Shifting Polls

[RecoveringJournalist]

I've begun to see a lot of publicity about a recent perceived shift in polling toward Trump and away from Harris. I would be lying if I said it didn't make me nervous. What have you all been seeing and does anyone have ideas on real ways to counter this shift (if it's real?)

[JohnSJ]

1. They are basing this on one poll which is nonsense because

all the polls are within the MOE, and effectively saying it is too close to call.

I don’t believe Trump is gaining votes, or Harris is losing votes

[stopdiggin]

6. I don't think this is the true picture. Nor is it about a single poll.

I will agree that almost everything swing state is within MO. (and has been mostly all along)

[obamanut2012]

8. It is 100% based on one poll with sketchy methodology

[stopdiggin]

12. The OP is fairly clear in NOT referencing a single poll

And talks about unspecified 'media' plural. So I don't know where we come up with that assertion.

[RecoveringJournalist]

19. Not one poll

OP here. No, my OP is not about a single poll. It's a thought based on several polls which keep changing.

[JohnSJ]

11. That is effectively what will determine the election. The popular vote is meaningless.

because of the electoral college.

[Johnny2X2X]

2. The race has been remarkably static

Harris is up a few points nationally and leads in the key swing states.

GOTV is all that matters now. Whoever has the better ground game is going to win.

[pat_k]

52. Yes. GOTV is all.

I really think/hope more Trump supporters are ambivalent enough about how awful he is to prompt them to just stay home.

[Patton French]

3. The corporate media always pushes a horse race narrative.

It's what they do. Ignore the polls and get out the vote.

[stopdiggin]

4. you're correct

the polls are nowhere near as favorable as they were 3 weeks ago. That's just the way it is.

People can discount and disavow what those findings actually mean - and I predict there will be some of that here - but that is a different argument that what the polls are coming in with.

Reasons? Guess is as good as mine. But I'll go with the very real pull of misinformation - which is practically ubiquitous in the dumpster fire that is today's social media. Case in point, ridiculous lies about hurricanes and relief efforts ...

[RJ_MacReady]

10. they aren't credible

Large swaths of people don't simply flip flop the way polls suggest. Polling in general is a dead metric.

[ScratchCat]

36. This here

Large swaths of people don't simply flip flop the way polls suggest.

The only way the poll would be legitimate is if it polled the exact same people over a time period - which they wouldn't do because it would show the same results every time.

[RJ_MacReady]

39. I could see some people

Being wishy washy sure but this idea laege swaths of people will flip on a dime over a certain issue is absurd to me. "I am no longer am voting goes Harris because the honeymoon is over". Who actually thinks like this?

[Fiendish Thingy]

17. It's NOT "just the way it is"

The averages are deliberately being manipulated by a flood of Red Wave pollsters, just as they were in 2022.

https://www.democraticunderground.com/100219537649

[democratsruletheday]

18. amen....they're flooding the landscape with bogus polls

like you said, JUST like 2022 and we know how that turned out.

[Zeitghost]

48. It turned out

With us losing by 3%.

[qazplm135]

22. I mean the average has her up mid 3s

She was maybe up 4 to low 4s three weeks ago.

Which isn't even a 1 pt change.

When you have a MOE of 3-4 or more for most polls, a 1 pt change is statistically just noise.

Let's assume Harris is up 3.

You'd expect, statistically to see some polls with her up more and some with her up less and evena few with Trump up.

The reality is, we see way more up than down right now

[WarGamer]

24. 538 has Harris +2.6% and RCP has Harris +2.0 and 270towin has Harris +2.7%

[OnDoutside]

5. I'm trying to remain logical and stay with the facts

Lots of voters who popped their immortal coil since 2020 were republican

Lots of new voters now, are young.

500,000 voters registered in the 48 hours after Taylor Swift endorsed Harris.

Millions of previously dem voters were energised by Biden withdrawing and Harris becoming the nominee.

Trump has been more of a nut case since she became the nominee.

[obamanut2012]

7. Wow! You have incredible self restraint

No way I could only post 160 posts in nine years. What's your secret? I am not being snarky, but sincere. I wish I could not post so much and only read DU.

