The NYT/Siena poll is far from perfect - and in general I worry more about people expecting it to be perfect than I worry about it getting insufficient credit.
But the poll is very different from the others - and the differences help explain its record

https://x.com/NateSilver538/status/1833495962578465266

Staying on the LV question: from its inception until last week's CNN/SSRS state polls (give them applause), the NYT/Siena poll was the only public poll incorporating self-reported vote intention and a model of turnout based no vote history

FiveThirtyEight released a new model today, showing Harris with a 58% chance to win. But it's clearly a very different model and I think it's important to hear more about the differences The previous model made Biden the favorite bc it gave 4:1 weight to fundamentals > polls -- a view that would make Trump stronger today and perhaps still ahead.
Now it gives 4:1 weight to polls > fundamentals -- which would have made Trump a large favorite before

Kamala Harris puts the Sun Belt back in play, with the race tied across AZ, NC, NV, GA
AZ: Harris 50, Trump 45
GA: Trump 50, Harris 46
NV: Trump 48, Harris 47
NC: Harris 49, Trump 47
nytimes.com/2024/08/17/us/… The poll basically shows the race returning to "normal," with Harris obtaining far larger margins among Black, Hispanic and young voters, propelling her to nearly a 10 pt gain across the four states
nytimes.com/2024/08/17/ups…

Response rates by party looked fundamentally normal, including in Pennsylvania where we have the best partisanship data out there

https://x.com/Wertwhile/status/1822225193638535599

That said, it is worth noting that the sample does look a little blue. It's about a net-1 pt more Democratic, Dem-leaning and 2020 'Biden compared to the last time we polled these three states in May. This could be an indication of a shift in response patterns, though it could indicate change in attitude as well

Kamala Harris leads Donald Trump, 50-46 among likely voters, in each of Pennsylvania, Wisconsin and Michigan in the first Times/Siena state polls since she became her party's nominee
nytimes.com/2024/08/10/us/… Sometimes, it's hard to explain why things move around. This is time, it's easy: after three extraordinary weeks, Harris is popular.
Her favs are >49 in each state; she does well on every trait we measure.
Trump's favorables are up too -- highest level ever in NYT/Siena polling in these three states. Not enough against a popular opponent.
nytimes.com/2024/08/10/ups…

Trump 48, Harris 47 among likely voters nationwide in our first Times/Siena national poll since Biden left the race
nytimes.com/2024/07/25/us/… It's an ordinary result, but there are a lot of shifts beneath the surface.
-- Trump favorability surges to 48%, up 6 points since post-debate and his highest ever in a Times/Siena poll
-- Harris favorability surges to 46%, up 10 pts since February
nytimes.com/2024/07/25/ups…

Not to be an assignment editor, but at this point I could use a thorough analysis and explanation of the findings of the model, not just a description of the methodology

https://x.com/gelliottmorris/status/1812505261950587082

The findings of the model do not intuitively follow from the description of its methodology. The QT is the latest example: I would have expected an adjustment based on data in other states to hurt Biden in WI, as WI has yielded some of Biden's best numbers this year.

Kamala Harris ran ahead of Joe Biden in New York Times/Siena College polls of Pennsylvania and Virginia, taken before the shooting on Saturday.
Biden v. Trump
PA: Trump+3, 48-45;
VA: Biden+3, 48-45.

Harris v. Trump
PA: Trump+1, 48-47
VA: Harris+5, 49-44
nytimes.com/2024/07/15/ups…
This is not the first time Harris has run ahead of Biden in our polling. She also ran ahead of Biden in Times/Siena battleground state polls last November. In each case, she outruns Biden among young and nonwhite voters

Perhaps the most important poll you've never heard of came out today: the Pew NPORS study, a large mail survey with financial incentives and a 30% response rate. It's important enough that I had to open it when the email arrived.
pewresearch.org/methods/fact-s… The survey is important because it's used as a 'benchmark' -- its results are used as targets for weighting by other polls. Pew uses it to weight their usual surveys, and other polls (like CNN/SSRS, KFF, Ipsos) do too.
(We don't use it, but I do compare it to NYT/Siena data)

