I don't click on right wing sites.

So overestimation may have been based on $$ amounts ?


✌🏻

from
Seven data-driven lessons from the 2025 elections
Democrats outran their polls and swept statewide races from Georgia to New Jersey, on an agenda of affordability and a broad anti-Trump backlash
G. Elliott Morris
Nov 05, 2025

Excerpt:
The other poll that almost nailed Virginia was… This poll that I did the weighting for! Instead of weighting by past vote, we came up with models of what the electorate should look like, including by party ID, and weighted the LV sample to those benchmarks. We had Spanberger +13 statewide, and also nailed several of the close House of Delegates races. In contrast, when I weighted our survey by 2024 past vote instead of party, Spanberger’s margin was just 6 points.

So, in summary about the polls, the best pollsters learn from their mistakes and use rigorous, transparent methods that are adaptable to different non-response environments. The list of pollsters who perform well and use the same methods over and over again is very short. Even good firms that don’t adapt die. Just ask Ann Selzer.


The article contains a lot more analysis
https://substack.com/home/post/p-178051880

FiftyPlusOne:

https://fiftyplusone.news/polls/approval/president

Did you mean to use the phrase “a disturbance in the force” (that’s the actual Star Wars quote you might have been thinking of)

the recent increase in his disapproval numbers.


Qui-Gon Jinn: With your permission, my master, I have encountered a vergence in the Force. Yoda: A vergence, you say? Mace Windu: Located around a person? Qui-Gon Jinn: A boy. His cells have the highest concentration of midi-chlorians I have seen in a life-form.

You have to expect a confused response.

Most folks only watched those once, when they first came out 25+ years ago.

I’ve probably watched the original trilogy a dozen times, and have watched the sequels multiple times.

I could barely get through a second viewing of the Phantom Menace, and have never rewatched Episodes 2 and 3.

And, I'm not a SW nut. Just a casual fan.
Episode 1 is what I've watched the least.

Then you must surely love Andor and Rogue One.

Didn't find Andor that interesting. Not sure I even watched all of it. I may have; maybe not.

Which is, in turn, the prequel of events immediately before episode 4.

It is the most political of all the Star Wars movies and shows- it is the story of the birth of the rebellion, and how diverse people from all walks of life came together to resist the Empire.

Maybe I did watch the whole thing, because it seems familiar.

So, if you don’t remember watching it over the summer, then now’s the time to catch up.

“Polls could be overestimating President Donald Trump’s support following a “huge” miss in the New Jersey gubernatorial race, pollster Nate Silver said.”

Things like the percentage of various groups in the overall electorate. X percent women, y percent black, z percent Republican, etc. That is used to weight the poll if they didn't get enough of each group in their randomly selected respondents.

In 2024 Harris only won New Jersey by 6 points. The polls might have estimated a higher Republican turnout because of that being the most recent election in the state. In particular while Trump has out performed polls when he's specifically on the ballot, in the 2018 mid terms Democrats out performed polls because of an enthusiasm gap since most Republicans don't have Trump's charisma with his voters.

Silver knows why the polls were wrong - he includes GOP biased polls in his aggregate averages, which is why Trump’s approval rating are always 2-4 point higher than on reputable sites like G. Elliot Morris’s site, FiftyPlusOne.

https://fiftyplusone.news/polls/approval/president

(Morris is the guy who ran 538 for its last two years after Silver was fired for allowing 538’s averages and projection model to be deliberately and openly manipulated by Republican pollsters)

IMO, Trump’s support wasn’t overestimated by pollsters, but voter outrage over the economy and Trump’s lawless impunity was.

BUT - since they are not asking every single person, they are an extrapolation (interpretation) from the actual data mapped against the population as a whole.

How accurate that extrapolation depends on how well those sampled model the population as a whole, and how accurate the adjustments made for over/under sampling certain populations are.

It's sot of like flipping a coin 10 times and getting 3 heads and 7 tails, then extrapolating from that that if the coin were flipped a million times 300,000 would be heads and 700,000 would be tails - when the reality is that with a million flips (assuming a balanced coin and uniform flipping) the numbers would be closer to 500,000-500,000.

That's an over-simplification - but that's the gist of how extrapolations from actual data (the number of heads in a limited sample) might not match the actual results.

added more weight to the |GOP side to counter bias claims from the right.

It could be that they went too far with their claims.
Now we need new polls to see where Trumps ratings really are.

😂😂😂

those people who extrapolate, interpret, analyze and summarize the meaning of the poll data. He wasn't being very eloquent or precise in his choice of words and the ambiguity left room for it to be interpreted as though the poll data itself was making a conscious decision to "overestimate" (which of course is impossible).

so are the polls.

More like pollpaganda these days.

IMO

"Instead, the best explanation for 2025 is that voters didn’t know what they were getting with Trump 2.0 last November, but now they do — and they don’t like it."

Uggghhh.

When it is not election season (which seems like an ever shrinking amount of time as I have grown older), the pollsters will only poll "registered" voters or even "all adults".

That can affect polling outcomes significantly.

There are many polls.

Pick the one you like and swear by it.

Swear at the others.

.................... That's why candidates sometimes refer to mystery polls that no one can locate to check.