[TeslaNova]

16. This is only my 284th post in 16 years

[RecoveringJournalist]

20. Not as hard as you think

I only post something if I haven't already seen it here. I'm far from perfect. I'm sure I've missed a few things. But I figure why bombard this site with duplicate OPs. And for the most part if it's something I'm thinking about, I find that it's already here somewhere. The less is more ideology.

[Prairie Gates]

9. Trump has some "good" commercials

I say good as in effective. Some of them are plain evil from a moral standpoint ("She's with they/them. Trump's with you/us" ), but they are effective for the audience they're intended to hit. They're really blanketing the swing states with these. The McCormick ads in PA are also troubling. It's October, and they're absolutely pushing out everything.

The Harris response has been pretty good, and has probably significantly blunted the effect of the Trump/RNCC ad campaigns, but they are effective campaigns, so you're going to see erosion. Hopefully, Harris does something with the billion dollars that pushes the thing way in the other direction, but time is short.

[PeaceWave]

27. Not being in a swing state, I haven't seen many of them. But, those I've heard are evil.

[RobinA]

40. I Live In PA

It's pretty bad here in swing state country. They are all narrowing from the fairly narrow gap they were at two weeks ago. I'm planning (or trying to) how the hell I am going to get out of bed and make it to work Wednesday morning. And I realize it probably won't be decided by then, but Trump will have won or be within striking distance at that point.

[GreenWave]

13. Every week somebody drags in a horrible poll.

I would call that poll, "a shifty poll".

[ColinC]

14. My take is this:

National polls will be within the margin of error for national polling, or outside by no more than 3-5 points. Many state polls however are going to have errors favoring Dems by possibly up to 20-30 points depending on what averages are used. Take most polls with a grain of salt.

[tableturner]

45. I disagree...polling errors will favor Republicans, just like in 2022 and recent special elections.

I think we will see the same thing we saw with the recent elections, even in strongly Republican areas, which was a very negative reaction to the abortion decision, GOP extremism, and their fascist leanings (sometimes MORE than leanings).

Then add in the wild enthusiasm and juiced up campaign contributions when Harris became the nominee (to say the least!), and the reality that there are many voters who voted for Trump four years ago who will abandon him this time for the reasons noted above.

It will only take a few former Trump voters jumping to Harris to seal the deal, and I do not see how Trump will avoid that (in fact, with his irrationality, his hateful language, his diminishing cognitive abilities, and his hostility to women being shown daily by him, it actually looks like he is trying to do the opposite).

[ColinC]

47. Polling errors favored Dems in 2022 for statewide polls

At least in statewide senate races. Generic polling was the other way around barely, but the statewide swing races had an average 5.6 point margin error favoring Dems.

For example:

Pennsylvania senate understated fetterman by about ten in the RCP average, and WA understated Cantwell by about 15.

I haven’t looked at the recent specials but I thought they favored Democrats as well -if I can recall.

[tableturner]

50. We are I think actually saying the same thing...see below...

What I meant was that polls in 2022 falsely showed better results for Republicans than what actually happened. Therefore, to me those were polling errors that falsely favored Republicans to do better than they did.

I think we are making the same point. They understated the Democratic vote, so that means that the errors falsely favored the Republicans because their results were worse than the polls suggested.

[ColinC]

57. Oh right 100%

In fact I didn’t realize how off they were in 2022 until I looked at it recently. Repugs wouldn’t have even gotten close to taking back the house if not for disproportionally gerrymandered districts.

[ColinC]

49. In fact here is what would happen if the 2022 average senate polling error was transferred into swing states for Prez

https://www.democraticunderground.com/100219552363

[Fiendish Thingy]

15. What is happening is Red Wave pollsters are manipulating the averages.

And it’s only going to get worse.

https://www.democraticunderground.com/100219537649

[Amishman]

38. While that is a factor, it's more than that

Emerson has recently put out polls with Trump up in WI, PA, AZ, and tied in MI and GA. They are an established reputable pollster (538 has at 2.9 / 3 stars - 9th highest on their list).

Quinnipiac also recently put out some worrying polls with Trump up in MI and WI. They are also a mainstream decent pollster (538 has at 2.8 / 3.0 - 17th highest on their list).

Yet we have those on DU claiming they are bad, biased, red pollsters just because we don't want to hear what they are reporting.

We need to realize that this is going to be very tight and there is a very real chance Trump wins.

