The point of my ¯\_(ツ)_/¯ all day isn't to be nihilistic. My view is this: You have the polls. Then you have a bunch of reasons the polls might be too bullish on Ds, then a bunch of reasons why they might be too bearish on Ds. 1/

Oct 6, 2026 · 8:54 PM UTC

27
16
248
237,538
There will probably be a poll error in one direction or another, and so one of these groups will be correct. And it will be received wisdom that it was obvious all along that their interpretation was the correct one. 2/
3
7
170
28,261
My basic stance: Take the polls, and then pay attention to these arguments to get a sense of why and how large the poll error might occur in either way. FWIW I do think Senate polling is too hot for Ds given the generic balloting, but that doesn't tell us which is right. 3/
6
6
142
27,952
I also think there are structural forces that make it tough to get beyond D+8 or 9, so I suppose I have a bias toward a D lean in the polls right now, but who knows. Structural forces do collapse.
13
4
128
23,955
Sort replies: Relevant Recent Liked
Replying to @SeanTrende
The polls ar predicting ahistorical results. These same polls were absolutely atrocious in the primaries. You figure it out.
2
21
4,664
Replying to @SeanTrende
This is not random error. The university credentialed are oversampled, the working class undersampled and a substantial number of Republicans simply refuse to participate.
7
10
4,379
Replying to @SeanTrende
Guy that believes that the errors will basically cancel each other out and you’ll get a healthy D+9.5-10.5 margin nationally like most polls suggest
1
29
6,572
Replying to @SeanTrende
"then a bunch of reasons why they might be too bearish on Ds" Such as? When's the last midterm or GE where polls overstated R performance?
4
7
1,533
Replying to @SeanTrende
The Blue Wave will fizzle. The hypothesis that polls aren't reaching rurals has merit. Rural folk are going bankrupt, and they're losing their farms, but they'll crawl over broken glass to vote Red so long as Trump keeps working to Make America White Again. Believe it.
1
3
1,597
Replying to @SeanTrende
I don’t see many reasons why they might be too bearish for Dems lol It sounds like pollsters have leaned into the fact that they don’t suffer consequences for repeatedly being wrong and wildly biased for Dems Mostly because aggregators shrug and “throw it in the average”
1
1
749
Replying to @SeanTrende
What are reasons they are too bearish on Ds?
1
1
3,123
Replying to @SeanTrende
You could be right, but the Ohio YouGov poll has unweighted recall sample of Kamala +14.5%
1
4
1,477
Replying to @SeanTrende
How many rural Texans do you really think have participated in a poll ??
3
2,567
Replying to @SeanTrende
@grok when was the last time a poll or even the avg of polls missed on a single senate race in the direction of the Republican? That is to say the R candidate polled higher than the Dem and lost the election ? By say 5 pts
1
4
Replying to @SeanTrende
“Too bearish on Dems” they’re releasing +8 slop polls in Kansas and Iowa. Stop this, you’re way too smart to act this stupid.
6
1,063
Replying to @SeanTrende
Stop being nuanced and reasonable and just tell me the polls say exactly what I want to have happen.
19
2,697
Replying to @SeanTrende
Trump has been running as a rough and tumble asshole businessman who can fix the economy for 10+ years. Then he kneecapped the economy with an unneeded war. He's never been in this position.
10
899
Replying to @SeanTrende
Polls are bullshit. They don’t capture the actual sample anymore. DO NOT TRUST POLLS!!!
1
423
Replying to @SeanTrende
1. Generic D candidates ran +dd ahead of Trump in 2025/6 special elections (mostly before Iran war started) 2. Trump polling 6ppts below where he was in 2018 3. Trump 2 has been a fiasco for farm states (input costs up bc of Iran war, exports counter tariffed) Biblical disaster incoming for GOP
11
3,621
Replying to @SeanTrende
I think some Democratic supporters are counting every lead, no matter how small, as a win. Get out the vote efforts may get some candidates over the line. The big unknown is who the undecideds have voted for in the past and whether when the moment comes, will they move?
659
Replying to @SeanTrende
Sometimes, the polls are pretty close and especially the state polls.
146
Replying to @SeanTrende
Basically, we’ve reached the point where there is still a month until the election and not a lot new to say before we see actual outcomes
2
1,940
Replying to @SeanTrende
IPEV is up 250% in Cuyahoga County. That's a empirical observation and not a normative one.
325
Replying to @SeanTrende
I disagree the polls are too bearish on Dems. Literally everything has fallen in line for them.
723
Replying to @SeanTrende
So when should we determine a poll error to be a poll error? Collins in 2020 is a classic poll error. Everyone showed her losing handily. But if a race shows a 0-3 point lead consistently between two candidates (maybe one leads in 4 polls and the other 2) it’s a MoE result
1
671
Replying to @SeanTrende
Wait — are you seriously insinuating that a pre-game Pointspread doesn’t perfectly predict what the final score will ultimately be!?!? But in spite of that, it’s still a useful indicator of who’s likely to win, and by how much?? Madness!!!!
2
3,342
Replying to @SeanTrende
Yea cause you don’t know shit. That’s why you post this cope.
3
1,549