I concur with the general ambiguity of the race based on schizophrenic polling data, but will say the profile of the late-breaking vote seems Trump-friendly/challenger-friendly. Anybody who has yet to be persuaded in the Democrats' direction by the critical mass of abortion rights messaging seems unlikely to be motivated by it in the next 25 days. Either these voters aren't being correctly counted by pollster modeling or.....the Democrats are in deep shit.
As for Ohio, I'd be very skeptical of polling there showing Harris overperforming. I was gonna say that I can't remember the last time pollsters didn't badly undersample Republicans in the Buckeye State, but then I remembered when they last got it right....in 2012. Ever since then, Republicans have decisively outrun the polls in every statewide Ohio contest. I'd mentally prepare myself for a double-digit Trump win and a decisive Moreno win.
I promise I don’t just seek out your posts to disagree with them, but I have a completely different view of the late breaking voters. Who are the undecided Trump voters? Are there really that many people who haven’t made up their mind about him?
Undecided Harris voters though? Yeah, I can imagine a lot of them. People are still making up their mind about her, partly because of the ludicrous claim that she is avoiding the media and not providing any details of how she will govern, but also because of the way she became the candidate, and, sadly, because she is also a woman of color.
I think it is far more likely than not the undecided voters will break for Harris by a significant amount.
As far as OH polls, agree with you there with the exception of 2022 Senate. As the election approached Vance’s lead increased and the final aggregate (6.2) was spot on (Vance won by 6.1). So it’s possible that the errors of 2016, 2018, and 2020 have been addressed. A double-digit Trump win I could see, but I think Moreno’s ceiling is a 2-3 point win.
Yeah Harris got into the race in July more voters seem not sold on her than undecided on Trump. most of the focus groups are some form of "I can't vote for him but I don't know enough about her" again who knows maybe they're all missing Trump shy voters but based on what limited data we have, it seems like most of the undecideds have ruled out Trump but have not committed to Harris yet.
Between this theory of the undecides, the pollsters' expectation that they've solved the 2016 & 2020 D+ leans, and Harris' supposedly* bigger turnout machine, I think Harris' vote percentage is more likely to grow than Trump's.
But that could be countered by the sheer volume of lies spewing from Trump and spreading unskeptically throughout the RW media sphere. Plus, I put an asterisk above, in case Musk's $180 million turnout operation makes up for the RNC shortcomings.
Exit polls show that the late-deciding voters broke to Trump in 2016 and even more decisively to Trump in 2020. It won't be the same subset of voters but there's certainly precedent for Trump closing well. So who are they? Low-information populists, mostly white and not terribly political but who fall squarely into that disproportionately sized quarter of the electorate who is socially conservative and economically liberal insofar as they have any cogent ideology at all. They want all the jobs to stay in America but make sure to buy the cheapest, Chinese-made option whenever given a choice. They want to expand government to do a number of things on their behalf but bristle at the prospect of paying one more penny in taxes. They want to get record annual raises but still think goods and services should cost what they did when they were children. They have some problems with Trump, but "....oh that Trump economy was wonderful...kind of...I think....when was Trump President again?"
Undecided Harris voters? It's a tougher call for me. Maybe center-right women who don't agree with her on policy but don't like Trump and would love to see the first woman President. And yeah, maybe a small sliver of misogynistic male Democrats who've voted for a black man and (probably) a number of white women before but who draw their line at a black woman. I still think this is a microscopic number, but I often underestimate people's prejudices.
Ultimately I think Harris pulls would-be undecided voters from the smallest ideological quadrant of the electorate.....socially liberal and economically conservative. They have an oversized footprint in the chattering class and in corporate boardrooms so we overestimate their electoral muscle before elections. I don't like her chances of picking off more late deciders than Trump does.
I checked it out and you're right that the final round of 2022 Ohio Senate polling wasn't far off. I had remembered Ryan running closer to Vance but those polls must have been earlier in the cycle.
Can agree with your take on the Trump voter, but I doubt very many people fitting your description are still undecided, they’re all in for Trump and probably have been since Biden was the nominee.
But for Harris, what about Haley voters? First time / infrequent voters who are still deciding whether they’re going to make the effort to actually vote? Dead-end Biden supporters who are still upset he is not the nominee (I doubt there are many of them but possible). I can think of a lot of folks who will probably end up supporting Harris who may not be ready yet to admit it, but still don’t see those people for Trump.
Considering Trump's grew his 2020 vote by nearly 18% from 2016, I'm not at all persuaded that he's the third-place option to the couch for significant numbers of undecideds.
