I don't believe these polls Steve Kornacki is talking about.

[madaboutharry]

Who are these polling companies even reaching? Certainly not young millennials and Gen Z. Are they calling people on landlines? I don't even have a landline anymore.

I think former republican congresswoman Barbara Comstock, who just spoke on MSNBC is right. She called Trump a loser and said he is going to remain a loser.

[Polybius]

1. Which polls?

What did they say?

[madaboutharry]

3. He was at his "Big Board" showing that Biden was only 1 point

ahead and it was just terrible for him because in 2020 he was further ahead and now that Trump is under indictment for 91 felonies, well...everyone just likes him more!

He was really annoying.

[brooklynite]

9. He was annoying because you didn't like what he said?

If you disagree with his findings, tell what you think the correct margin will be.

[JohnSJ]

17. The problem is that there are a lot of people who don't answer calls from numbers they do not

recognize, and those people are not adequately factored in.

That number can never be reported in a poll because they cannot determine why a call is unanswered.

State polls should be taken more seriously than National polls, because that is how elections are determined.

[brooklynite]

18. But now you're asserting without evidence that polling firms can't address that issue.

They believe they can. So do the candidates (including Democrats) who hire them.

[JohnSJ]

25. How? If someone doesn't answer a call how do they know it is because they aren't available,

or they don’t answer calls from unknown numbers. They can’t

Show me the evidence they can determine how many people don’t answer calls from numbers they don’t recognize

I have never seen a poll indicate that variable

If you have it I would like to see it

[brooklynite]

29. I'm not a pollster so I can't give an answer to satisfy you.

The key point is not that people don't respond to phone calls like they used to; it's whether pollsters can find alternative processes to compensate. The bottom line is that Democratic candidates don't seem as concerned about the accuracy of the data as you are, and if they did, one would expect that they wouldn't continue to hire pollsters.

You're welcome to discount polling data, but then ask yourself on what basis you're making your own judgements about the likely election results.

[JohnSJ]

31. They use it because that is the only tool they have. There are variables that they base on

assumptions based on past elections. Demographic turnout, favorability, etc.

Independents will determine the outcome of 2024

Medscape had an article suggesting screening for cancer doesn’t save lives.

https://www.medscape.com/viewarticle/995803?ecd=WNL_trdalrt_pos1_230830_etid5805113&uac=143706SN&impID=5805113

Assumptions are made which are flawed because there are variables that cannot be determined

The article’s point is that the metadata analysis is flawed, not that screening shouldn’t be done

[Aepps22]

52. Weren't You?

Weren’t you the same person back in the 22 midterms that kept telling us about how terrible the Dems would do based on your private conversations with James Carville? I agree with you on the polls but you did the exact same thing saying that Jamie Harrison was failing and that Carville was gloom and doom on some exclusive conference call.

[Cha]

62. Good question.

[LowerManhattanite]

69. I do remember that.

It’s (and other odd, embraces of oldhead DLC-grade goofiness and the like) a reason I take certain commenter’s words with a pound of salt. It’s entertaining stuff, but not to be held as serious.

[MorbidButterflyTat]

71. +1

[Demsrule86]

33. The last election showed clearly how terrible polling firms have become...some are now partisan

and they play games by 'weighing' things and manipulating the data...you can't trust any of them...and folks don't answer the phones these days... a waste of time and national polls are the worst...I remember you defending dreadful polls when Fetterman ran...we we were right he did in fact beat Lamb and Oz.

[Celerity]

51. no it did not

The last election showed clearly how terrible polling firms have become

no


The Polls Were Historically Accurate In 2022

https://www.democraticunderground.com/100218239554#post49

[Celerity]

50. a lot of polls (some of them with quite good accuracy records) do not use live calls

[JohnSJ]

58. They use both, online and calling, including cell phones, and I think the demographic in

Last edited Fri Sep 1, 2023, 11:46 PM - Edit history (1)

those polls under represent young people

An assumption has been made by pollsters for years that young people are less likely to vote.

[Celerity]

59. there are all types of polls, including pure online panels

I still see people (not you) here who think online means a Drudge style smash the button as much as you want open poll

[JohnSJ]

60. Still online polls have to solicit people to participate, and the question is how many people in

that online poll don’t want to spend 5 minutes or more answering political questions?

[Celerity]

63. Many are very good at scientifically assembling a large and widely representative panel.

[uponit7771]

46. No is because he's presenting polls out of perspective, Joe Biden is doing par for this century

[MorbidButterflyTat]

70. He was annoying

because he's a drama queen claiming trump is even more popular since being indicted multiple times.

