Here’s a free link to Josh Marshall’s members-only article on polling.
I’m glad he has the patience to write this all out because all I can personally muster is to say that the polls have gotten to be way more black magic and secret sauce, which I sometimes shorten to “ignore the polls, they’re more bullshit than science at that point, just do the work.” Because I seem to be out of patience at the moment.
But lucky for us, Josh Marshall isn’t!
I wanted to share with you a few ideas, possible insights and caveats about campaign polls. These aren’t original to me in most cases, just some general points, observations, etc.
First, herding: Herding is the phenomenon in which even professional, good-faith polling operations start grouping together in the latter stages of a campaign because you don’t want to go too far out from the consensus numbers. Right now the national top lines have been between 2 and 4 points in Harris’ favor for a couple months. If you do a poll that gets you plus 10 in either direction, you’re going to think or are liable to think something’s wrong with your numbers. Somehow you’ve just got a spoiled set of data. Maybe you don’t release that poll or maybe you look again at the numbers and decide there are too few of some demographic subset and you re-weight that and it brings the topline back close to that 2-4 range.
It’s also the case that voter choice gets more settled in the final weeks of a campaign. So maybe the voters are actually herding themselves. There are lots of possibilities. But the general point here is that there are factors which can drive even ethical and professional pollsters in this herding direction.
It goes without saying that the more fly-by-nightish operations probably do this a lot. It’s sort of assumed that a lot of the junk pollsters let their freak flags fly for most of election season but then try to bring them in line with the herd in the final weeks to protect their reputations.
Implicit in the herding discussion above is what I’ve noted several times recently: the stability of this race is one of its most defining features. That could just be how settled the public mind is on this choice. It must be some of that. But I’ve heard another possibility proposed recently: Maybe pollsters are now modeling so aggressively that they’ve wrung all the movement out of the numbers?
Remember in earlier discussions I’ve noted that a poll is based not just on the 500 to 1,000 people polled but also on the pollster’s theory of the electorate. This is the art part of the art and science of polling. Modeling has become more aggressive as non-response has become a bigger and bigger problem. So you’re not just weighting men and women and Black people and white people but affluence, education levels, city vs. rural, how people say they voted last time, perhaps party identification. At a certain point you’re basically just saying you need 49% Trump supporters and 49% Harris supporters. I jest of course. But there could be something to this: maybe the race is so stable because the variability has been squeezed out through aggressive modeling. No one is proposing this as more than an idea, a “what if.” But it’s worth having in the back of our minds.
Next there’s an article you should read by Nate Cohn of the Times. It gets at a key methodological question about this year’s polling. Historically, it was always considered bad practice to weight polls based on how voters say they voted in the last election. Basically, people don’t always remember or tell the truth about how they voted. And it tends to pump up the advantage of the party that lost the last election. In other words, by traditional standards and accepted methodological considerations, by adding this voter recall weighting the pollsters have built a pro-Trump advantage into every result. But in an effort to more effectively avoid missing hidden Trump voters, most but not all pollsters decided to weight for voter recall for 2024. This isn’t obviously a bad idea, as Cohn explains. Polls did underestimate Trump in 2016 and 2020. And there are various ways pollsters have tried to account for the reasons why traditionally this isn’t a good idea to do. Cohn’s point and one I think he’s right to make is that that decision leads to two very different stories of the election. I’ve argued here and on the podcast that there’s as good a reason to think pollsters over-adjusted after 2024 as there is to expect Trump to again outdo the polls. Depending on how the race plays out it might be this specific decision that explains why one group or another of pollsters missed the mark.
Open thread.
Update: A great list in the comments below from hrprogressive:
There are a myriad of problems that come with trying to provide “an accurate snapshot of how a state or national electorate might vote”.
1) The obvious miss of 2016 which all but coronated Hillary as President before she lost.
2) The obvious vested interest of the corporate-owned MSM in “the horserace”
3) The obvious vested interest of the pollsters commissioned by same in making sure there “is” a horserace.
4) The vested interest of the right-leaning business community/shareholders in oversampling their voting base to drive the narrative they are more popular than they really are.
5) The very real problem of having a base of “non-Trump Republicans” who aren’t full-blown cult members but for whom the idea of “voting for a Democrat” is still anathema.
6) The very real structural problems the fascist right have created by making it harder to vote, chiefly impacting the core constituency of the Democratic Party.
7) The very real concerns a lot of Millennial / Gen Z voters have that basically amounts to “what have you done for me lately?” when it comes to the Democratic Party not being nearly progressive enough for their tastes
8) The ratfucking problem of a very certain subset of terminally online leftists who are either accelerationists and/or have been disingenuous about the I/P issue.
9) A very specific subset of “undecided voters” who are, politely speaking, morons who do not understand the stakes of this election and/or what might happen to them and the people they care about if the Trump Reich comes to pass.
10) The very ugly truth that there is a subset of this country – it’s not “50%” but it’s not insignificant – who actually, truly, do want to see a Christofascist Theocratic Dictatorship and it doesn’t matter who’s at the top of the ticket, so long as they can seek out their specific agendas and aims, whether it’s eschatological, or just hypernationalistic or whatever.
11) The very real methodological problem of technology moving past how “issue polling” has even been conducted, and that the base of people willing to participate in “traditional polling” skews very old, and very conservative, and a lot of younger, more liberal opinions, just aren’t being captured, but, there hasn’t yet been a real substantive movement to address this in a way that is scientifically sound while more accurately representing the actual populace.
Baud
Liberals need to learn to get mature when it comes to (1) the media and (2) polling. IMHO.
frosty
@Baud: Does mature mean doing something else besides this:
When in danger or in doubt, run in circles, scream and shout!
Baud
@frosty:
Hard to say. It’s a catchy rhyme. I don’t know if I have anything that can compete with that.
hrprogressive
There are a myriad of problems that come with trying to provide “an accurate snapshot of how a state or national electorate might vote”.
1) The obvious miss of 2016 which all but coronated Hillary as President before she lost.
2) The obvious vested interest of the corporate-owned MSM in “the horserace”
3) The obvious vested interest of the pollsters commissioned by same in making sure there “is” a horserace.
4) The vested interest of the right-leaning business community/shareholders in oversampling their voting base to drive the narrative they are more popular than they really are.
5) The very real problem of having a base of “non-Trump Republicans” who aren’t full-blown cult members but for whom the idea of “voting for a Democrat” is still anathema.
