*****Breaking -- Harris' Lead Grows When Eliminating Biased, Republican Leaning Pollsters

[Chasing Dreams]

I'm sure many of you are familiar with Simon Rosenberg / The Hopium Chronicles. He is a renowned Democratic Strategist who correctly predicted that the so-called red wave of 2022 would not materialize. Part of his analysis then -- and now -- relies on removing Republican leaning pollsters from polling averages and analytics:

It’s Time To Acknowledge Red Wave Polling As A New Form Of Partisan Political Activity - One of the central arguments we made in our battle against the red wave in 2022 was that the explosion of GOP and right wing aligned polling made the polling averages unreliable. As I wrote to you on Thursday, it turned out to be true as the polling averages, gamed by a flood of right aligned polls, told us, falsely, that a red wave was coming. This problem of there just being too much right-aligned polling is still with us in 2024 and it may about to become a serious problem.

So how many polls and pollsters that we see in the averages have GOP or right-wing affiliations? I count at least 22:

American Greatness, Daily Mail, co/efficent, Cygnal, Orbital Digital, Echelon, Fabrizio, Fox News, Insider Advantage, McLaughlin, Noble Predictive, Public Opinion Stratgies, Quantus, Rasmussen, Redfield & Wilton, Remington, RMG, The Telegraph, Trafalgar, TIPP, Victory Insights, Wall Street Journal (some have also suggested I add SoCal Data and The Napolitan Institute. Will look into that).

https://www.hopiumchronicles.com/p/winning-the-house-with-derek-tran

This afternoon I decided to calculate the unbiased poll averages Nationally and in the Seven Swing States using the following methodology:
1. Only polls from September 1 to present.
2. Only likely voter (LV) polls.
3. Only Harris vs. Trump, with the exception if the unbiased pollster only had a LV poll with third party candidates it was included.
4. Eliminate all Republican lean polls, as well as polls that just popped up without any background on who they are.

Here are the results (formatted the best I could)

----------------------- Known, Unbiased Pollsters __________ 538
----------------------- Harris Trump Harris Lead_________ Harris Trump Harris Lead
National______ 50.2% 46.6% 3.6% ______________ 48.4% 45.9% 2.5%
Pennsylvania_ 49.4% 47.5% 1.9% ______________ 47.8% 47.3% 0.5%
Michigan_____ 49.8% 46.8% 3.0% ______________ 47.9% 46.4% 1.5%
Wisconsin____ 50.3% 47.9% 2.4% ______________ 48.4% 46.8% 1.6%
Nevada_______ 49.0% 46.3% 2.8% ______________ 47.8% 46.9% 0.9%
North Carolina 48.4% 48.4% 0.0% ______________ 47.3% 48.2% -0.9%
Georgia_______ 47.3% 48.8% -1.6% ______________ 47.1% 48.4% -1.3%
Arizona_______ 47.3% 48.6% -1.2% ______________ 46.8% 48.1% -1.3%

The table shows a 0.8 to 1.9 % bias toward Trump when compared to 538 in terms of our net lead in the National, PA, MI, WI, NV, and NC polls. There doesn't appear to be much of an effect in AZ or GA. Still, the overall effect of the Republican leaning pollsters "flooding the zone" is to do what they intend: depress Democrats, and make things appear closer than they are.

Remember, the LV models for the known pollsters are designed to replicate 2020 turnout. There is no Dobbs effect included. We have outperformed polls since 2022 and they have under performed. Including Trump in the 2024 primaries. WE ARE AHEAD.

And, my fellow DUers, let's stop posting the junk polls from the right wing sources noted above.

[senseandsensibility]

1. I've been posting about this too

Simon Rosenberg is a GEM and has saved my sanity the past few weeks. He's on twitter, and I often see him on Lawrence's show. Very level headed and backs everything up.

[Chasing Dreams]

2. Yup, I've seen your posts

Thanks for helping inspire me to do this analysis. I'm very tired of the right wing huckster polls!

[senseandsensibility]

3. Thanks for doing this!

Those swing state polls though. Can't help the feeling that they are WAY too close any way you look at them. Still, as Simon says, I'd rather be us than them.

[wryter2000]

40. If you remember how we have been overperforming polls

It isn't so scary.

