How can all the polls be wrong in just one direction

[edhopper]

by double and triple the Margin of Error?

Ohio was a virtual tie in the polls and Trump is up by 8%

PA, where Biden was up by 6% has Trump is now ahead

WI, Biden up by 10% and it's too close to call

Same in MI where Biden was ahead by 8% in the polls.

There were no States where the polls were wrong under counting Biden

There's something happening here...

[marylandblue]

1. People are definitely lying to pollsters. They've decided to mess with us.

Last edited Wed Nov 4, 2020, 11:34 AM - Edit history (1)

Why should they tell pollsters the truth? What's in it for them?

[edhopper]

2. Everywhere?

In the whole country?

In every State?

[d_b]

11. Yep. Everywhere

For months on end. Polling outfits kept calling these republican jokers — lots of em. Operation Trolljob.

I’m kidding, of course.

[marylandblue]

12. No, but in the key states, and just a certain group.

We've been calling them deplorables for years. If 5 or 10 per cent of them really are deplorable, of course they might lie. Stick it to the liberals and the scientists, right?

[uponit7771]

15. Voter suppression works big time and is minimized to the nth degree by pollsters and dem party

[CousinIT]

5. I suggested people lying to pollsters - and the idea was laughed off here on DU.

I mean there's no proof of it or any link to such a story but CLEARLY polling in the US has gone to total shit in the last 3-4 elections.

I'm guessing that people (Trump cult) are lying to pollsters. Maybe they should just look at registered voters and historical data. Leave the liars out of it.

[Statistical]

14. Registered voter vs Likely voter only moves the needle 2% of so.

Registered voter is likely better than trying to predict who will vote but even still the models were way off. Like easily 5 pts conservatively across the country.

[Mike 03]

17. How come the lying never went in the opposite direction?

I live in an area where people voting for Biden would be ashamed or afraid to admit it because it is so red.

What about those adjusted polls that showed even if the polls were off as much as they were in 2016 Biden would still have a huge win?

I'm just asking. Because I'm having trouble believing the "everyone lied to pollsters in exactly the same percentages in every battleground state and none of the Biden voters did" theory.



I do have a theory about the "Shy Trump Voter" but it has to do with the Christian right.

[dalton99a]

21. +1. I also believe people pulled numbers out of their asses at some point

and there was a herd mentality among pollsters

Hence my motto: "Fuck the polls. Fuck the pundits. GOTV."

[Steelrolled]

33. I think you are right.

I know people who said they lie to pollsters (although many robocall polls are thinly disguised campaign ads).

I think some people view pollsters as part of the establishment, and would lie to "stick it to them." It is not unlike voting for Trump.

[marybourg]

30. I'm starting to agree with you.

[Amaryllis]

3. Why are the polls wrong is the wrong question. The right question is why do the results veer so far

off the polls.

[edhopper]

4. This is part

of the same question.

[Mike Niendorff]

6. Bingo.

It's almost as if some voting machines are being manipulated or something ...


MDN

[TexasBushwhacker]

7. My the GOP is encouraging people to respond as "undecided"

I also wonder how they come up with their call lists. I've been voting in presidential elections since 1976 and I've never gotten a poll call.

[question everything]

8. There were several op-eds that many chose to hide their Trump preferences

Right now on CBS, polls - again - found that the economy was most important, ahead of COVID-19 and racial justice

[Rice4VP]

9. Polls in PA were mixed. If you looked at RCP recently

Trump was up in some of the Pennsylvania polls

[Azathoth]

19. Trump wasn't up in a single non-partisan poll

Outfits like Trafalgar and Susquehanna and Rasmussen are explicitly Republican operations.

Every reputable non-partisan pollster consistently had Biden up, some outside the MoE.

[Rice4VP]

28. Well...they are still polls. In any case, Pennsylvania

still have millions of votes to count so Biden might eek it out. I think winning back those three and possible Georgia and getting Arizona gives him a little mandate

[BComplex]

10. Again.

Every time republicans pull Karl Rove into the mix, we have the most UNUSUAL issues happen with our elections.

What a coincidence.

[uponit7771]

13. ***VOTER SUPPRESSION !!*** It works, pollsters minimize and dems doing so by saying "just vote"

... as if "just voting" overcomes the inability to do so.