Donald Trump leads Joe Biden by 6 points among likely voters nationwide in a New York Times/Siena College poll taken after the debate. He leads by 9 points among registered voters.
nytimes.com/2024/07/03/ups… This is a 3 point shift from the Times/Siena poll taken pre-debate -- that's almost exactly the average shift toward the winner of a presidential debate over the last few decades.
The bigger shift in the polling is the one that's been unfolding slowly for four years

Previous respondents to recent NYT/Siena surveys shift 2 points toward Biden in post-verdict re-interviews nytimes.com/2024/06/05/ups… A recontacting study involves a trade off: it's less representative, but it confirms shifts at the individual-level.
The folks we recontacted were relatively engaged, educated, whiter, and older than the electorate as a whole -- which may underestimate the swing to Biden

I won't be able to go to @AAPOR (a polling conference) this year, so I want to mention here what I would have liked to have discussed had if I attended:
Vote history. Pollsters have long known that vote history is a strong predictor of propensity to respond to surveys. What's new (at least in Times/Siena data this year) is that it's now an extremely powerful predictor of vote choice, controlling for other variables used in weighting

One thing I've been experimenting with since our GOP oversample this summer: weighting our polls by each partisanship subgroup, which has the consequence of ensuring that each subgroup used in weighting has the right number of Dems and Republicans The main downside is that our estimates for self-reported education by voter file party are modeled, and that's something I've had pause about. I'm gradually getting more comfortable with it, as the party x edu tallies for the typically weighted sample seem consistent with subgroup

I happened to be looking a lot at Pew data last few weeks, even before this most recent partisanship study, so I wanted to share a few interesting observations about trends I noticed in their data
pewresearch.org/politics/2024/… One thing I noticed: subtle but persistent, multi-year differences between the partisan splits by demographic on the Pew ATP -- the mostly mail-to-web panel they use for this study -- and the Pew NPORS study (the one-off high-incentie mail survey they use for weighting the ATP

*Tosses meat into cage* nytimes.com/2024/02/14/ups… A few outtakes:
-- By our (rough and preliminary) estimates, this looks to be yet another zero-persuasion (off Biden '20) special. We'll have to see final vote history, but at least in Nassau it looks just as we'd expect given the party reg turnout

lol well did I get replies to this!
A few notes on special elections / clarifications

https://x.com/Nate_Cohn/status/1747441818612945200?s=20

1) Specials are driven by turnout. The data is unequivocal as long as I've been looking at them with our rich data, going back to 2017. That should not be remotely surprising, as the as the people who know about/vote in specials are highly parstisan -- just like all of you!

We have a new NYT/Siena national survey, and it's an interesting one -- with the public sympathetic toward Israel but disapproving of Biden on the issue and split on whether Israel should continue military operations.
It also has interesting '24 numbers...
nytimes.com/2023/12/19/ups… Trump leads 46-44 among RVs, but *Biden* actually leads in our first measure of the likely electorate nationwide, 47-45.
The split is driven by a huge gap in vote choice by turnout history: Biden+6 among '20 voters; Trump+22 among 2020 nonvoters
nytimes.com/2023/12/19/ups…

One thing I've seen over the last few days: a lot of people asserting that, in a variety of different ways, pre-election polls aren't very useful for demographic subgroups
I have to completely disagree. Stepping back, it's my long-standing view that the pre-election polls are the best basis for post-election estimates -- and, in particular, better than exits.
For ex, all these estimates were based on pre-election polling:
nytimes.com/2016/06/10/ups…

One year out, Trump leads Biden.
The poll finds Biden is pretty badly damaged, but that there's still a path for Democrats to win.
Whether Biden can capitalize is the big question
nytimes.com/2023/11/05/ups…
Since the last election, Biden lost the advantages that let him beat Trump. Simply put, he's no longer the broadly appealing, well-liked candidate with an advantage on favorability and so on who beat Trump four years ago

Trump 54, DeSantis 17 in the first NYT/Siena poll of the cycle. No one else is over 3%
nytimes.com/2023/07/31/us/… The poll amounts to a fairly comprehensive takedown of the DeSantis campaign theory of the race, with Trump enjoying strong advantages on DeSantis' main arguments (electability, getting things done), while his issues (woke) seem to lack much punch with conservative voters

nytimes.com/2023/06/01/ups… Quick outtake for the scatterplot lovers. Black is 2012; red 2016; blue is 2020. Data uniformly adjusted to match election result