But I'm not going to stop making calls to voters until we have won.

[wryter2000]

39. I also had a fight on my hands

When I insisted the race was not tied.

[Fiendish Thingy]

4. Further essential reading regarding the current state of polling:

https://prospect.org/politics/2024-09-25-polling-imperilment/

[Chasing Dreams]

8. I read the article a week ago.

What can I say, I’ve been following polls since i was a boomer kid. Yup, fractions opened me to a fantastic world of BAs, ERAs, bowling averages and yes, polls. I don’t take the results THAT seriously, but I get really pissed off at republican pollsters flooding the zone with bullshit.

[Prairie Gates]

5. Get rid of everything you don't like, and the results improve!

[Chasing Dreams]

7. Nope. Read the article I linked and it will make more sense.

And I sincerely mean that. The more people who understand this insidious right wing strategy, the better. Thanks.

[TheRickles]

23. But what is the definition of "a GOP affiliation"?

If that's the criterion by which a poll is accepted or rejected, it seems a bit iffy. I'd rather see an average of polls that have proved most accurate in recent elections, regardless of their political leaning. The better the track record, the more trustworthy they would be. I've seen 538 do such an aggregation in the past, but not recently.

[SomedayKindaLove]

6. Post Dobbs...

And post January 6th. Trump lost R support on January 6th, I know 2 who will never vote for him again after that.

[BWdem4life]

9. WI + MI + PA 270

NV is likely to stay blue. PA is the one they would have to steal (and then keep NC and GA).

If they did take NV and AZ and GA (plus PA) they could afford to lose NC.

But that all hinges on PA which is polling Blue.

I am cautiously and nauseously optimistic.

[littlemissmartypants]

16. Hear! Hear!

Especially to the "nauseously" part.



❤️

[Self Esteem]

10. Even getting rid of the Republican-leaning polls indicates how close of a race this is.

[Blaukraut]

14. Close enough to lose via suppression, theft, and litigation

We're toast.

[Chasing Dreams]

17. I hear you, but we have an arsenal too

An arsenal of great elected officials including Biden, governors and attorneys generals of PA, MI,WI, NC, NV, AZ who control the National Guard and much of the litigation. We have professionals, they have lunatics.

[littlemissmartypants]

18. I'm not toast.

I'm more of a semisweet chocolate chip cupcake with cream cheese Icing.

Speak for yourself, toast.



❤️

[Elessar Zappa]

37. No, we're not. People were saying that in 2018, 2020, and 2022.

And almost every election and liberal ballot measure these past four years has outperformed what the pollsters said.

[GreenWave]

28. Imagine if such polls were never included to push the undecided...

I suspect bigger leads. Still where are the Harris bumps for the DNC and the debate clobbering?

[liberal N proud]

11. Election night is going to be scary as hell

I will need something to get it out of my mind

[Chasing Dreams]

13. If you live in a cannabis legal state

then I highly recommend Keef Cannabis-Infused sodas. Get a case. This may go on for days or longer.

[Self Esteem]

15. Yes. I think people need to prepare themselves for a very long election night.

I get wanting to be confident and optimistic about November but the facts remain: this is likely to be a very close election and that means it could take days before every state is called.

I too would love for a 2012-like night - and who knows? Maybe we get it (tho, Obama was polling far better in his must-win states than Harris is). But it's just hope at this point to expect an easy night. And I fear those who have convinced themselves this will be a cakewalk - like some saying Harris will win Pennsylvania by 15 or so points (absolutely not happening) are going to be the ones most negatively impacted by a close election.

I am absolutely prepared for a long outcome, especially after 2020. If she somehow is declared the winner early, hell yeah! But I'm not expecting it.

[aggiesal]

12. Only poll that matters is the one taken on 11/5/24 n/t

[littlemissmartypants]

21. That's right, aggiesal. 💯😸❤️

[slightlv]

19. With all the vote suppression techniques being employed

by maga politicians and elected "officials", I just came to the conclusion a long time ago not to listen to the polls. Besides, as a former test creator, I know intimately how questions can be worded to get the response someone wants. Much more important to me is how she is greeted in cities which she visits, the numbers at her rally, and how she manages her "day job" while running her campaign. Yes, she has a lot of helpers and aides in both sectors of her life right now, but that would be the same way it would be if she was Prez. No person is an island, as the poem goes. It's how you motivate people and encourage them to help that counts, and she's showing the world how it's done. I'm so proud of her -- and while she's doing all this, her family doesn't seem to be suffering even a small bit. Most of us would find this kind of multitasking too overwhelming. She finds it just another day. She'll bury trump in the cesspool from which he rose!