6) The very real structural problems the fascist right have created by making it harder to vote, chiefly impacting the core constituency of the Democratic Party.
7) The very real concerns a lot of Millennial / Gen Z voters have that basically amounts to “what have you done for me lately?” when it comes to the Democratic Party not being nearly progressive enough for their tastes
8) The ratfucking problem of a very certain subset of terminally online leftists who are either accelerationists and/or have been disingenuous about the I/P issue.
9) A very specific subset of “undecided voters” who are, politely speaking, morons who do not understand the stakes of this election and/or what might happen to them and the people they care about if the Trump Reich comes to pass.
10) The very ugly truth that there is a subset of this country – it’s not “50%” but it’s not insignificant – who actually, truly, do want to see a Christofascist Theocratic Dictatorship and it doesn’t matter who’s at the top of the ticket, so long as they can seek out their specific agendas and aims, whether it’s eschatological, or just hypernationalistic or whatever.
11) The very real methodological problem of technology moving past how “issue polling” has even been conducted, and that the base of people willing to participate in “traditional polling” skews very old, and very conservative, and a lot of younger, more liberal opinions, just aren’t being captured, but, there hasn’t yet been a real substantive movement to address this in a way that is scientifically sound while more accurately representing the actual populace.
So,
TLDR; there’s a lot of real issues with any polling in this day and age, and it is neither advantageous for left-leaners to scream about “being down 10%” which is defeatist and also “omg we’re up 10% we’re gonna get 400 EV’s and a trifecta”
It’s impossible to believe either.
But it’s also impossible to feel good about the outcome.
El Cruzado
Rick Perstein had a more historical —as is his wont— view on these issues a couple weeks ago here
MattF
Also, IMO, people do not have a good sense of the behavior of random noise. We are now at a point where there are hundreds of polls— consequently many polls will have results that are many standard deviations from the true answer. It’s noise, people, and finding patterns in noise is easy and incorrect.
Omnes Omnibus
For you, perhaps.
Nancy
I’ve been reading Talking Points Memo about as long as I’ve been reading Balloon-Juice, 20 ÷ or – years. I’ve learned a lot over that time from both sites. Went from being someone who didn’t “like” politics to whatever it is I am now.
Obsessive perhaps.
Currentky waiting on a batch of postcards to maybe convince some swing state undecideds.
HumboldtBlue
Mark Robinson arrived at an even yesterday and immediately left after being encountered by these women.
kindness
I’m inundated with texts pleading for donations. At this point I’m not answering any unknown calls/texts and am just blocking any number I don’t know. Some of those may be viable polls. That I don’t get represented doesn’t bother me though. January 20th can’t come quick enough (because we all know this thing ain’t gonna be over 11/6).
zhena gogolia
@Omnes Omnibus: So what is your present mood?
TS
I’m the family worry wart – I keep thinking if the polls had MVP 10 points ahead – I would be saying they are accurate – but she is showing only marginally ahead – although I cannot believe she is at 50/50 with trump – I worry that this is accurate – or the polls are wrong in the other direction.
I was worried until Ohio (then a bell weather) went to Obama in 2008 – I could not believe he would win. I’m hoping this is my reward for all my worry about Kamala this time around.
Omnes Omnibus
@zhena gogolia: I think we are going to win.
Melancholy Jaques
@Baud:
Many liberals need to ignore polls and media and focus on the dull but necessary work of campaigning. An example often shared in this forum is the postcards. None of us need to receive them, but a lot of people do. Studies show they work. Media never mention them.
Between now and election day, the only thing that matters is making sure our people vote.
zhena gogolia
@Omnes Omnibus: You and my husband. Nobody else around me. I guess I should listen to you guys.
zhena gogolia
@Melancholy Jaques: I have 300 ready to go to NV on the appointed day.
Omnes Omnibus
@Melancholy Jaques: Yep. Worrying about the polls does fuck all good for anyone’s mental health.
Steve LaBonne
Thanks for putting that up. I Iinked to it near the end of a dead thread and I don’t know if anybody saw it. I think it makes really good points.
Melancholy Jaques
@zhena gogolia:
Because you are awesome!
Geo Wilcox
@TS: Ohio was not a big deal but Indiana sure as shit was.
WaterGirl
@HumboldtBlue: I’m a bit surprised that Robinson recognized the Handmaidens reference.
SatanicPanic
🤷♂️. Who knows. I think it’s worth paying attention to polls but they’re not all that valuable either.
Steve LaBonne
@hrprogressive: As Josh points out, the miss in 2016 was only about a point and a half, which really is about as good as it gets. 2020 was the real miss (4 points). Pandemic conditions, as Josh says, make it questionable to draw any conclusions from that election, yet that is what has motivated them to bump up the number of Trump voters in their samples as a “correction”. Really, the depth to which response rates have fallen, and the amount of kludging required to try to compensate for that, mean that “ignore the polls” is the best advice.
HumboldtBlue
Here’s an excellent breakdown of just how fucking bad the NYT has been covering politics in the past few years.
Anoniminous
The only evidence prediction of this is election is:
Based on the past two elections the result will be decided by ~80,000 votes spread across some combination of Arizona, Georgia, Michigan, Nevada, Pennsylvania, and Wisconsin.
The structural prediction of this election is:
New and black voters in Michigan and Pennsylvania should carry those states for Harris and to 260 EVs. Thus her taking any of or a combination of Arizona, Georgia, or Wisconsin puts her to or beyond 270.
Yes I’m leaving North Carolina and its 16 EVs out of the picture. Trump absolutely needs it, Harris doesn’t. If it goes D that probably Harris has a solid shot of taking all of the above and maybe – just maybe – Florida and Iowa and … Ohio??????
WaterGirl
@hrprogressive: That’s a great list!
Anoniminous
@Omnes Omnibus:
I agree. Harris has many paths to victory. Trump only has one.
ETA: and now I have to commit seppuku for agreeing with a lawyer.
Omnes Omnibus
@Anoniminous: For any any opinion you have ever expressed, you can probably find a lawyer that will agree with it.
Glyph2112
Here is the question that haunts me. When they come out with a poll that has Trump moving up at all, who are these people who have now decided to vote Trump? I mean he is who he is. In fact it seems like he is actually getting more repulsive. Nothing he is doing should add new voters. On the contrast, I see Harris trying to reach new voters and a ton of republicans are either warming up to her or are so turned off by trump that they are voting for her.
