Thoughts on some narratives surrounding current polling.

[SnowLabrador]

Many of my fellow Democrats are in denial. They think that current polling is either inaccurate or is bound to improve for Biden over the next few months. Here are the narratives they come up with, and here's why I think they're wrong:

"People don't know Trump is the nominee yet" - That doesn't mean they can't have an opinion of him. Polls ask people who they'd vote for in a hypothetical matchup assuming they are the nominees. It's clear to see that Biden has gotten considerably more unpopular since 2020, and Trump significantly more popular by virtue of Biden being toxic.

"What about special elections?" - These are low-turnout affairs without much at stake. Nobody but us follows them, and the turnout dynamics are completely different from the quadrennial elections.

"Polls are overestimating Trump with minorities" - That's what they said before 2020, when Biden was supposed to win by a landslide. He barely squeaked by, in large part because BIPOC voters swung toward Trump from 2016. Look at Miami-Dade County, which continued zooming rightward.

"But downballot Democrats are doing okay! There won't be that much ticket-splitting!" - Things are always true until they're not. I'm 23 years old, and when I was born in 2000 it was likely unthinkable that polarization would one day be as virulent as it is today. It's not too crazy to suggest that things might change soon, just like they have changed so drastically in my 23 years on this planet.

"But Trump's criminal charges!" - Anyone who's been paying attention knows Trump is a criminal , and has known so since prior to the 2020 election. People either don't care or are already voting for Biden. While some people point out that polling suggests Trump's numbers will tank if he is convicted, I would remind everyone that with all of Trump's trials being delayed by SCOTUS, it looks increasingly improbable that there will be a verdict before the election, and an acquittal (which could happen!) would likely cause his numbers to shoot up.

The cold, hard truth is this:

Democrats have three options. They can either push Biden to get his act together and stop alienating the voters he needs, push him to drop out and install someone else as the nominee, or accept that Trump is going to win. Given Project 2025, the latter should not be considered an acceptable option.

[It's Time.]

Turnout is gonna tank this year, so I would expect ticket splitting to go up slightly. Low-propensity voters who vote straight-ticket are going to stay home.

[heatcharger]

Thank you. Hopefully you aren’t censored for “dooming”.

[Tekken_Guy]

Polls like Quinnipiac and Emerson are showing clear movement toward Biden from his polling nadir last fall. NBC is an outlier in that regard.

[ProgressiveModerate]

I agree with some of these points, but one thing you don’t respond to is the midterm election results. In 2018, we saw the early signs of Hispanics shifting right for instance as Dems underperformed Clinton in RGV and Miami-Dade. We also saw the sign of leftwards suburban shifts.

However, in 2022 we didn’t really see minorities at large shift right, more so just bad nonwhite turnout. I think bad nonwhite turnout is definitely something that could hurt Dems in 2024, but polling is showing a significant swing right there just weren’t signs of in 2022. Sure you can make the argument low propensity nonwhites who didn’t show up in the midterms are disproportionately more likely to back Trump but that becomes a stretch, especially since even if that were true those voters might not show up in 2024 either.

Also polling always struggles with black voters; in 2020 there were many many polls showing Trump getting 20 or 30% of the black vote which obviously never materialized.

[President Johnson]

There are polls of higher and lower quality. But even the good ones are just a snapshot in time and over last few cycles with an incumbent have always underestimated the sitting president by around three, four points in the battleground states this far out. That doesn't mean it's going to be the case this time, but I prefer to go with fundamentals this far out, compared to national polls that are between Trump +8 and Biden +6.

If polls are the same in September or October as they are now, I'd be much more worried.

[DrScholl]

Many of my fellow Democrats are in denial. They think that current polling is either inaccurate or is bound to improve for Biden over the next few months. Here are the narratives they come up with, and here's why I think they're wrong:

"People don't know Trump is the nominee yet" - That doesn't mean they can't have an opinion of him. Polls ask people who they'd vote for in a hypothetical matchup assuming they are the nominees. It's clear to see that Biden has gotten considerably more unpopular since 2020, and Trump significantly more popular by virtue of Biden being toxic.

"What about special elections?" - These are low-turnout affairs without much at stake. Nobody but us follows them, and the turnout dynamics are completely different from the quadrennial elections.

"Polls are overestimating Trump with minorities" - That's what they said before 2020, when Biden was supposed to win by a landslide. He barely squeaked by, in large part because BIPOC voters swung toward Trump from 2016. Look at Miami-Dade County, which continued zooming rightward.

"But downballot Democrats are doing okay! There won't be that much ticket-splitting!" - Things are always true until they're not. I'm 23 years old, and when I was born in 2000 it was likely unthinkable that polarization would one day be as virulent as it is today. It's not too crazy to suggest that things might change soon, just like they have changed so drastically in my 23 years on this planet.

"But Trump's criminal charges!" - Anyone who's been paying attention knows Trump is a criminal , and has known so since prior to the 2020 election. People either don't care or are already voting for Biden. While some people point out that polling suggests Trump's numbers will tank if he is convicted, I would remind everyone that with all of Trump's trials being delayed by SCOTUS, it looks increasingly improbable that there will be a verdict before the election, and an acquittal (which could happen!) would likely cause his numbers to shoot up.

The cold, hard truth is this:

Democrats have three options. They can either push Biden to get his act together and stop alienating the voters he needs, push him to drop out and install someone else as the nominee, or accept that Trump is going to win. Given Project 2025, the latter should not be considered an acceptable option.