[littlemissmartypants]

22. She's spectacular. ❤️

[lucca18]

20. Wow! Thank You!

I appreciate your time and effort!
BLUE WAVE!

[kerry-is-my-prez]

24. I don't agree with some of the polls you have listed.

[Chasing Dreams]

35. They aren't mine - they are from research led by renowned Democratic Strategist Simon Rosenberg

Please go to the link I provided, and carefully read his analysis. Thank you!

[AdamGG]

25. I hope this is true

It could make sense, but I'm not sure if WSJ or even Fox should be lumped in with the hackiest partisan pollsters. They're long term players and don't want their results to turn out to be embarrassingly off.

Is the point of the alleged strategy to boost pro Trump turnout because they think he's in a dead heat rather than trailing or to give him an excuse to dispute the results (or I guess possibly both)?

[WarGamer]

26. The problem is... ONE State of the Blue Wall and Trump probably wins.

GA and AZ look like they're going back... I think the storm seals NC as GOP hold.

Blue Wall sweep and Trump is flushed.

1/3 and Trump is POTUS v2.0

It's damn close.

GOTV.

[displacedvermoter]

31. If Robinson is trailing by same double digit

margin he is now come election day, and the full FEMA effort is up and running by then, not sure how you see NC being "sealed".

And not sure that Lake trailing Gallego in AZ by another double digit margin is a clear indication that state is "going back". You are anticipating pretty dramatic ticket splitting in both states.

[WarGamer]

32. 4 weeks... we'll know soon.

[Doodley]

27. Why wouldn't the pro-Trump pollsters cheat? Pro-Trump media and pro-Trump lawmakers cheat all the time.

[Amishman]

34. Don't think in terms of cheating, think in terms of data model bias

All pollsters normalize their data against a model of what they think the electorate will look like - number of R/D/I/O and a bunch of other factors.

Why do they do this? because they can't control the exact distribution of those who choose to respond plus people who are politically engaged are far more likely to both to spend time responding to a poll. Lots of little factors why they can't just use the data collected as-is.

This is where the rub comes in - republican leaning pollsters have data models that are more bullish on various R favoring factors, which is why they consistently slant R versus the rest. Here's the rub - they could be right. Trafalgar was the most accurate pollster in 2016, and in recent years was a A or B ranked pollster by 538 and they are hands down the most R biased. Why? Because we've had several elections over the past decade that have had big polling misses that understated R support.

Modeling the electorate is as much an art as it is a science, and unexpected aggregate misses do happen.

I do not agree with the approach of ignoring polls from right leaning pollsters, it gives a false sense of security. It's like claiming your football team was undefeated - if you ignore the games they lost.

[Chasing Dreams]

36. Accuracy is measured by pollsters' LAST survey each cycle.

The issue with right wing pollsters flooding the zone with junk results — except the last one each cycle — is that they are designed to muck up dem strategies about where to invest time, staff, and money during the election; depress dem turnout and enthusiasm; and generally muck things up.

Trafalgar and others will magically shift their polling in the end to claim accuracy. And no one remembers the bullshit they spewed along the way.

[GoCubsGo]

29. I recommend this week's Science Friday on NPR with regards to polling.

It's in podcast form, so you can listen on-demand. Ira had on a polling expert, who had some really interesting insight on polling. He said most of the polls out there are crap, and explained why.

[displacedvermoter]

30. Thing is a supposedly knowledgeable person like

Robert Reich writing about the "tightening polls" causes unnecessary anxiety, and lessens enthusiasm with less informed voters.

Great numbers work, thanks!

[Emile]

33. My thoughts all along.

[kerry-is-my-prez]

38. Napolitan is a news network that uses RMG to do its field work.

I have the impression that they are one and the same.