It also doesn’t help that since I donated to Harris, I get a TON of emails asking for money, and they all seem to say “We are behind in the polls and we need money”. This adds to the anxiety of where the race is.
Anoniminous
@hrprogressive:
If black participation rate in Michigan and Pennsylvania had equaled 2012 Clinton would have carried both states and been elected. My opinion is the fall off was a combination of the ‘She’s Got It In the Bag’ pre-election triumphalism and her racist attacks on Obama in 2008.
HumboldtBlue
There are many small, petty, shitty human beings in the GOP, but none smaller, pettier or shittier than Puddin’ fingers DeSantis.
Sister Golden Bear
@WaterGirl: Rule #34. Check his browser history.
Splitting Image
A big problem with polling is that traditional polls (“who are you planning to vote for”) assume a single electorate with a large number of swing voters. The questions they ask reflect this.
The current electorate is now a group of 81 million Democrats and 74 million Republicans, going by the 2020 results. The number of people who have gone from one group to the other since then might be in single digits. If Harris turns out all of Biden’s voters but Trump’s vote reverts to his 2016 levels, the result will be 81 million to 62 million, or 57%-42%. A landslide.
Conversely, if Trump maintains his 71 million votes from 2020 but Harris reverts to Clinton levels from 2016, the result will be 74 million to 65 million, or 53%-46%. This is the GOP ceiling. The real result will be somewhere between these two extremes.
So what you’re really measuring is enthusiasm. To get a picture of what the result will be, you need to be conducting two different polls to measure how likely Democrats are to approach their maximum, and how likely the Republicans are to approach theirs. A traditional-style poll, asking the traditional questions, doesn’t really answer this.
Or in other words, you used to assume that people would vote, and you were polling them to find out who they were voting for. Now you know who they will be voting for, and you are polling them to find out if they are going to vote. The independent and dependent variables have reversed.
All of this is on top of the cratering response rates, which makes it harder to conduct either type of poll.
Steve LaBonne
@Glyph2112: The anxiety can be tamed by absorbing what Josh is saying and realizing that the little blips in the polling that are worrying you are unlikely to be connected to reality. As bad as polls have become from a prediction standpoint, they are somewhat better at detecting trends, and as Josh points out the big story of this race as seen by the polls is that there hasn’t been any trend since the convention.
Another Scott
[ quietly-whistling ]
Ctrl-F “Dobbs”
Not Found.
[ /quietly-whistling ]
;-)
Forward!!
Cheers,
Scott.
bbleh
@Melancholy Jaques: exactly this. “The polls” are all SO close that, between issues of modeling and simple statistical “noise,” anybody who says there’s a clear leader just doesn’t know what they’re talking about. That’s not to say they’re “garbage”; it’s just that the race is too damn close.
So … LET’S GO MAKE IT LESS CLOSE. The end.
Sister Golden Bear
@Melancholy Jaques:
A-fucking-men. There is no magic poll, nor one weird trick. Only outwitting, outworking and outlasting Republicans.
Baud
@Omnes Omnibus:
For a fee.
Melancholy Jaques
@Omnes Omnibus:
If you can pay the retainer.
Baud
@Glyph2112:
“Should” means nothing more than projecting your values onto other people. It’s a problem Democrats have when they ignore the fact that a lot of people hate us and out all the blame on poor “messaging.”
Omnes Omnibus
@Baud: @Melancholy Jaques: In the words of Billie Eilish, “Duh.”
Citizen Alan
@WaterGirl: knowing what a fucking idiot he is, he might well have just thought they were witches and ran away in fear.
Omnes Omnibus
@Citizen Alan: Come on, man. Witches have pointy hats. Everyone knows that.
Falling Diphthong
Early in the season I saw “Polls are now a way to self-prime the betting markets” and that has really hung true.
Melancholy Jaques
@Baud:
I think all the people that hate us are committed Trump supporters or non-voters. I think the biggest factor in Harris’s favor is the people who are tired of him and his act.
SiubhanDuinne
In case it hasn’t been posted, Ethel Kennedy has died, age 96, from a stroke. RIP.
JML
@Anoniminous: the numbers of states in play for Harris is the best structural aspect for this election. The fact that the GOP candidate has so many fewer “outs” to call upon somewhat makes up for some of their other structural advantages (like the overvaluing of votes in states like WY).
NC and Georgia would be such fascinating case studies if the stakes weren’t so freakin’ high. It’s hard to get excited about delving into the minutiae of what is moving or not moving voters in those states, turnout patterns, ballot access, etc when if things go south and TFG wins the USA as we’ve known it could in fact get dismantled.
But if these states do go blue, it potentially impacts national politics in some profound ways. these are states that are growing, not shrinking. these are states with significant electoral votes. These are states that have impact on their neighbors (I’m looking at you, SC).
Sometimes flipping a state, represents a trend that can’t be turned back for at least a generation. CA went that way, going from being a GOP stronghold to no-chance in literally a decade. Can NC and/or GA be like that? And can texas get lined up the same way?
Jackie
@HumboldtBlue:
I watched that “live,” and my immediate reaction was very similar to yours.
What an asshole. 😡
M31
@HumboldtBlue: hahahaha they filmed the original Handmaid’s Tale in Durham, there used to be weird barricades set up in neighborhoods with tanks with the eye-on-a-pyramid driving around, and I know some people who got complaints that their wedding was spoiled because the Duke chapel had a huge gallows set up in front of it
lol might actually be original costumes
Omnes Omnibus
(this is the full Bluesky post, okay?)
Omnes Omnibus
@SiubhanDuinne:
I still think it was shame that killed her.
lowtechcyclist
@hrprogressive:
What miss?! I call bullshit.
After the final debate, and before Comey got into the act, the polling average gave Hillary a 6-point national lead. Was that not accurate at the time?
She won the popular vote by 2 points, even after Comey’s disclosures dominated the news for the last week and a half before the election. Is it improbable that that could have caused a 4-point swing?
Certainly the polling averages showed that steady plummet in Hillary’s lead during that painful week and a half.
On the morning of the 2016 election, according to Nate Silver, based on the polling, Trump had about a 30% chance of winning. Imagine the election’s decided by a dice roll, and a 7 or 8 gives Trump the win: that’s what Silver was saying. So his actually winning was hardly a miss.
