It's the Pollsters, Stupid

Saturday, November 02, 2024

Dead-heat poll results are astonishing " and improbable, these experts say

It's possible the tied race reflects not the sentiments of the voters, but rather risk-averse decision-making by pollsters

Posted by Corky at 11:40 AM | 17 COMMENTS | permalink | Comment on This Entry |

Alternate links: Google News | Twitter

Comments

Admin's note: Participants in this discussion must follow the site's moderation policy. Profanity will be filtered. Abusive conduct is not allowed.

"Large numbers of surveys would be expected to show a wider variety of opinion, even in a close election, due to the randomness inherent in polling.

The absence of such variation suggests that either pollsters are adjusting "weird" margins of 5% or more, Clinton and Lapinski argued " or the following second possibility, which they deemed more likely.

"Some of the tools pollsters are using in 2024 to address the polling problems of 2020, such as weighting by partisanship, past vote or other factors, may be flattening out the differences and reducing the variation in reported poll results," they write.

Either explanation "raises the possibility that the results of the election could be unexpectedly different than the razor-close narrative the cluster of state polls and the polling averages suggest", they added."

Add in the way the Media promotes the 'Close Race!' to push profits, and it could still be a race that's not all that close.

#1 | Posted by Corky at 2024-11-02 11:49 AM | Reply

It's possible the tied race reflects not the sentiments of the voters, but rather risk-averse decision-making by pollsters
Posted by Corky

#2 | Posted by censored at 2024-11-02 12:11 PM | Reply | Newsworthy 1

... It's possible the tied race reflects not the sentiments of the voters, but rather risk-averse decision-making by pollsters ...

... or maybe the pollsters are just reporting what they are being told by those they poll?

#3 | Posted by LampLighter at 2024-11-02 12:41 PM | Reply

or maybe the pollsters are just reporting what they are being told by those they poll?
#3 | Posted by LampLighter

Large numbers of surveys would be expected to show a wider variety of opinion, even in a close election, due to the randomness inherent in polling. The absence of such variation suggests that either pollsters are adjusting "weird" margins of 5% or more, Clinton and Lapinski argued - or the following second possibility, which they deemed more likely.

"Some of the tools pollsters are using in 2024 to address the polling problems of 2020, such as weighting by partisanship, past vote or other factors, may be flattening out the differences and reducing the variation in reported poll results," they write.

Either explanation "raises the possibility that the results of the election could be unexpectedly different than the razor-close narrative the cluster of state polls and the polling averages suggest", they added.

#4 | Posted by censored at 2024-11-02 01:50 PM | Reply

@#4 ... due to the randomness inherent in polling. ...

But shouldn't such randomness tend to iron out the differences between the pollsters rather than emphasis those differences?

Isn't that why pollsters use random sampling?

#5 | Posted by LampLighter at 2024-11-02 01:56 PM | Reply

But shouldn't such randomness tend to iron out the differences between the pollsters rather than emphasis those differences?
Isn't that why pollsters use random sampling?
#5 | Posted by LampLighter

That's the desired result, but there would still be deviation, especially given the relatively small sample size for many of these polls.

Nate Silver: Cheating' Pollsters Are Putting ----er on the Scale'
"Oh, every state is just +1, every single state's a tie,' No! You're ---- herding! You're cheating! You're cheating!" said Silver

Silver slammed GOP-leaning polling firms for continually putting Trump at an advantage in many states, accusing them of making a prediction that won't put them "too far on a limb."

He added, "Your numbers aren't all going to come out at exactly 1-point leads when you're sampling 800 people over dozens of surveys. You are lying! You're putting your ---- ----er on the scale!"

According to Silver, "If a pollster never publishes any numbers that surprises you, then it has no value."
www.msn.com

#6 | Posted by censored at 2024-11-02 02:08 PM | Reply | Newsworthy 1

Polls, Damn Polls, and Reality

And...

About 77 hours left: What's your prediction?

#7 | Posted by Hans at 2024-11-02 02:21 PM | Reply

@#6 ... Silver slammed GOP-leaning polling firms for continually putting Trump at an advantage in many states ...