But in 2020 he was not a convicted felon, he had not attempted a coup, he was not responsible for the end of Roe v Wade. Trump picking up votes from 2016 makes sense when we look back at it. Plenty of people who didn’t vote for him in 2016 saw what he did and liked it, so wanted to keep him as president. Fortunately, more people were turned off by what he did. There’s also the incumbency factor, which may be over stated but was probably responsible for some of that gain.
But what has he done since that is going to attract new voters? Potentially Dobbs could be rewarded, but considering how far away Republicans are running from abortion that seems highly unlikely. Folks who miss the Trump economy, sure, but as I said I think they’re already for him.
Do you know if there is any analysis on the composition of undecided voters? I strongly suspect the demographics would favor Harris but admit I haven’t seen any data.
I'm not aware of any expert analysis on the composition of undecided voters. Aside from focus groups, I suspect it's a guessing game just like poll modeling generally.
Considering the incumbency factor, the cult of personality cementing itself and COVID not really being blamed on him (since it was a huge global issue) I’m not surprised he picked up votes compared to 2016
Of course, Biden grew Hillary's 2016 raw vote total proportionally 31% more than Trump grew his, despite starting from a higher base. So thinking Trump's the first choice isn't a strong bet, either.
Here is a Republican pollster who tried to asked when ( ealry, couple weeks, couple days, day of election) LV decided on their candidate. Grain of salt but now is the time Harris can make her case to those that have positive approval of her. Trump’s largest cohort are voters that decide on the day of the election.
That is why i think right now a lot of R sponsored polls are coming out now to try to flood the field with a false IMO narrative that Trump has momentum. People that have a favorable opinion of Harris are deciding now and these polls and stories are trying to blunt that with the undecided voters.
I've always assumed the Haley vote would ultimately break toward Trump even if Harris got a decent percentage of them, but it's a deceiving metric because it's unclear how many of the Haley voters were just Democrats and left-tilting independents trolling Trump. As for your other groups, it's hard for me to reconcile statistically significant numbers of voters who still haven't seen enough to disqualify Trump but will see enough in the next 25 days.
It's always been a reasonable thesis that people would head to the polls and the gravity of the situation would become real to them, rendering them incapable of filling in that circle for Trump. But since that didn't happen in the past and his tactics have been normalized for nearly a decade, it's even tougher to imagine late-breaking voters having crises of conscience preventing them from voting for him this time.
I feel like I’ve seen an analysis showing no appreciable difference in Haley votes in open vs closed primaries which would indicate that the effect of trolling is minimal but it’s been months so not sure where. But agree with Henrick’s comment above that for a lot of undecided voters it’s going to be Harris or no vote. But again, will admit that is not based on any actual data
IIRC you’re correct about there being little material difference between closed and open for Haley numbers
Now, I think that Harris winning 40% of Haley supporters is probably on the bullish side, but even 20-25% of them flipping would be a major problem for Trump considering he generally trails with Indies in most polls, too. Only question is how much the electorate Is wind up being and that’s what’s very hard to properly model
I hadn't even considered dead-end Biden voters. You expressed skepticism of there being many, and I'd see your bet and raise it to... pretty much zero. Interesting thought though
Haha yeah, I was just remembering the conversations on DK when following the announcement Biden wouldn’t run and there were quite a few folks upset, saying they wouldn’t vote for his replacement, etc. I’m sure the vast majority of folks will end up voting for Harris, but I could see some holding out in polling. Enough to actually affect the polls? Probably not. But at the same time they seem just as, if not more, likely than someone who hasn’t made up their mind on Trump. I mean there’s at least a grudge for motivation there. The only reason I can’t think for having an unformed opinion on Trump is a potato brain.
Not many, but in a really close race, just a few thousand in a state like NC could make a difference. And I could definitely imagine some people who were cool with voting for an old white man but have to bring themselves to vote for a Black woman.
That's a very coherent post, but do you really think there isn't a significant number of Haley voters and other principled Republicans and Republican-leaning independents who are wrestling with the question of whether to vote for a Democrat for President for the first time?
I think there is. The unresolved question is how many of these people there are proportional to the ongoing collapse that seems to be happening among downscale voters. There could well be more than I think, but the dot diagram of economically conservative and socially liberal voter allocation leads me to believe there's fewer of these types of voters than conventional wisdom suggests. Before Trump, it strikes me that Mitt Romney was counting on the same demographic that Harris is now to pull him over the finish line but there simply weren't enough of them.
Fair. It’s possible I’d put it on the very end of the bell curve with about equal probability of a 7 point win for Brown. I think the current polls are probably pretty accurate and Brown ends up holding on by a point or two.
I might have to think about it a little more but initial thought is an Osborne win in NE is about as likely as a decisive (5 pt) Moreno win in OH. That’s probably a little bit too optimistic but not by much.