No really, tell me how the polling was a miss.
dr. luba
I went to election training today. Since it was at 10 am on a school/work day, it was mostly retired folks. I’ve been doing this since 2008, so it was mostly just review for me, but there were a few new wrinkles:
1. same day registration is now allowed in MI
2. voting your absentee ballot at your precinct on election day is also now allowed
3. voting your absentee ballot during the week of early in person voting is also now allowed
4. write-in votes will be tabulated by the state
What I found interesting was a discussion afterwards I had with two ladies I’d never met before. We got started talking about the changes, and after a few other things, we got into Dobbs and women’s rights. The level of anger from those two about SCOTUS overturning Roe was amazing. Not just that, but women’s rights in general.
Any of those experts out there claiming that economic issues will be the major concern of voters in this election aren’t talking to women. Especially older women…..who are pissed off, and who vote.
Trump/Vance/MAGA seem to think that older women won’t care about Dobbs, since we are menopausal so it doesn’t concern us. They are wrong. We have daughters, nieces, granddaughter that we care about. And we know that Dobbs is just the start. Many of us (not me, personally) remember when contraception was inaccessible without a husband’s permission, and when women had no economic rights (owning property, signing contracts, having bank accounts or credit cards).
We won’t go back!
Jackie
@Omnes Omnibus: Yup. Bobby jr sent her to her grave early. She would have lived to 100…
zhena gogolia
@dr. luba: Good to hear.
Falling Diphthong
So I don’t do most polls anymore. But yesterday I answered a poll from the Massachusetts Dept of Health. Perfectly legitimate, about Important Stuff You Should Know About Citizens. And what I learned is:
• If the answer to “How long will this take?” is “Oh, I’ll just ask the questions fast,” that means “30-40 minutes.”
• Even with all that time, the questions were not well aligned. e.g. I had to pick either 1 year or 2 years, I couldn’t say “1-2 years ago, I think.”
What I learned is not to answer the next poll. (Which I suspect is a broad: Fewer people respond –> We’ve got to make do with those so ask them more stuff –> Those people stop picking up.)
TaMara
@TS: I sobbed when they called Ohio. I AM NOT A SOBBER. Surprised me, but that’s how tense I was…I could not do 4 more years of Republican “leadership”
HumboldtBlue
I just came across this story… and if I am being completely honest, fuck this silly old man.
trollhattan
@Jackie: She was deprived of
notvoting for her own son for president. Pity.Steve in the ATL
Received this from the State Bar of Georgia. Just one more indicator that lawyers are better than everyone else*:
*may not apply to Baud, Omnes, and me
trollhattan
@HumboldtBlue: I’ll wager the old fucker refused to get a covid vaccine. Talk about a long dragged out suicide method. “Follow Trump to an early grave. You can DO this.”
trollhattan
@Steve in the ATL: “fallen tree disputes”
Do they fight each other whilst on the ground?
OTOH that’s a solid initiative, good on them.
Omnes Omnibus
@Steve in the ATL: By this point, you should know the difference between “may” and “does.”
Falling Diphthong
@kindness: I get occasional political voter and fundraising texts. (At least now all D, whereas in the spring I got a couple of Trump ones.) My son–who recently moved from a blue state to another blue state–gets way more. Enough in aggregate to be off-putting.
I get way more unsolicited text “job offers” than he does.
catclub
Isn’t there a poll that repeatedly samples the same group? of course, then you have to wonder about a group of people who let the pollster repeatedly sample.
Sister Machine Gun of Quiet Harmony
@hrprogressive: Some of those undecided voters are non-Trump Republicans. If Harris were center left, instead of being somewhat progressive, they would have already decided to vote for her. Some are so propagandized, they would never vote Dem, but there’s more out there than you realize who voted for Biden because they thought he was more moderate
WaterGirl
@Splitting Image:
That’s another smart, take, too!
frosty
Same as in 2016. First time voters coming out of the woodwork. I worked a poll that year and saw a lot of people being asked for ID, which meant it was their first time. To have that happen in my county was a bad sign.
It’s my biggest worry this year. There’s anecdata that it’s happening again and all I can do is work to make the same thing happen on our side.
Falling Diphthong
@HumboldtBlue: One thing that’s really been bored into me the past few years is the power of conspiracies as a self-identity. They don’t respond to logic, to the dismay of the earnest people who thought facts would help. Pile up enough evidence and the person will just switch to another conspiracy.
See for example “Humans have no effect on the weather” –> “The weather is controlled by Democrats’ secret magical weather machine.”
Geminid
The Wason Center released a poll of 800 likely Virginia voters on Monday. I think there will be another one taken the week before the election.
Anyway, the poll showed Harris leading Trump 52-41, with 4% undecided and 3% saying they would vote for someone else. Harris led among Independents 49% to 34%. Senator Tim Kaine led retired Navy officer Hung Cao by 20 points.
I like to look at the Wason Center’s demographic findings. This poll found that 49% of likely Virginia voters had a high school education or less, while 51% said they had a college education or more. Sixty-nine per cent identified as White and 20% as Black or African American.
Self-identification by party was Democrat, 36%; Republican 31% and Independent 31%.
When asked to pick one of five ideological categories to identify themselves, 6% chose Very Liberal; 9% chose Liberal; 36% chose Moderate; 24% chose Conservative; and 11% chose Very Conservative..
So in a voter pool where 35% identified as Conservative or Very Conservative and only 15% said they are Liberal or Very Liberal, Harris still led Trump by 11 points
Ed. The Wason Center is affilitated with Christopher Newport University. Rachel Bitecofer used to direct their polls before she parted was with CNU in the Spring of 2020. Now Doctor Becky Bromley Trujillo does. She also researches the intersection of environmentalism and politics.
Doug R
@hrprogressive:
There’s this of course:
Russia is trying to exploit America’s divisions over the war in Gaza
Hoodie
@Baud: I don’t know how to think about that. The Trump strategy in the campaign has been pretty mystifying. The campaign seems to have become more extreme with each month, which might indicate that they’ve doubled down on a motivate the base strategy. Maybe they’re picking up a few voters who may have thought Trump wasn’t racist enough? Or does the Willie Horton play win over vacillating voters who can be scared about crime? I don’t see Trump trying to pivot to more moderate positions on any issue.