[Azathoth]

16. There's definitely an underlying mechanism at work across all the reputable pollsters

Three possibilities:

1) All pollsters made exactly the same mistakes estimating the composition of the electorate and weighting their samples

2) Many Trump voters refused to talk to pollsters, so the samples were skewed

3) Trump voters lied about their voting intent

I think the answer is some mix of all three, but I have a hard time believing that every single pollster suffered from that level of self-sabotaging group-think, especially after the 2016 polling disaster. So options 2 and 3 are definitely real parts of this.

[Tommymac]

20. 4) Putin

Both cracking of the physical technological infrastructure...

and Disinformation Campaigns targeting precincts backed by him.

Just like 2004. 2012. 2016. and probably 2008 but that was a true landslide under poor circumstances for the Putin backed Rethugs. The chances of them tipping their hand in that one was too great...they knew they could get the Congress in 2010 and obstruct everything after that.

[Azathoth]

24. No. Just no.

Putin didn't do this. There was massive turnout across the country. By most naive metrics, this was a Democratic dream election in terms of turnout.

Texas absolutely shattered turnout records. And it went heavy for Trump, basically imploding the myth that Texas can be a blue state with high enough turnout.

Like it or not, the support for Trump runs *much* deeper in this country than any poll captured.

[Tommymac]

29. Yes. Just Yes.

If folks continue to hide their heads in the sand when it comes to using old outdated insecure 1990's technology in a world filled with 21st Century Threat Vectors...

why are we surprised when the same results keep happening cycle after cycle after cycle...it's the definition of insanity.

And it's been PROVEN Putin interfered in 2016 using modern internet disinformation tactics to target vulnerable precincts with pinpoint accuracy to sway just a few voters in each precinct, but over the whole state that meant a slim win for his Puppet...why on earth wouldn't he use them now????

No. I refuse to believe that many Americans are this stupid. A lot are - too many...but not to the extent we say this election.

[llmart]

23. +1

Also, why did a lot of the GOP Senators not seem very worried about winning reelection? Why did McConnell just chuckle away in that confrontation? And most importantly, why did they ALL back trump on everything? Because he offered them a sure "win" if they continued to support him. He's spent his lifetime surrounding himself with criminals and he has done that again. Those people will stoop to nothing to win.

[KPN]

25. This question needs a thoroughly researched

explanation. In my lifetime, polls taken as a whole have not been this far off the mark in two presidential elections consecutively. Anecdotal explanations don’t cut it. A poll is a survey based on research methodology that has been developed, validated, replicated and shown to be statistically reliable in social sciences now for many decades.

To not be open to this question and not be curious about a reliable, complete and accurate explanation seems naive given the electoral shenanigans we’ve experienced since Nixon.

[Steelrolled]

36. It is "thoroughly researched" each time

by the same types of people who design and run the polls. Not surprisingly there is always an "answer" that the predictions were fine. Just like economists, weather people, etc. People have learned to take their predictions with a big grain of salt, but somehow believe the politics pollsters.

[GoCubsGo]

26. I'm going to go out on a limb here, and say that it's not the polls that are what's wrong.

It's the behavior of one party, that loves to suppress voters, and otherwise lie, cheat and steal.

[My Pet Orangutan]

27. AZ was on the money.

[Steelrolled]

31. Political polls are wrong for the same reason that COVID-19 predictions are so wrong.

Humans are complex, and the issues are mired in politics (the enemy of truth). There are all kinds of problems that our science just isn't that good at - predicting the stock market, sports, weather, and elections. But there are still people in the business of doing it, the best they can, and they are still getting paid.

[ElementaryPenguin]

32. Results from ALL of the exclusive mail-in states EXACTLY MATCHED the polls!!!

Only in-person voting in swing states were skewed in ONE DIRECTION - towards Trump!

Russia likely hacked our precinct totals precisely as in 2016!!

What was to prevent them?

WAKE UP, AMERICA!!

Electronic voting is not secure - and it's costing us our democracy!

[Steelrolled]

38. I don't think that electronic voting is the issue

However, having now gone through a mail-in election, I think that aspect of voting opens some easy ways of fraud. Of course, it is hard to know.

[Baltimike]

34. Because they have been flipping votes for more than 20 years now, and we act like

it's impossible or that "they would NEVER"

Oh yes they would and we better start saying it or it will just continue...

[ElementaryPenguin]

35. Exactly!

[BlueIdaho]

37. Political Polling is a joke.

I’m done paying attention to them.