[Fiendish Thingy]

42. The Quinnipiac and Emerson polls have far less impact on averages than the dozens of other shitty polls

Two polls among 20+ Red Wave polls released in the past week won’t counter the brazen manipulation of the averages, and dozens more shitty polls are yet to come.

The Quinnipiac polls are likely outliers- they have higher than usual MOEs, and the samples included more republicans than Dems (don’t know the actual breakdown of partisan affiliation in WI and MI)

[Thrill]

21. They do need to start hitting back with tougher ads

The nice stuff doesn’t work in Presidential ads. We get non stop negative Trump ads in NC. Literally none from Harris about Trump being a Felon and Fraudster and his record as President.

I don’t understand it. I’m not sure what the strategy is here. You have every right to be concerned

[dameatball]

23. I am leery of methods, biases and sample sizes in polls. The only poll that counts is in November.

[sky_masterson]

25. I've watched this Two Way (Mark Halperin zoom)

and he is all pushing internal polls showing Harris losing, and Trump surging.
I also saw Cenk Unger go off about the same thing. Panicking.
All the people in the chat and Zoom meeting were eating it all up, but it seems very gaslighty to me.
I tend to believe people like Rachel Bitecofer who says to ignore the polls because they aren't moving up and down like stock markets,people know who they are going to vote for.
Personally I don't see how millions of women will all the sudden forget to vote to restore Roe v Wade.
They also seem to go with the impression that Harris is Hillary. They are different people, different times, and Harris is doing better than Hillary did at this point IMHO

[obamanut2012]

29. Mark Halperin is a wingnut and Cenk is not a liberal

Why in the world would you listen to them?

[sky_masterson]

31. I always want to know what the other side is thinking/doing.

Hell, I've watched hours of RSBN Trump coverage. I'm not being swayed, I'm keeping an eye out. I can normally only take about 20 minutes of Trump before I click out though

[kansasobama]

34. I watch Mark and then it was easy

It was packed with his followers and I got the feeling Mark wants dictatorship.

That was it for me.

[sky_masterson]

41. He is very dishonest in his coverage

He does have a few pretty good Harris supporters that often bring the truth, but it is a big circle jerk.

[clutterbox1830]

26. Trump is always attacking while Harris is mostly defending.

Hurricane damages and perceive lack of relief is getting to people even though that not the case.

Trump lies about the hurricane relief efforts and that's getting media cover. It leaves Harris team is defense mode.
It puts Trump more in the spotlight. It also make Trump look to be the candidate of change. Instead Harris should also be always attacking; reminding the people that Trump withheld aide who city/counties that didn't vote for him or mentioned how stupid he is when he suggested nuking a hurricane will disperse it.

On sidenote, I live in the Pittsburgh area and I don't think Harris ad strategy is very effective. While watching the last Steelers game, there were 4:1 Trump to Harris ads during this game. Steelers game by far highest rating show on TV in half of Pa area. It is not even close. However, it was disappointing to see mostly Trump ads being shown.
Also, Harris's ads are not very effective imo. It always starts off with mention of Trump Project 2025. If people are not aware of it by now then it just turn people off. Just mention that Trump is loser, an idiot, and weird and given examples. It gets people talking.

That is my opinion.

[Dee0909]

28. Scared

I am also seeing the polls moving towards Trump and it is frightening. They are saying that Union Members, the Muslims in Michigan and Black Men are leaning towards Trump. I simply cannot get my mind wrapped around this. As far as Muslims they say they want to punish the Biden/Harris administration and that Donald Trump said he would end the war! What is wrong with these people? I am scared to death! Trump Will plow Gaza under!

[obamanut2012]

30. Your concern is noted

Seven posts since 2018?

[lindysalsagal]

32. These are deliberate winger polls meant to sow mistrust. Setup for the challenges to come.

https://www.democraticunderground.com/?com=view_post&forum=1002&pid=19537649

Here are the winger polls to ignore:
American Greatness,
Daily Mail, co/efficent,
Cygnal,
Orbital Digital,
Echelon,
Emerson - National, AZ, NC, PAFabrizio,
Fox News,
Insider Advantage,
McLaughlin,
Mitchell Comms - MI
The Napolitan Institute.
Noble Predictive,
OnMessage - PA
Patriot Polling - PA
Public Opinion Stratgies,
Quantus,
Rasmussen,
Redfield & Wilton,
Remington,
RMG,
SoCal Data a
The Telegraph,
Texas Gun Rights -
TXTrafalgar,
TIPP,
Victory Insights,
Wall Street Journal

[-misanthroptimist]

33. Shifiting polls sells ads.

Business is business.