Just like you are annoyingly putting words in others' mouths, and bossily telling them what to do.

[brooklynite]

76. He's not claiming Trump is "more popular". He's claiming Trump is more popular WITH REPUNLICANS

Which is true.

[womanofthehills]

21. Worrisome - tied in latest polls

Morning Consult has Biden up one point and You Gov has Trump up by one.

[Demsrule86]

34. Bullshit polls by bullshit pollsters...we don't elect directly thus state poll should be better but

you can't trust any of them.

[Polybius]

66. It's way too early to poll

The election isn't this November. Even if it were, that would still be two months out. I ignore those polls too.

Now, if Biden and Trump are tied on October 30, 2024, then I would worry. Would you?

[LakeArenal]

2. I haven't believed a poll since Comey screwed Clinton.

[Bev54]

4. It is all BS, I turned it off and then he was on again and I turned him off again.

Sick of their bullshit.

[montanacowboy]

5. I turned that off right quick

I am fed up and sick and tired of "these so called" polls and fuck Steve brown pants. I cannot stand that screwball with his '"magic" chalkboard. This shit he is spewing is going to become a self fulfilling prophecy and we are having it banged into our heads day in and day out how Biden can't win and he is so old and the Democrats are worried, blah blah blah.

Will the mainstream media love it when Dump is elected and he shuts down all the blabbermouths? Will they love what they created then?

[inthewind21]

22. Yet

when a favorable one for DEMS come out they are plastered and cheered all over as if they were the holy grail.

[Demsrule86]

35. I don't believe any of them.

[wnylib]

64. ?

[MorbidButterflyTat]

72. Bull

poopie.

[inthewind21]

23. What

THEY created? You seriously think Trump is nothing but a media creation? Did they single handedly create all his cult members as well?

[Demsrule86]

36. Trump's base is not large enought to win a General.

[wnylib]

65. It was in 2016.

While I am skeptical about the polls, I think that we cannot afford to assume that Trump could not squeeze out a win, due to the electoral college.

[onecaliberal]

6. It's complete and total bullshit.

[Emile]

7. Talking heads want a close election for their job security.

[brooklynite]

10. And you're telling us it WON'T be a close election?

Unlike all the recent ones?

[Emile]

13. A lot has happened since the last election. Recent elections

have favored Democrats.

[brooklynite]

15. We're not arguing about who's likely to win; we're arguing about the vote margin,

[Emile]

16. No you are arguing.

[brooklynite]

19. The complaint in this thread is that polling pointing to a close election is incorrect.

The alternative to that position is that the election will not be close. Which do you think it is?

[onenote]

45. 2022 is a recent election. And it shows a closely divided electorate

It resulted in a split Senate and a narrow repub house majority. In terms of total vote, in house elections repub votes outnumbered votes for Democrats by 50.6% to 47.8% and in the senate, it was 49.0% Democrats and 49.1% repubs.

[Emile]

48. Wasn't the Republicans expecting large wins in 2022? The Red Wave

that fizzled.

[onenote]

56. Not getting large wins isn't the same as the election not being close. In fact its the opposite.

[Celerity]

53. In 2022 US House national number, Rethugs outvoted Dems by over 3 million total

https://en.wikipedia.org/wiki/2022_United_States_House_of_Representatives_elections

Popular vote

Rethugs 54,506,136

Dems 51,477,313


Rethugs +3,028,823

https://en.wikipedia.org/wiki/2022_United_States_House_of_Representatives_elections_in_California

remove CA (capped by the Electoral College in terms of impact for POTUS)

Popular vote

Dems 6,743,737

Rethugs 3,859,666

Dems +2,884,071

and we were outvoted by almost 6 million nationally

[Demsrule86]

37. No, I don't think it will be close.

[brooklynite]

38. I do and so does the DNC.

[Polybius]

67. It'll probably be within 3 points

And no candidate will get over 315 EV's. Just my guess, maybe you'll be right and I'll be wrong.

[MorbidButterflyTat]

73. Why do you keep doing that?

This is what Emile said: Talking heads want a close election for their job security.

JFC.

[maxrandb]

8. Is he wearing khaki pants and a polo shirt?

[MorbidButterflyTat]

74. Yes

he's just dreamy.