Much ado about poll results that still are within the margin of error.

For example, the most recent Rasmussen Reports poll has fmr Pres Trump ahead by 2%, with a margin of error of 2%.

... Your numbers aren't all going to come out at exactly 1-point leads when you're sampling 800 people over dozens of surveys. ...

Is it really exactly 1%, or is that the result of rounding error? Maybe one poll shows 0.6% and then the next shows 1.4%.

Both round to 1%.

Taking a step back, I do agree that my opinion is that the polls are showing a close race when I would not want the race to be this close.

But I'm not going to go around and make up conspiracy theories why I think the polls are so close.

#8 | Posted by LampLighter at 2024-11-02 03:16 PM | Reply

www.youtube.com

Pres historian takes a few Youtube Shorts seconds to confirm the Media Bias he's seeing.

#9 | Posted by Corky at 2024-11-02 04:25 PM | Reply

#8 | Posted by LampLighter

Are you ignoring the likes of the Unskewed Polls guy from 2012? Why are you making ALL the pollsters seem like they're up to snuff?

#10 | Posted by rstybeach11 at 2024-11-02 04:38 PM | Reply

@#10 ... Why are you making ALL the pollsters seem like they're up to snuff? ...

That's not my intent or opinion. Some pollsters are better than others, and some have a political tilt (can you say Rasmussen Reports, I knew you could). However, even with that in mind, the polls indicate a tight race.

Before I start going with the conspiracy theories as to why the polls show a tight race, I'd rather wait until the election results are counted and compared to the polls.

Until then, I do have one inkling why the polls might be shown to be incorrect.

It has to do with the annoyed women that Danforth has noted may be undercounted. That combined with the Republican view that if a wife of a Trumper actually vote for VP Harris but tells her hubby that she voted for fmr Pres Trump, that is grounds for divorce. In the voting booth is the only place where her vote would be secret. She would not want to tell the pollster on the phone her actual vote while her hubby is listening.

#11 | Posted by LampLighter at 2024-11-02 06:21 PM | Reply

"It has to do with the annoyed women that Danforth has noted may be undercounted."

Clarification: HAVE BEEN undercounted. In every election since Dobbs.

Otherwise, spot-on.

#12 | Posted by Danforth at 2024-11-02 06:28 PM | Reply

@#12

Thanks for the correction.

I agree. Too soon to tell if they will be under counted in this election.

#13 | Posted by LampLighter at 2024-11-02 06:34 PM | Reply

It's the Russian FAGATS, stupid

#14 | Posted by LegallyYourDead at 2024-11-02 11:03 PM | Reply

But I'm not going to go around and make up conspiracy theories why I think the polls are so close.
#8 | Posted by LampLighter

Well, I'm not going to argue how statistics work. That's a semester of somewhat advanced math and even then half the class won't get it.

Instead, I'll just copy/paste what Corky copy/pasted from the article in #1 and I copy/pasted in #3: "Large numbers of surveys would be expected to show a wider variety of opinion, even in a close election, due to the randomness inherent in polling. [...] "

#15 | Posted by censored at 2024-11-03 09:59 AM | Reply

@ LampLighter

TonyRoma has posted an article that goes into more detail here:
Silver: Swing State Polls are Only Close Because of Herding
drudge.com

I found helpful the stat that Tony copy/pasted ("Based on a binomial distribution - which assumes that all polls are independent of one another, which theoretically they should be - it's realllllllllllllly unlikely. Specifically, the odds are 1 in 9.5 trillion against at least this many polls showing such a close margin."

#16 | Posted by censored at 2024-11-03 10:22 AM | Reply

Anyone still believing polls is a fool.

They've dropped the ball the last three election cycles and keep trying to tweak their formula to get it right.

But, I think, the problem isn't their formulas, it's that they haven't adapted to modern changes in how we communicate. No amount of tweaking is going to make up for bad sampling.

#17 | Posted by jpw at 2024-11-04 09:31 AM | Reply