I think that's the question, not the answer to "What do we do next?"

Or, DID we lose? Did we actually lose, or did more people vote for Clinton & Kaine in swing electoral college states than Trump and Pence? If they did, and the election was stolen after voters cast a vote, then the problem we face is different from "Why did all those nuts vote for Trump?" If Democrats actually won, then the question is "How do we assure the safety of our elections?"

And if the answer to the first question is NO, we didn't lose, then it will do us little good to tinker with our message or find more and more exciting candidates, because with the best candidates and the most powerful message, we will still "lose" in 2018.

If we actually won, the problem is to secure the accuracy of our elections, which is a big enough problem. If we actually lost the election, then we have to find a way to communicate with all those sincere Trump voters. Sincere, but woefully misguided.

So, I don't see how we can decide on our strategy until we know what the problem actually is.

[joshdawg]

12. Well, according to the electoral college, we lost. Any way you look at it,

a lunatic republican is in the White House.
I get what you are saying, but again, why belabor the point that republicans won and Democrats lost.........again.
The point of my post was to ensure we have viable candidates with the right message. That is our focus. Regaining the House and the Senate. It can happen.
Have a good day.

[planetc]

15. And the electoral college is controlled by.....?

[mythology]

16. We lost

There is no evidence of votes being switched.

[planetc]

17. Unless, of course, we didn't lose.

I don't think you're interested in the highly suggestive evidence that votes were switched, erased, or simply hacked. Evidence that started coming in in 2000. Your mind is made up.

[sharedvalues]

8. Microtargeting on social media said to lie to pollsters?

Some speculation that the social media manipulation 45's campaign did (and Russia did) told people to lie to pollsters. So the right lied to pollsters. And the left got a little complacent, which may have been intended.

[colsohlibgal]

9. Not Just 2016

2000 and 2004 as well. Our election procedures are a joke, scattershot and ripe for hanky panky. Then there is the Electoral College....which gives way too much weight to voters in sparsely populated republican stronghold states when you calculate votes per population.

[DemocratSinceBirth]

10. The models were off more than the polls.

Miracles happen. Black swan events happen. Trump hasn't repealed the laws of probability.

[Bernardo de La Paz]

11. Need more accountability and security in vote tabulating. Damn voting machines.

Damn the instant gratification need for counting that leads to vote machines in the first place.

If the machine is compromised -- the Republicon companies selling them have no accountability or transparency -- then it can just jigger the vote results for that machine by 10% in favor of Republicons.

If the computers used to read from the machines are hacked then they can jigger the results after reading them.

Needs to be a lot more auditing and transparency.

Remember: about a 40,000 vote swing (about 80,000 vote margin) would have won Clinton the Electoral College in addition to her winning the Popular Vote.

[yallerdawg]

14. Comey - yeah, it mattered.

Exit polls found that late-deciding voters backed Trump by substantial margins in Florida, Michigan, Pennsylvania and Wisconsin in what broke the back of the Clinton campaign’s path to victory. The report notes: “This can be seen as good news for the polling industry. It suggests that many polls were probably fairly accurate at the time they were conducted.”

[karynnj]

18. Not necessarily Comey

While there may be some who decided based on Comey, they could also have been conservatives, republicans etc who reluctantly cast a vote based on the Supreme Court/ right wing policies, who still did not like or respect Trump. The exit polls showed that those who did not like EITHER candidate, mostly voted for Trump. I suspect this is because by the time of the election, very very few Democrats had not aligned with Clinton - leaving the remainder mostly right leaning.

I do not think any of the exit polls bothered to ask an open ended question to the late deciders as to WHY they made that choice.

What I also don't know is whether any of the pollsters have gone back and analyzed their data on regular opinion polls to compare JUST FOR THE GROUP WHO SAID UNFAVORABLE TO BOTH, how they answered the head to head question. I never saw this breakout - even when I looked at all the crosstabs posted -- and I really really wanted to see it, so I did look. Throughout the entire general election, both the favorable/unfavorable and the head to head race questions were asked for many samples. As BOTH candidates had unfavorables above 50%, there were people who disliked both.

I assume that the numbers of people disliking both are small in all the studies, but I would love it if a statistician compiled a list that showed the number fitting this category and how they answered the head to head question on each study with both AND the study's overall for each candidate- sorted by time. The raw data has this -- and it could tell much about what happened. One possibility is that all or most of the Democrats "came home" long before the election. This would not be a surprise as, among Democrats and the left, even those who would not have picked Clinton over all possible choices circa 2015 would have easily picked her over Trump. This was NOT true on the other side. there were prominent republicans who said they could not vote for Trump.

[yallerdawg]

20. Pennsylvania Monmouth Poll - Wednesday, November 02, 2016

At time of this poll, Hillary had a 4% lead.

9.      Have you read or heard about Friday's news that the FBI is now looking into newly discovered emails from Hillary Clinton's time as Secretary of State, or have you not heard about this recent news? 
94%     Yes, heard
  6%     Not heard
 
10.    Has this recent news changed your mind about who you will vote for in the presidential election, or has it not really changed your vote choice?
  4%     Changed mind about vote
89%     Not really changed vote choice
  1%     (VOL) Don't know
  6%       Not heard

 
4% changed their mind. Another 7% hadn't heard or were thinking about it.

Trump 48.8% vs Clinton 47.6%. 1.2% difference.

https://www.monmouth.edu/polling-institute/reports/MonmouthPoll_PA_110216/

[karynnj]

21. Very interesting -- Thanks

[StevieM]

22. 4 percent changing their minds is an 8 point swing. Thanks for the link. (eom)

[StevieM]

23. The polls weren't off--the polls changed. Comey rigged the election.

And not for the first time, either.