[karynnj]

11. I saw his coverage of the polls

I think that there is a real problem with polling that goes beyond landline vs cell phones. Many serious articles have explained that the pollsters are lucky if they got even 5% of people in their sample to respond - the articles referred to response rates of 2 to 3 percent. Years ago when repeat calling got response rates well over 50%, pollsters made the assumption that those who respond in a given demographic category are similar to those who don't. Obviously, this is a necessary assumption to make, but given the VERY low response rate, that assumption becomes harder to accept. In addition, the pollsters need to have population estimates for the demographic cells -- including Democratic, Republican, Independent. If fewer people now identify as Republican are fewer than those "population" estimates, estimates will be distorted. (I suspect that former Republicans would now identify as Independent - making the independent category more conservative.)

[JohnSJ]

14. Exactly. How many people answer calls from numbers they don't recognize? I sure don't, especially

with all the spam calls out there.

[JohnSJ]

12. I have no idea, but I believe that people who don't answer calls from numbers they do not recognize

because of all the spam calls occurring, are not adequately factored in.

I do not answer calls from numbers I do not recognize, and I think a lot of others don't either. The number of people who don't answer pollster calls can't be known. I suspect those people who answer calls from numbers they don't recognize are also from an older demographic.

I think young people are under sampled for the same reason.

National Polls are not a good reflection of a Presidential Election, because that is not how president are elected, which is why I take state polls more seriously.


I think if women, young people, and minorities come out in a big way in 2024, we will win.

[chia]

20. Agree 100%

[honest.abe]

26. +1

I think the pollsters won’t admit it because it’s a threat to the entire polling industry.

[Tree Lady]

61. Pretty much everyone I know sceens calls

because of all the spam so who are they getting? Super old folks?

[Arthur_Frain]

24. If it helps, think for a moment about the likelihood of who responds to these polls.

Whether it’s a land line or not, I never had time for any surveys when was in my 20s-30s.

Also, I don’t know why anyone would want to respond honestly to any phone canvasser.

People who take polls are looking for validation. That’s what they get, and it’s confirmation bias writ large.

[Emile]

28. The 2016 election proved Republicans are easily conned.

So who in all likelihood would answer a phone from an unknown caller?

[yardwork]

27. We can't trust the polls and it's too early anyway.

My advice is to forget the pills and work like hell to get out the Democratic vote.

[BlueIdaho]

30. M$M has a vested interest in "horse race" polling results

There are millions of dollars at stake in feeding the polling monster. It’s as predictable as clockwork and deeply cynical. Ever since M$M disbanded their news divisions and folded the programming into their entertainment divisions they have lost credibility. Today the all news all the time networks see it as a necessary evil to drive viewership.

As the saying goes - M$M isn’t there to deliver programming to you, the programming is there to deliver you to advertisers.

[LiberalFighter]

32. IMO the numbers are misleading.

The bloated crap got 45.9% of the votes in 2016.
Biden got 51.3% and the bloat got 46.8% despite being the incumbent.

When Steve Kornacki comes on they are only talking about how *rump does with Republican voters. And they are only about 30% of the voters. There are the Democrats and Independents. If any Democrats don't vote for Biden they won't be voting for *rump. If he gets any support from Independents I doubt it will be as many he received in 2020.

*rump killed off a lot of his supporters during Covid.
There are people that are no longer Republicans. It may not seem like a lot but they are likely to be actual voters. And if they vote for Biden that is a two vote switch advantage for Biden. If they vote for someone other than a Democrat or Republican it is only a one vote advantage for Biden even if he doesn't receive their vote.

How many of those idiots will really vote for *rump again if they think 2020 was stolen and 2024 will be stolen too? A good message to send to them all.

Then there are those in prison that can't vote and in some states if they are a felon can't vote at all.

[BComplex]

39. I don't believe those polls, either, madaboutharry. Anyone with two brain cells to rub together

can plainly see that since the 2020 election (where Biden won by about 7 million more popular votes) the entire republican party, and trump in particular, has lost huge numbers of supporters. Partly from trump fatigue, partly from republican insanity, and partly from abortion/book-burning/LGBTQ/voter disenfranchise issues.

Where the pollsters got their numbers during the "red tsunami" that was going to happen in 2022, are still getting the wrong numbers in their polls.

[honest.abe]

40. I think the whole polling industry is in trouble.

Many people especially young people simply wont answer or respond to pollsters or wont even answer calls from numbers they dont recognize. I am on old f'kr and I also dont answer calls from unknown numbers. And even if for some reason I mistakenly answer the call and its a pollster I hang up. i dont want to be bothered. I think Im not that unusual.

So the data is extremely skewed to only people who are willing to take unknown calls and take a poll.