Some of this suggests they’re just trying to milk the die-hards for more cash (supposedly their small-dollar donations are down something like 40%), which would seem to go along with all the scammy promotions he’s been doing lately. This latest rumor about trying to campaign in blue states might be bullshit, but it could also be an attempt to milk a last bit of money from Trump donors in those states because they’re tapped out in the regular red-state haunts. The whole thing bears several of the hallmarks of the late stages of a con.
catclub
The item about all the things Republicans have done to make voting more difficult
is interesting because I think those also generate a backlash against those restrictions that almost or nearly completely counteracts any effect. When you can tell potential voters that someone is trying to keep them from voting, oppositional attitudes might surface.
[Of course, this depends on being able to tell them.]
I have seen very few reports of ‘this group of voters is obviously down due to restrictions’. Maybe I have not been alert enough.
dww44
I cannot add to my anxieties by even reading polls. I can only report the anecdotal evidence around me:
1. In the last 3 weeks there have been lots more Trump signs, a couple of which were add-ons ( yards already with a Trump sign) with the message “Pray for Trump“.
2. A long time neighbor put a Trump sign in his front yard. Never done that before in 40 years of living on this dead-end street.
3. The Trump signs are placed into the yards of homes which are obviously middle class and upper middle class and are interestingly near by to homes which have Harris-Walz signs. There are a lot less of the latter.
I am of mixed feelings about what I’m seeing. Are the MAGA/Trump supporters panicking or is this evidence that they aren’t going to allow this state to go Blue again, cause they they think they own the levers of power here?
As a white person, I am in a distinct minority and kinda lay low with family and friends, but am truly distressed by my inability to gently get close family to view Trump with clear eyes, particularly after Jan 6. These are people who gave us Dems a hard time in 2004 for being anti-American vis-a-vis the Iraq War.
So, I think the way I get through the next weeks is go volunteer and knock on doors. I’ve been a volunteer in the national elections and the state elections since 2008 but haven’t thus far as I’m very busy with an arts non-profit and just am no longer young, But doing something is probably better for my blood pressure than stressing out for the next 20 plus days.
catclub
@Hoodie:
The entire strategy is tearing down Harris. No positive Trump campaign.
The Trump sign I saw was ‘My dog is smarter than Harris’ Not sure that’s a winner.
trollhattan
@dww44:
Those signs should read “Prey for Trump.”
VFX Lurker
Reproductive rights are an economic issue. A big one. Maybe these “experts” don’t realize this.
VFX Lurker
I deleted my duplicate comment.
catclub
@trollhattan: If they pray ( or prey) but don’t vote, that is a win.
Lyrebird
@Geminid: Thanks Geminid, you are always bringing the key deets, and I am always learning from them.
I will still argue, though I don’t live in VA anymore, that VA is likely to have a higher percentage of identified conservatives voting against Trump, bc the concentration of proud conservatives who care about actual foreign policy is higher there than some other places. Like in (coughs) Langley and so forth.
I thank the FSM for them, especially if they follow Cassidy Hutchinson’s (sp?) lead and keep the orange menace far far away from any seat of power.
and
@Another Scott: Cheers to you too!
HumboldtBlue
Just got a series of photos of my grandnephew, and at two months old he’s a dead ringer for Tim Walz! Has his big wide grin and everything!
catclub
@Falling Diphthong: I am rarely contacted for polling. But _my_ first question is usually “How much are you paying me?”
Lyrebird
..and as you say, at least some memory recall! I hope that when I’m retired I’ll be involved with LWV or at least write a whole heck of a lot more postcards…
And rock on for working elections!
WaterGirl
@dww44: They pay people to attend his rallies. I would not put it past Trump to PAY people to put up signs.
catclub
89% of white voters in Mississippi voted against Obama in 2008. I think if only 83% vote against him, he wins. I was in that minority.
Sister Machine Gun of Quiet Harmony
@Glyph2112: I think Vance doing well enough in the debate brought some Trump supporters back home. It made him more likable to them. He was a liability, now less so.
dmbeaster
Polling has gone off the deep end for a variety of reasons. It always has a core problem with the following:
Any form of statistical analysis would look at these factors, and conclude garbage in, garbage out on polling results.
To compensate for these known problems, pollsters have “models” that allegedly informs them how the polling sample should look. If the pool of people that participated in creating polling data does not match the “model,” then they simply cook the numbers based on the model assumption. This is not a terrible practice, and is a necessity given the basic problems with polling. But it requires that the pollster have a really good sense about creating a good “model.”
There is no meaningful empirical data for creating such models. It does seem that in the past, pollsters who developed a lot of familiarity with a certain electorate could create good models. But it seems that pollsters are really struggling with this one based on modern trends in the electorate.
An example was the Field Poll in California run by Melvin Field, which I saw in action for years as I have lived here since 1969. I think that he had a better sense of the electorate than most politicians, and could read polling data with great accuracy. This was someone who in those times, could create very effective models for his polling data. The Field poll shut down two years after Melvin Field died in 2014.
Even when pollsters get it “right,” that could just be a statistical anomaly like a stopped clock.
The one thing that polls are really good for is showing trends. You may be uncertain about the bottom line numbers, but the same pollster using the same methods can probably accurately detect and report a shift in the probable behavior of the electorate. Of course, this assumes that they are not “herding” as discussed by Marshall.
Bottom line. I think that the biggest predictors of outcomes are other factors. This is a turnout election, and Harris seems on top of this, while Trump has stopped bothering.
HumboldtBlue
Sister Golden Bear
@Hoodie: Some of the campaign appearance oddities may also be Trump’s ego (and some subconscious sense that he’s going to lose), like going to Coachella and especially Madison Square Garden. For the former, one theory is that he wants to brag afterwards that he draw a larger crowd than famous music festival. (Doubtful since the festival has drawn up to 250,000 people.) For MSG, apparently Trump has long wanted to appear there.
Obviously none of this is mutually exclusive with other potential reasons, such as his handlers strategy of lowering his visibility during the final weeks of the last two elections—and especially keeping him away from swing states. There’s a reason Harris keeps encouraging undecided voter to actually attend a Trump rally.
Melancholy Jaques
@Sister Machine Gun of Quiet Harmony:
But I have been assured by every leftie that Harris center right, probably a closet Republican!