1. There was like one poll that showed only 55% of Democrats believing this. It hasn't been a huge narrative at all, but on the margins it could account for response bias.

2. Democrats had a history of performing poorly in special elections particularly because of low turnout but there has been a big correction of that.

3. We all know that the Cuban vote in Miami-Dade County leans heavily Republican and it's swing in 2016 was probably a fluke. Other minority groups saw less swing.

4. Once again, Democrats have struggled down ballot in off or midterm years, but proved that wrong in 2022. The President tends to drag the downballot down, but somehow that didn't happen and that calls into question whether or not any discontent with Biden translates into enough low turnout this year re-elect Trump. I wouldn't count on it.

5. Prior to 2020 Trump didn't have actual charges filed against him. It's a huge anchor for any candidate to have charges against them and the delays only cause a situation similar to what Clinton in 2016 faced, the innuendo and did/could/would narrative.

In short I wouldn't bet on a Democrat with that much baggage winning even if the polls did say so despite those polls being a positive for my party.

[Mr. Third-Wayist]

Many of my fellow Democrats are in denial. They think that current polling is either inaccurate or is bound to improve for Biden over the next few months. Here are the narratives they come up with, and here's why I think they're wrong:

"People don't know Trump is the nominee yet" - That doesn't mean they can't have an opinion of him. Polls ask people who they'd vote for in a hypothetical matchup assuming they are the nominees. It's clear to see that Biden has gotten considerably more unpopular since 2020, and Trump significantly more popular by virtue of Biden being toxic.

"What about special elections?" - These are low-turnout affairs without much at stake. Nobody but us follows them, and the turnout dynamics are completely different from the quadrennial elections.

"Polls are overestimating Trump with minorities" - That's what they said before 2020, when Biden was supposed to win by a landslide. He barely squeaked by, in large part because BIPOC voters swung toward Trump from 2016. Look at Miami-Dade County, which continued zooming rightward.

"But downballot Democrats are doing okay! There won't be that much ticket-splitting!" - Things are always true until they're not. I'm 23 years old, and when I was born in 2000 it was likely unthinkable that polarization would one day be as virulent as it is today. It's not too crazy to suggest that things might change soon, just like they have changed so drastically in my 23 years on this planet.

"But Trump's criminal charges!" - Anyone who's been paying attention knows Trump is a criminal , and has known so since prior to the 2020 election. People either don't care or are already voting for Biden. While some people point out that polling suggests Trump's numbers will tank if he is convicted, I would remind everyone that with all of Trump's trials being delayed by SCOTUS, it looks increasingly improbable that there will be a verdict before the election, and an acquittal (which could happen!) would likely cause his numbers to shoot up.

The cold, hard truth is this:

Democrats have three options. They can either push Biden to get his act together and stop alienating the voters he needs, push him to drop out and install someone else as the nominee, or accept that Trump is going to win. Given Project 2025, the latter should not be considered an acceptable option.

1. There was like one poll that showed only 55% of Democrats believing this. It hasn't been a huge narrative at all, but on the margins it could account for response bias.

2. Democrats had a history of performing poorly in special elections particularly because of low turnout but there has been a big correction of that.

3. We all know that the Cuban vote in Miami-Dade County leans heavily Republican and it's swing in 2016 was probably a fluke. Other minority groups saw less swing.

4. Once again, Democrats have struggled down ballot in off or midterm years, but proved that wrong in 2022. The President tends to drag the downballot down, but somehow that didn't happen and that calls into question whether or not any discontent were Biden translates into enough low turnout this year re-elect Trump. I wouldn't count on it.

5. Prior to 2020 Trump didn't have actual charges filed against him. It's a huge anchor for any candidate to have charges against them and the delays only cause a situation similar to what Clinton in 2016 faced, the innuendo and did/could/would narrative.

In short I wouldn't bet on a Democrat with that much baggage winning even if the polls did say so despite those polls being a positive for my party.

Also he missed out the fact that polls are contradicting eachother. I don’t care where they come from but when Biden +6 and Trump +5 happen within 3-4 days of each other, that does not exactly make me want to trust polls more.

[GeorgiaModerate]

https://twitter.com/admcrlsn/status/1754240064647786652

[Reactionary Libertarian]

https://twitter.com/admcrlsn/status/1754240064647786652

Important to note that if Trump is tied in PV he’s almost definitely winning EV. 2022 showed us that doesn’t have to be the case, but it’s very likely to be that way in 2024. So Trump is still decently favored here.

[BG-NY (permanently retired)]

https://twitter.com/admcrlsn/status/1754240064647786652

G Elliot Morris, not unbiased at all, developed the new methodology and compiled the polling rankings for 538.

[Annatar]

G Elliot Morris is a democratic hack, not sure why anyone should take seriously anything he writes including his assessment of what polls are accurate. Nate Silver is a liberal as well but he tried to keep his analysis neutral.

[Progressive Pessimist]

Turnout is gonna tank this year, so I would expect ticket splitting to go up slightly. Low-propensity voters who vote straight-ticket are going to stay home.

Isn't that good for Biden?

[GeorgiaModerate]

The new 538 pollster rating methodology is very transparent and seems like a sound method of evaulation.  If you want to criticize their ratings, do so on methodological grounds, not because you don't like Morris's political leanings; doing the latter makes someone as much of a political hack as they accuse Morris of being.

Overview: https://abcnews.go.com/538/best-pollsters-america/story?id=105563951

Detailed methodology: https://abcnews.go.com/538/538s-pollster-ratings-work/story?id=105398138