[Recycle_Guru]

41. it is all about getting a reliable sample plan and weighting it correctly

anyone knows the statistics part is just math--getting a valid sample representative of the population you're measuring is very tricky.

[Sympthsical]

42. Polls are fine

I had no problem understanding 2022 based on the polls.

They're just snapshots. The only poll that matters is the one at the ballot box. However, politicians and groups can use these snapshots to fine tune messaging, see if they've been neglecting something, figure out where to spend money.

Which is why Democrats use pollsters regularly.

If polls weren't at least vaguely accurate, no one would use them.

But I do like the, "I feel like it works this way, so I'm going to believe it does" explanation of polling that happens every single time there are numbers people dislike. Who needs math? I have vibes and soulthoughts!

Let me toot my own horn. Note the date:

https://www.democraticunderground.com/?com=view_post&forum=1002&pid=17271873

I got my information from the same polls people were screaming about until the very end.

[Celerity]

49. The Polls Were Historically Accurate In 2022

https://fivethirtyeight.com/features/2022-election-polling-accuracy/

Let’s give a big round of applause to the pollsters. Measuring public opinion is, in many ways, harder than ever — and yet, the polling industry just had one of its most successful election cycles in U.S. history. Despite a loud chorus of naysayers claiming that the polls were either underestimating Democratic support or biased yet again against Republicans, the polls were more accurate in 2022 than in any cycle since at least 1998, with almost no bias toward either party.

Of course, some pollsters were more accurate than others. And today, we’ve updated the FiveThirtyEight pollster ratings to account for each pollster’s performance in the 2022 cycle. Our ratings are letter grades that we assign to each pollster based on historical accuracy and transparency. (You can read exactly how we calculate pollster ratings here.) They’re one of many tools you should use when deciding how much stock to place in a poll.

Before we reveal the best- and worst-rated pollsters, let’s start with our regular review of polling accuracy overall. We analyzed virtually all polls conducted in the final 21 days before every presidential, U.S. Senate, U.S. House and gubernatorial general election, and every presidential primary, since 1998, using three lenses — error, “calls” and statistical bias — to conclude that 2022 was a banner year for polling.

In our opinion, the best way to measure a poll’s accuracy is to look at its absolute error — i.e., the difference between a poll’s margin and the actual margin of the election (between the top two finishers in the election, not the poll). For example, if a poll gave the Democratic candidate a lead of 2 percentage points, but the Republican won the election by 1 point, that poll had a 3-point error.

snip

[Sympthsical]

54. I found their pollster ratings really useful in 2022

I'm a hobbyist and not an analyst, so I use them more to get a general sense of things for my own edification. And part of that is looking at bias. I remember noting at the time that some people were really pushing Morning Consult and YouGov collaborations on DU. I noted that they were pretty biased to the Left and got a lot of push back for that (one person kept screaming at me - they're also in this thread). But a lot of the blue Dobbsian wave was predicated on Morning Consult stuff.

Having that history of performance and being able to see methodologies is useful for filtering out bias and noise and adjusting expectations.

One of the biggest things I see are people making claims about landlines and cell phones and things. But the results I always found completely skewed were the online surveys. I don't know why it is, but it appears that Democrats are all about taking online surveys.

There aren't a lot of polls I just dismiss outright, but online surveys need to come with a salt truck.

[Celerity]

55. Some the most accurate polls use at least partial online methodology

https://projects.fivethirtyeight.com/pollster-ratings/

[Sympthsical]

57. I left out an important component

I meant paid online marketing surveys.

The paid marketing ones were always kind of messy. I looked into them a while back, and it seems this kind of polling frequently gets run through marketing research firms paying people to fill out online product and service surveys all day. Like SurveyJunkie and things.

Morning Consult and to a smaller degree YouGov have relied on those, and so their data always seem a bit wonky to me. And historically, it has been. They were crazy off in 2020, iirc.

I think it's just that they haven't found a good way to account for and adjust results from people who are the type to take a lot of online surveys.

[budkin]

43. DO NOT WATCH CABLE NEWS

It's simply not a credible source of news.

[Mr.Bill]

44. I don't know anyone under 30 who answers their phone at all.

As in never.

[Azathoth]

47. Trump will get 45% of the vote even if he's sitting in a federal prison

That is just the reality of the country we live in right now.

A majority will break for Biden though.

[Paladin]

68. As usual, Kornacki is WAY too happy, reporting about these latest polls.

Fuck him and his new blue-gray slacks (I guess he finally wore out the khakis). I absolutely loathe him.

[MorbidButterflyTat]

75. Right there with ya.