Glyph2112
@frosty: The big difference between 2016 and 2024 is that in 2016, I bet a ton of those people thought Trump was a serious person and would “pivot” to become presidential. You would have to be living under a rock to not be exposed to his BS over the last 8 years. People know who trump is, so I just can’t believe there is a multitude of undecided voters out there.
And again, I see a bunch of republicans moving to Harris, even if its just for this election but don’t see any former Biden voters moving to trump.
Maxim
@WaterGirl: I’m not sure we can safely assume that he did.
frosty
@Glyph2112: Nope. The ones in my neck of the woods wouldn’t care about “Presidential”. Think instead of the “common clay of the new West”
David 🐝KHive🐝 Koch
I don’t think attacking the polls is a good idea as it could alienate Polish voters in the Rust Belt
lowtechcyclist
@Glyph2112:
This is why I have a separate email address for all my political and charitable contributions. I get forty or fifty new emails a day at that address, and they don’t fill up my family-and-friends email box.
Captain C
@Falling Diphthong:
“I know better than all those sheeple out there, who laugh at me for my total lack of education.”
dww44
@catclub:
@catclub:
Not voting is not gonna happen with the Trump voters here. they will vote.
Captain C
@Geminid:
I initially misread this as TCFG leading Harris by that much, with the Senate race the same. I was wondering what the pollsters were on when they did this, and then read to the bottom to the bit about Rachel Bitecofer’s (previous?) involvment and decided a reread was in order.
trollhattan
@Glyph2112:
Agree. A LOT of folks who should have known better presumed Trump’s election persona was an act crafted to gather votes and did not reflect the candidate. “Now, he must pivot to the center.”
Sadly for us all, a good fraction of those opinionators have never admitted much less corrected for their grievous error, and here we are. Dude could win.
HumboldtBlue
The Case for Kamala Harris. The Atlantic’s endorsement
trollhattan
@David 🐝KHive🐝 Koch:
Walz could host a kielbasa feed.
Captain C
@Glyph2112:
I suspect there will be a small few, and their reasons will come down to misogyny and racism.
Trivia Man
Most suspicious to me – the last several days 80% of the wingnut chatter is about polks.
trump us leading!
he’s widening the lead every day!
kamala internal polls look terrible!
so much more of this compared to scary immigrant talk feels like bluster.
Kosh III
“Christofascist Theocratic Dictatorship”
Like Ted Fraking Cruz’s father? And a whole lot more.
Trivia Man
@Omnes Omnibus: for a fee many will agree with anything
trollhattan
Trump has no idea, but this would reverse part of St. Ronaldus’ elimination of the consumer interest deduction. I can see the Jarvis people having severe cognitive dissonance right now.
Kosh III
@HumboldtBlue:
There are many small, petty, shitty human beings in the GOP, but none smaller, pettier or shittier than Puddin’ fingers DeSantis.
You aren’t in Tennessee and had to endure 25 years of Marsha Airhead Blackburn.
HumboldtBlue
Falling Diphthong
@Captain C: Wise advice re engaging conspiracy theorists: If someone has made, say, flat eartherism the core of their identity, then that is really not going to be shaken by your Very Sound Argument.
Melancholy Jaques
@Captain C:
And the belief that the president determines the price of groceries.
Hoodie
@Sister Golden Bear: The blue state campaigning thing is likely just bullshit. The appearances would have too great a likelihood of ending up embarrassments (e.g., low attendance), which undermines the whole ego thing. I get the sense that his campaign relies totally on getting out the knuckle draggers and about all they have left in their quiver is to call Harris a n****r b***h. Kind of reminds me of that old Buck Henry gag on SNL where he plays a guy running a call-in show who eventually starts threatening to kill puppies when no one calls. Of course it’s pure conjecture, but I can see Harris walking away with this one if nothing really bad happens over the next few weeks. Trump may be maxed out at something like 45-46%, which means she could win by as much as 6-9 points. It’s conceivable that polls won’t pick that up because they’re based on Trump’s prior campaigns where some might have been deluded into thinking he’s really moderate and normal.
zhena gogolia
@Trivia Man: And preparation for claiming the election was stolen.
Baud
@zhena gogolia:
Hope so. That means he didn’t win legitimately.
HumboldtBlue
Yet another example of why the mainstream corporate media is now useless when it comes to political coverage.
lowtechcyclist
@HumboldtBlue:
The nerve of Evangelicals for Harris, quoting Billy Graham accurately! No wonder Franklin Graham has his panties in a twist!
Matt McIrvin
@Falling Diphthong: Like I said in another thread, I don’t really think I can convince any of those people, but I wonder if it’s possible to intervene at an earlier stage and keep people from partaking of the conspiracy drug. There are always normies reading. Some of them will be kind of uninformed about the subject matter and some conspiracy theorists are well-spoken, and can make them wonder if there’s anything to it. And once they get in, they sample those powerful emotions, maybe they get in deeper.
I don’t know, this is sounding like old War on Drugs talk but it does seem like people collect more and more conspiracies once they really get into it.
The modern right has made this their brand, but this is not just on the right. Sometimes I argue with people here who have fallen into what I think are conspiracy theories on our side. Hell, my daughter’s called some of my more unhinged political ranting “conspiracy theory” and maybe she’s right (I tend to express florid fears about what people might do in the future rather than claim they are doing them now, and keeping a lid on that is hard–our right tends to gloat like Bond villains about the terrible things they’re planning when they get the chance, so often it’s just repeating the worst items on their wish list).
Baud
@lowtechcyclist:
Love they neighbor.©
Baud
@Matt McIrvin:
As I said in that other thread, it’s a human failing, not a right wing failing.
What makes the right wing special is how they’ve built a political movement based on a reinforcing and amplifying human failings.
WaterGirl
@Maxim: Just saw a bunch of scary women and bailed?
Amy!
What about push polling? In the global trend of enshittification of everything, my experience has been that pollsters, within the past [insert random number] years, have increasingly adopted slanted language, exclusion of topics/issues and promotion of other topics/issues, with slanted or carefully constructed multiple choice answers. They’re fouling the results deliberately. My training was as an historian (including research in archives), and have a near-religious reverence for data collection. But some time in the last year or so, I just stopped answering polls (and I’m rural, with a landline, and answer my phone (still) and sometimes also respond to texts from strangers (not young, in short)). Part of the reason for this is that several of the ones that I responded to last were infuriating, one to the point that I started declining to answer and the (actual human) pollster told me she would just pick a response (I hung up).
