FWIW, within the last five years, both my kids had lessons and essay assignments in high school about King’s letter.
Sorry this is from a paid site, but here’s an article from the Atlantic: (gift link)
The 2016 election lives in popular memory as perhaps the most infamous polling miss of all time, but 2020 was quietly even worse. The polls four years ago badly underestimated Trump’s support even as they correctly forecast a Joe Biden win. A comprehensive postmortem by the American Association for Public Opinion Research concluded that 2020 polls were the least accurate in decades, overstating Biden’s advantage by an average of 3.9 percentage points nationally and 4.3 percentage points at the state level over the final two weeks of the election. (In 2016, by contrast, national polling predicted Hillary Clinton’s popular-vote margin quite accurately.) According to The New York Times , Biden led by 10 points in Wisconsin but won it by less than 1 point; he led Michigan by 8 and won by 3; he led in Pennsylvania by 5 and won by about 1. As of this writing, Harris is up in all three states, but by less than Biden was. A 2020-size error would mean that she’s actually down—and poised to lose the Electoral College.
More generally, nationalist populism (mostly RW but some LW in Latin America - see Chavez, H.) has been on the upswing since the passing of the two-bloc geopolitical model c. 1990.
Though IMO it had been there all along, WW2 and the Cold War merely shoved it to the wings for a while, but once the threat seemed gone, it was ready to come forth to fill the ideological vaccuum.
This “thermodynamics” law has been holding us back for too long! Congress should act now to unleash the potential of cold fusion and free energy!
Wait, “free” energy? That sounds kind of socialist…
Should we assume that same 2020 error is taking place in 2024? Post Jan 6th and post-Dobbs, post several other things … I’ll just say that Trump support will have to prove to me yet a third time that it’s being under-represented.
That article is certainly a depressing bit.
FWIW it links to this as its source which is free. And lots of detail.
Bottom line though is that they don’t clearly understand why it was such a bad performance. Okay. You’ve succeeded in making me even more anxious! Thanks. 
Excellent post.
Well said — I will point this out to my Political Geography students this semester.
The most immediate and obvious example was the wars following the breakup of Yugoslavia — but, as you said, much of the resurgence of nationalism elsewhere was facilitated by this too.
I mean, this is the Straight Dope, fighting ignorance and all that stuff. I’m a big believer that understanding our world is always a good thing.
Sidebar as this reminded me…polling just came out here in Ontario, that the Conservative provincial government under Doug Ford, possibly the most corrupt administration I’ve seen in Canadian politics in my lifetime, is set to win a majority at the next election, partly thanks to there being two large left(ish)-of-centre parties and just the one conservative one.
Worldwide, I see this and despair. As for the States, anyone still throwing support behind Jill Stein or Cornell West at the expense of a Democratic vote is either dumb as a bag of hammers or a GOP plant, I’m sure.
In this case it is more a better appreciation of our not understanding! That report is very depressing in its recognition of the depths of our ignorance. Of their listed possible hypotheses several would be expected to repeat in 2024 (and would have been expected to not impact polling during 2022 midterms). Their conclusion explains the reason for the ignorance well: it is impossible to know much about those who did not respond by studying those who do.
So yeah. He could win by polling being systematically off in a presidential cycle understating Trump support even to a much smaller degree than it did in 2020.
We would need to be Harris up over 5 in PA for example to get over that possibility. Being scared is the rational place to be.
Yup, we are no where close to where we need to be.
I’m off to lunch and then every Thursday I man the local Dem HQ in my town.
This IS the country that almost passed a law that pi=3.2. Think of it! With a bigger value for Pi, cars are more efficient! They go further with each revolution! The gas savings alone would be…well, more that you can imagine!
madmonk28 linked to the Atlantic article I’ll mention.
Yeah, polling in 2020 was off, so the pollsters reweighted things close that gap you mention. Who knows if they did it right or not. The Red Wave of 2022 never materialized. Maybe we are doing better than we think vis-a-vis the polls; maybe the opposite is true.
This kos diary addresses the Atlantic article and some polling failures since 2020:
I hate to be “that guy” but my takeaway from all those articles is that we simply can’t trust the polls to be very accurate, and we have no idea which way the errors will fall. Sure, in those cases where one candidate has a huge lead in the polls we can be fairly confident, but so many results at the presidential level are just too close to call.
Also, the article specifically is about polling when Trump is on the ballot, not special elections and congressional races. But you’re right, polling is a mess and we won’t have a way of knowing how much and in what direction until after the election.
I basically agree.
I’m inclined to think that we have the upside (i.e., the polls are more likely to break our way) as we have all the positive factors on our side, and Trump has his base and a drone of fear-based messaging on his.
Further, if things are looking up in Texas and Florida, can things be all that shitty in other places? I’m trying to see the big pic, but I don’t know what will happen.
Yeah that is a horrible article. Best to read the source that the Atlantic article is based on. At least the summary and the conclusion section and maybe also the section on the state by state performances.
It is sobering and that Kos gloss over take is one only possible without reading it. It is a strenuous effort to see what they want to see.
All we can say with any confidence is that despite concerted effort to correct for Trump supporters relative non participation in polls the pollsters were systematically off both times Trump was on the ballot, worse in 2020. And the experts don’t really know why. They have dismissed multiple explanations but that doesn’t solve the problem.
Maybe they have the correction right this time. Maybe behaviors have changed. Maybe they over correct this time. Maybe. But the track record is as @madmonk28 states. Being confident that it will other than that again would be a bit silly. Hopeful that it won’t be? Sure.
So yes. Polls are nice. Leads in polls are nice. Leads in swing state polls are nice. And they may even reflect reality. We just can’t be very sure that they do.
Well said. And here’s an example of a “flip” just from the last hour, which should probably quell the “break out them cigars, boys - he’s in an unrecoverable death spiral now” talk.
Last update: 12:30 p.m., Thursday, August 29 : Although we wouldn’t advise worrying too much about the difference between a 52/48 race one way versus a 48/52 race the other way — it’s not a big difference — this wasn’t a good day for Kamala Harris in our model, as Donald Trump is the slight favorite for the first time since August 3.
There’s one big reason for that — Pennsylvania, which is the tipping-point state more than one-third of the time and where it’s been quite a while since we’ve seen a poll showing Harris leading (including two new polls today).