About the polls: they were wrong on election night because GOPers lied.

[joshcryer]

They either outright refused to say they were going to vote for Trump or actually lied that they weren't voting for Trump. It was the first case in many years that the pollsters failed to mitigate liars in the poll results. 11% of Trump supporters were ashamed to admit they were voting for him.

So keep that in mind as we look at polls. If 11% of people lie to the pollsters come Nov it can easily skew the results, yet again. We got a lot of work to do.

[Fullduplexxx]

1. Or it was deliberate to skew poll results

[WeekiWater]

2. The polls in the last Presidential election were very close to reality.

There were some discrepancies but overall they were very close. Most national polls in the final day(s), contrary to popular belief, were too friendly to Trump. Clinton won by two and most of the recognized polls released on the final day had Trump winning. If Trump supporters wouldn't admit to wanting to vote for him how high to you think these polls would have shown Trump up?



This is one of the least understood and most often misrepresented aspects of the last election. Comey made a big difference and it showed.

[SWBTATTReg]

6. I agree. Comey influenced the elections adversely and should face charges that he did so too...

A bad decision made just on the eve of elections.

[ck4829]

9. +1. The polls reflected the people, the electoral college didn't

[JCMach1]

12. This

[former9thward]

16. The final polls on Monday were almost all Clinton.

Bloomberg Clinton +3, CBS Clinton +4, Fox Clinton +4, Reuters Clinton +3, ABC/Washington Post Clinton +4, Monmouth Clinton +6, Economist Clinton +4, Rasmussen Clinton +2, NBC Clinton +6. The only polls showing Trump with a lead were LA Times tracking poll and IBD tracking poll --both of which were attacked on this site.

https://www.realclearpolitics.com/epolls/latest_polls/president/

[WeekiWater]

17. Final polls were released on Tuesday.

Two of the three went to Trump. Almost every poll release on Monday were very close. Almost every single one was within the margin of error. The polling was very good.

"Clinton won by two and most of the recognized polls released on the final day had Trump winning."

11% weren't lying.

[former9thward]

18. Nope.

The final polls of almost all of the main polling companies --- the ones constantly cited here -- were released on Monday. The tracking polls finished Tuesday -- and they were always attacked on this website.

[WeekiWater]

19. Got ya. I was not differentiating between the method of poll being released.

It really shows how accurate the polling was. Just about every single poll was within the margin of error. Tracking seemed to be more friendly to Trump.

Not an 11% lied issue. More like a really accurate job.

[Polybius]

35. Rasmussen of all polling got it exactly right?

Now I'm worried. I always thought that they severely over-estimated Republican support. They consistently are the only ones to give Trump in the high 40's.

[a kennedy]

3. Yup......not believing a poll except the final results poll on Nov 6 or possibly Nov 7th.

been fooled way to many times before.

[Crutchez_CuiBono]

26. +10 Polls are bullshit often.

>They are used to tell us what THEY want us to believe, not reality. Ask two people....you have a 'poll'.

[DemocratSinceBirth]

4. Polling wasn't awful in 016

But sometimes folks attribute to polling a level of precision that doesn't exist, even with averaging.

[The Velveteen Ocelot]

5. I don't think that will be much of a factor in these elections.

People aren't ashamed of voting for candidates for representative or senator, most of whom aren't as controversial as Trump or as embarrassing to admit to voting for, especially if they're an incumbent. If people aren't ashamed of voting for Roy Moore they aren't easily shamed - and the polls for that election were pretty accurate.

[stopbush]

7. OR, the Russians changed votes at the ballot box, coupled with

Rs tossing out D votes to skew the results.

Yeah, I’m going with that. No need to find a way to blame voters/pollsters/candidates.

[Crutchez_CuiBono]

27. +1

[nini]

30. That's exactly what happened

Im tired of seeing that fact forgotten or ignored.

[bearsfootball516]

32. If it's a fact, you'll need to provide proof.

We know about the massive Russian disinformation campaign, but there no actual proof that they changed votes at the ballot box.

[Autumn]

33. This

Strange how some people want to put blame anywhere but where it belongs. Republicans who voted for him were proud of that fact , they didn't deny him, they embraced him .

[genxlib]

8. The thing that always scared me about 2016 polling...

While they consistently showed Clinton ahead, she usually stayed South of 50%. That means there were enough unclaimed voters out there to swing the majority. With Comey, Russia and 30 years of character assassination weakening her, there was enough gravity to tip that balance.

[Caliman73]

10. Correct.

The polling in 2016 failed to capture a pretty large segment of the population, enough for the nasty surprise, although 77 thousand votes across 3 states is a pretty tight margin.

[0rganism]

11. Trump isn't on the ballot for the midterms

i doubt any "candidate shame" effect is in operation now, and probably won't be in 2020 either as the GOP thoroughly embraces Trumpiness.

that said, the poll that really matters is a couple weeks away. go forth and shift the "likely voter" models in our favor.

[grantcart]

13. Please stop promoting the idea that the polls were wrong on my phone so can't check it

But memory is 538 combined popular vote Clinton + 2.8 and it came on at 2.1

Having said that polling by state or district is never as good as national because they are not as frequent.

Moreover there has never been this intensity on Dem side.

In some races they could be off 4-5%

[Loki Liesmith]

21. Correct take but no one listens

[redstatebluegirl]

14. NEVER believe polls anymore.

Not just the ones who lie, but it is heavily impacted by older adults with land lines. Since we got rid of our land line I have yet to be called for a survey. When we had ours I got tons of calls, almost once a day prior to a major election.

Older white adults tend to have land lines as well as cell phones. Until they figure out how to tap cell phones it will be mostly older Americans.

[USALiberal]

15. I remember people saying nate was wrong, etc. nt

[Loki Liesmith]

20. They weren't wrong

In nearly all states polls were correct. They were correct at the national level. Those states where they were off fell in the natural range of polling error.

[Calista241]

22. I suspect both my parents are Trumpers, but I know for sure that my Mom is.

A pollster called her the other day, and she told them she voted for Hillary in 2016 and would vote Stacy Abrams in November. She was embarrassed to admit to a stranger that she supported Repubs.

I also know that she’s already voted for Brian Kemp. How many people out there are publicly against Trump, but are secretly huge Supporters?

[UniteFightBack]

23. I thought the polls were within the margin of error on election day. nt

[edhopper]

24. Polls don't show

how many people who intend to vote are not allowed to, or whose vote is not counted.

Voter suppression is a main reason Trump is President. He lost the real election.

[spanone]

25. and for the mid terms there is no electoral college

[aeromanKC]

28. Odd that polls were more accurate pre-electronic voting machine

Just sayin ......

[Takket]

29. The big thing people miss in the polling...

Hillary was leading a lot of state level polls but there were at lot of 46% to 42% leads with all the undecideds still out there. If you aren’t polling in the 50s you haven’t won shit yet. And because if Comey the undecideds broke nearly all for drumpf.

[OrlandoDem2]

31. Polls don't vote. People vote! We must GOTV!

[Polybius]

34. 11%?

The polls were not that far off. Hillary won the popular vote by about 2%. Most polls showed her winning by 3-5%. They tightened the last week considerably.