It’s possible that I’m unique or at least unusual in perceiving a decline in the quality of the questions, and an increase it open partisanship, but perhaps others who have been willing to answer polls in the past but aren’t willing any longer (if any), could chime in?
TBone
@HumboldtBlue: thanks for sharing that great read 👍 and notice of virtual event.
Dangerman
Roughly 99%, give or take a point either way, of polls available to us through the media are shit.
Fortunately, the campaigns have the best pollsters money can buy (TCFG’s pollsters make sure the check clears first). Trump is acting like he is losing to me and he’s going bonkers; losing means he has to gin up weeks and weeks of lawsuits to try and overturn the election (and pay more fucking lawyers; shit, couldn’t Ivanka have at least had the courtesy of going to law school?), not to mention, he has to fight to keep his rotund ass out of prison.
HumboldtBlue
@TBone:
Huffman is my rep, I’ve talked with and interviewed him dozens of times. Great rep, great guy.
Steve LaBonne
The only thing we really know is that this election will be very close, something we knew 3 1/2 years ago- it’s a structural feature of our current political alignment. There is no additional information that is reliable and actionable- it was always going to come down to turning out our voters.
FelonyGovt
@Amy!: I haven’t answered a poll in years, but the last few times I did I similarly became infuriated by the slant of the questions. It was like a survey from a car dealer (how would you rate us- fantastic? or super fantastic?)
Geminid
@Captain C: Wason Center polling seems pretty solid. I thought it was worth mentioning Bitecofer’s prior role because she has a solid base of social science knowledge and can design and direct polls.
After she left CNU, Bitecofer transitioned from the academia to the professional campaign industry. I’m not sure she is working for anyone this cycle but after his successful 2022 campaign, Arizona Secretary of State Adrian Fontes spoke highly of Bitecofer’s assistance. She was working freelance out of Salem, Oregon and signed on with the Arizona consulting group Fontes hired.
dr. luba
@Lyrebird: Thanks! I became more active after the 2000 election, in which I voted absentee, and then spent election day and a few weeks after in the Australian outback, with only one radio station and no idea wha the hell was going on in the USA.
Over the next few elections, I worked election protection, GOTV (with a bunch of UAW workers–election day is a paid holiday for them), phone banking, etc. In 2008 it was suggested that we “younger” folks sign up to work as election inspectors and help out the older folks who were getting overwhelmed….. I took election days off and worked every election until 2020. I worked the primary that year; for the general, we were not allowed to require voters wear masks. Not worth risking my life in pre-vaccine times to help MAGAs vote.
Anyways, I’m one of the older folks now. I’ve even been assigned to the same precinct 3 elections in a row…..finally settling down. I’m learning to do the electronic poll book this year….as a back up person. I’m usually at station 1, being the greeter (due to my sunny disposition, I assume): checking IDs, helping people fill out their applications to vote, and directing lost and confused voters.
Working at the first station means I can help make sure everyone who wants to vote can figure out a way to do so. I am well versed in the nuances of voting rights. This year we won’t need provisional ballots, I am told–with same day voter registration, they should become a thing of the past.
NB: Michigan has gone from being one of the most gerrymandered states in the union, ruled by the GOP with a minority of actual votes cast, to a leader on voting rights. It began with a ballot initiative in 2018, another in 2022, and then more reforms passed by a blue legislature in 2023.
dr. luba
@FelonyGovt: It seems like all polling has become push-polling, no?
TBone
@HumboldtBlue: 💙 I just looked him up and he had me at:
SW
In 16 the fact that Trump was essentially an unknown entity for much of the population worked in his favor. Many wanted to try something different. This accounted for his outperforming the polls. After four years as President and four years as red state president in exile there are no voters who haven’t formed an opinion about him.
K-Mo
Good material in the OP and throughout the discussion. My personal take is that you can’t get too focused on the fine details- it will drive you crazy. Some results seem broadly reliable:
1) it’s very close (duh). Pretty much all of the swing state polls are essentially tied.
2) most voters are locked in. They are reliable D’s or reliable Rs. They may fall into the various categories described by hrprogressive but at the end of the day they are Rs and Ds.
3) There *has* been some shifting along the margins. Basically, Trump has peeled off a bunch of low-education folks that have voted D in the past but are disgruntled, and Harris has picked up a bunch of high-education folks who previously voted R but have realized that the things they used to like about their party are no longer true.
4) Because of #3, the electoral college will be closer to the national vote.
5) Women are fired up for Kamala. Men are mostly idiots.
6) There remains a significant group that either won’t answer polls, is torn about who to vote for, or is wishy washy on the whole concept of voting. IMHO these people are going on vibes and will do so in November. What makes me optimistic is that Kamala has intrinsically terrific vibes and Trump is despicable. In the meantime let’s do everything we can to win them over and get them to the polls.
catclub
@SW:
As I say repeatedly, Mark Burnett has a lot to answer for. If you asked a random person who was the ‘billionaire businessman’ on The Apprentice, too many would know. And that was how unknown Trump was.
Melancholy Jaques
@HumboldtBlue:
Likely going to produce a Streisand Effect. I didn’t hear of the ad until I heard of Franklin Graham threatening to sue.
catclub
That would be great.
catclub
Rotating tag for the blog?
Captain C
@Geminid:
Oh, I’m sure they are; I just misread the Harris-TCFG numbers as the opposite at first, which completely didn’t jibe with Kaine having a 20 point lead.
Matt McIrvin
@Baud: And just to be clear, a conspiracy theory in the sense that we’re talking about here is not just a theory that there is a conspiracy. I mean, Jack Smith’s superseding indictment of Donald Trump outlined excellent evidence for an actual conspiracy–that is real.
The distinction is in the evidentiary reasoning. Are you claiming there’s a conspiracy because you have real, solid evidence that this is going on? Or is it just because you wouldn’t put it past them, and believing puts you on the right team? The ones that are freed from the need for good evidence are often far grander, because why not, but that’s not the distinction.
I spend too much time thinking about these–the ones I find fascinating are the ones that hinge on bad ideas about science, and especially the ones that I would personally have had to be in on. Often a central feature is “the world is actually simpler than you think, and you don’t have to study hard to understand it”. I remember a whole book once that explained that the reason general relativity was difficult to learn was that physicists were intentionally obfuscating it to keep people from building machines to extract free “spacic energy”.