[-misanthroptimist]

35. Shifting polls sells ads

Business is business.

[crimycarny]

37. I follow Christopher Bouzy on twitter

I know some don't like Twitter so I'm not posting the actual tweets, but I follow a Twitter account Christopher Bouzy who has been incredibly accurate in his past predictions.

-- He predicted Biden would win Nevada and Wisconsin, and that when early votes were counted Biden would win PA
-- He predicted Harris would pick Walz as her VP
-- He predicted Oz would lose PA by +4 points when others were predicting Oz would win or it would be very close (Fetterman won by 4.9 points)
-- He predicted Kelly Loeffler would lose by 3 and Perdue would lose by 1 (GA Special 2021 election..Loeffler lost by 2, Perdue lost by 1)
-- He predicted the Dems would keep the Senate in 2022.

He is currently predicting a Dem win in PA and thinks Florida is in play.

Oh, and here is something he posted in October of 2020:

1. Biden will win.
2. Trump will claim victory early in the evening.
3. Trump and Barr will claim there was foreign interference and Trump will contest the results.
4. It is extremely important that Biden win by a landslide. Get out and Vote!!

[Newfrontier0101]

43. IEM market today: Harris 86-14

Forget the squirrely polls and Euro betting markets like the one led by cryptocurrency cretin and Trump stooge Thiel and flacked by Nate Silver, a guy so inept that his own polling aggregate company, 538, fired him. No way pollsters can find a representative sample in a culture bred to not answer unidentified calls and resort to manipulation based by unilaterally inflating Trump numbers bc of 2016. Nor can they properly chart the enormous upsurge of recent young voter registrations and Rep defections; they usually just go with identical rates for each party, 95-5 each. I'd bet Trump loses a good 10 pct of the Rep vote.
For a different skew, check out the only American political market allowed to operate, the Iowa Electronic Market, run by the business school at Iowa University. They're usually historically right on the nose and arent looking to game the odds; it's a non-profit, mainly for educational and research purposes. The numbers today in the winner take all market are Dem 0.858 Rep 0.143. They were way down with Biden then took off with Harris and have kept rising. Just so you know its no Dem front; they also predict a Dem House but Rep Senate.

https://iemweb.biz.uiowa.edu/iem_market_info/2024-u-s-presidential-winner-takes-all-market/

[SomedayKindaLove]

44. Not sure how polls work anymore

Yesterday someone on DU posted ten polls from PA. The aggregate had Harris around +1. Trump lead in one of those polls, +2. Two of the polls had them even. Harris lead in 7 of the polls (ranging from +1 to +4). If you added up Harris’ +’s, she was +19. While Trump was +2.

So is Harris given any credit for winning 7 of the 10 polls and drawing even in 2? Seems like that might mean something. But somehow cumulatively she was only +1. I don’t get it.

[Initech]

46. The billionaire criminal class is going to sell us out to authoritarianism so they can become trillionaires.

And fuck them and their billions.

[OrlandoDem2]

51. IT IS ONE POLL!! LETS WORK HARD FOR HARRIS/WALZ!!

[ThreeNoSeep]

53. The purpose of the OP so obvious

No links. No data. We will give your concern the thought it deserves.

[RecoveringJournalist]

58. Another Thought

Or maybe, just maybe, I'm a concerned Democrat who's scared of a Trump win. Feel free to research my profile. I have nothing to hide.

[doc03]

54. Tomarrow they will have a poll that shows Harris ahead, gotta keep it

a horse race for ratings.

[samnsara]

55. one poll scared me then i read it was for a very specific

population. so not so scared now but the headlines made a big deal of it.

[SWBTATTReg]

56. Here we go again, someone comes up w/ a shifting polls narrative. And tries to make hay w/ it. And after the

fact that numerous polls have come out already, and they base all of their recent observations and supposed shifts on this one singular thing. Ignore them. There's been way too many other polls pointing otherwise.