Glyph2112
@Dangerman: I agree that Trump knows he is in trouble. Thats why all of a sudden he is promising tax cuts and no taxes on OT, Tips, or auto loans. Also trying to back pedal on the whole abortion thing. The scary brown people only works on his base. His legal fuckery only works if the election is close.
Spanky
@catclub: It ain’t wrong.
Omnes Omnibus
@Trivia Man: Make an argument for you, sure. Not necessarily agree with you. There is a difference.
Geminid
@Captain C: Biden won Virginia by 10 points in 2020. That was the biggest winning margin for a Democrat since “F”LBJ in 1964. Aside from Youngkin’s win in the 2021 Governor race, Virginia has trended blue throughout this century and I won’t be surprised if Harris exceeds Biden’s 10 point margin.
EngineerScotty
@trollhattan:
What happens when your tree falls on your neighbor’s house, is actually a big issue for property owners and their lawyers.
Often times it comes down to the health of the tree before it fell.
Was the tree diseased, infested with termites, etc–something that should have been removed prior to something tipping it over? Then the property owner may be liable for damage to a neighbor’s property if it falls.
Otherwise, treefalls are generally considered an “act of God”, especially if knocked down by a strong windstorm, an ice storm, or other unusual weather event, and everyone is responsible for their own damages.
Six and seven figure lawsuits that hinge upon the health of a fallen tree can be a big deal.
Gretchen
I keep wonder how they’re doing this polling they’re so certain about. I’m in my 70s and never answer a call from a number I don’t know – most of them don’t even ring through. I can’t imagine that young people answer at a higher rate than I do. Is it just people in their 80s with landlines? Are they taking that one young black person’s answer and multiplying it to weight the poll?
EngineerScotty
A few weeks ago, folks with the Trump campaign were talking up their voting app, and their strategy of reaching out to yet more unactivated but Trump-leaning voters that were alleged to exist. (Actually, this might have been various longstanding GOP operatives being concerned about the lack of a traditional GOTV operation–or at least concerned about the fact that they hadn’t been hired; it’s not just the Tom Penns and David Alexrods on our side who like to occasionally throw shade at campaigns that don’t use them).
And in theory–it could work; 2020 was a high turnout election, and still 1/3 of eligible voters didn’t vote. Some consider it beneath them, some don’t like any of the choices on offer, and some don’t care about politics at all.
If you COULD activate a significant share of these, it would be a big advantage. Of course, these voters generally don’t vote for a reason. But a lot of GOP outreach over the past cycle and a half has been geared towards engaging this sort of voter–especially the “dumb as a box of hammers” variety. The whole Q nonsense seems ridiculous to us, but there are probably some voters who don’t care about things like global warming or abortion rights (in either direction) or the proper marginal tax rates, but do care about pedophile rings and other such stuff. But still–trying to do so is a high-risk play, especially if it replaces rather than augments traditional GOTV.
But here’s the thing–generally, voters who don’t usually vote won’t be registered to do so, and if they do decide to vote this November, in most states they will have to register. (In a few states, you only need prove you are a citizen and resident; there’s no registration step prior to being able to vote).
So–there are tremendous registration numbers being reported on our side, though possibly from partisan sources. Some partisan GOP sources claim that the GOP is turning out high numbers as well.
Do voter registration numbers suggest that the GOP is activating new voters in large numbers, larger than the usual turnover as kids become adults and old folks die? Or is this pie in the sky?
This happening, and/or large numbers of young men switching allegiance because they don’t want to vote for a woman, seem to be the only chances that Trump has.
Falling Diphthong
@Matt McIrvin: People collect more and more conspiracies once they really get into it.
This is an actual thing: like potato chips, it’s rare to have just one. Sometimes the conspiracies even contradict each other, and that just means you have all your bases covered.
I remember seeing research into why people believe conspiracies, and it was four things:
• When they learned the thing, they also learned the conspiracy to explain it.
• They’re afraid of something, and believing a conspiracy about it makes them feel in control.
• It’s an important group identifier.
• They feel like they have unearthed secret information that the regular sheeple are too dumb to understand.
The first might be amenable to new facts, especially if it’s not core to their identity–say they were raised by flat Earthers and never questioned it, but they haven’t spent any time memorizing the arguments for why the edge of the Earth isn’t solidly coated in luxury real estate for the view. The second might be reachable, especially if you have an argument about why they don’t need to be afraid. The last two groups have a ton of self-image invested in their conspiracies.
Ceci n est pas mon nym
In the science world, that is the time-honored principle called “changing the data to fit the theory”.
EM
Hey, so I used to work as a BDR, you know, one of those people making cold calls all day. And let me tell you, it was rough. I’m talking 50 to 80 calls a day, all cell numbers, and I’d be doing a happy dance if even 4 people picked up. Most days? It was like shouting into the void. I mean, there were times when I’d make 80 calls and get ONE person to answer. ONE!
So here’s what’s bugging me: How on earth are political polls still a thing? Like, seriously, how do they even work? If 98% of us are dodging calls from numbers we don’t know (and let’s be real, we all do it), how are pollsters getting anyone to talk to them? And it’s not just about answering – it’s getting people to actually take the survey, making sure they’re talking to a mix of folks from different backgrounds, areas, you name it.
It just doesn’t add up. Are these pollsters wizards? Do they have some secret sauce for getting people on the phone? Or are we all just kidding ourselves about how accurate these polls really are? I don’t know about you, but it makes me wonder if we need to totally rethink how we do this stuff in a world where nobody wants to pick up the phone anymore.
pabadger
@Omnes Omnibus: if she wins I will feel great. I like to celebrate success.
SW
@catclub: true but what they “knew” was made up by a TV show. It was primarily the people of NY who really knew him as a local buffoon for decades.
wenchacha
@WaterGirl: There’s probably a porn version or ten.
WaterGirl
@wenchacha: That explains it, then!
SuzieC
@Lyrebird: Older women remember marching in the streets for abortion rights pre-Roe. I did. We chanted “Women must control our fate. Not the church, not the state!” All of my female friends in our age cohort (70 to 75) remember and we’re mad as hell and we won’t go back.
Another Scott
@EM: 98% dodging? 2% answering??
That’s way, way too high.
Nate Cohn at FTFNYT in October 2022:
There’s no reason to believe the response rate has increased since then.
HTH!
Cheers,
Scott.