You could easily make the opposite argument - Trump has been around for 3 goes now - if people were going to support him they already would. The undecided voter could just be a Trump hater who is skeptical of Democrats/liberals/women/non-white people/Kamala Harris in particular who they still likely dont know much about.
There's no proof either way (the consistently shifting partisan divide in early voting has made judging anything off of it completely useless and polls arent proof).
To put it another way - Trump in two attempts hasnst topped 47% nationally, and theres no reason to expect him to this time.
One should expect this result - Trump will get between 47 and 50 percent in the swing states (PA, WI, MI, NV, AZ, NC, GA) and Harris will get between 48.5 and 50.5% in those same states. The actual outcome - who knows - each candidate could win all of those states or none or anything in between in almost any combination. Harris is more likely to win, but its by far not a given.
Moreover, consider the fact that Trump underperformed polls, often significantly, in just about every Republican primary this year. This alone indicates massive Republican discontent with their MAGA candidate.
It also indicates that something is seriously wrong with polling. Pollsters have adjusted for their perceived misses in 2016 and 2020 – but have, IMHO, failed to adjust to the post-Dobbs reality where it is Democrats who have overperformed in the Midterms Elections and almost every special election since Dobbs.
Case in point: Women account for approx. 55% of the Early Vote!
I concur with the general ambiguity of the race based on schizophrenic polling data, but will say the profile of the late-breaking vote seems Trump-friendly/challenger-friendly. Anybody who has yet to be persuaded in the Democrats' direction by the critical mass of abortion rights messaging seems unlikely to be motivated by it in the next 25 days. Either these voters aren't being correctly counted by pollster modeling or.....the Democrats are in deep shit.
As for Ohio, I'd be very skeptical of polling there showing Harris overperforming. I was gonna say that I can't remember the last time pollsters didn't badly undersample Republicans in the Buckeye State, but then I remembered when they last got it right....in 2012. Ever since then, Republicans have decisively outrun the polls in every statewide Ohio contest. I'd mentally prepare myself for a double-digit Trump win and a decisive Moreno win.
I promise I don’t just seek out your posts to disagree with them, but I have a completely different view of the late breaking voters. Who are the undecided Trump voters? Are there really that many people who haven’t made up their mind about him?
Undecided Harris voters though? Yeah, I can imagine a lot of them. People are still making up their mind about her, partly because of the ludicrous claim that she is avoiding the media and not providing any details of how she will govern, but also because of the way she became the candidate, and, sadly, because she is also a woman of color.
I think it is far more likely than not the undecided voters will break for Harris by a significant amount.
As far as OH polls, agree with you there with the exception of 2022 Senate. As the election approached Vance’s lead increased and the final aggregate (6.2) was spot on (Vance won by 6.1). So it’s possible that the errors of 2016, 2018, and 2020 have been addressed. A double-digit Trump win I could see, but I think Moreno’s ceiling is a 2-3 point win.
Yeah Harris got into the race in July more voters seem not sold on her than undecided on Trump. most of the focus groups are some form of "I can't vote for him but I don't know enough about her" again who knows maybe they're all missing Trump shy voters but based on what limited data we have, it seems like most of the undecideds have ruled out Trump but have not committed to Harris yet.
Between this theory of the undecides, the pollsters' expectation that they've solved the 2016 & 2020 D+ leans, and Harris' supposedly* bigger turnout machine, I think Harris' vote percentage is more likely to grow than Trump's.
But that could be countered by the sheer volume of lies spewing from Trump and spreading unskeptically throughout the RW media sphere. Plus, I put an asterisk above, in case Musk's $180 million turnout operation makes up for the RNC shortcomings.
No worries. Different takes are always welcome.
Exit polls show that the late-deciding voters broke to Trump in 2016 and even more decisively to Trump in 2020. It won't be the same subset of voters but there's certainly precedent for Trump closing well. So who are they? Low-information populists, mostly white and not terribly political but who fall squarely into that disproportionately sized quarter of the electorate who is socially conservative and economically liberal insofar as they have any cogent ideology at all. They want all the jobs to stay in America but make sure to buy the cheapest, Chinese-made option whenever given a choice. They want to expand government to do a number of things on their behalf but bristle at the prospect of paying one more penny in taxes. They want to get record annual raises but still think goods and services should cost what they did when they were children. They have some problems with Trump, but "....oh that Trump economy was wonderful...kind of...I think....when was Trump President again?"
Undecided Harris voters? It's a tougher call for me. Maybe center-right women who don't agree with her on policy but don't like Trump and would love to see the first woman President. And yeah, maybe a small sliver of misogynistic male Democrats who've voted for a black man and (probably) a number of white women before but who draw their line at a black woman. I still think this is a microscopic number, but I often underestimate people's prejudices.
Ultimately I think Harris pulls would-be undecided voters from the smallest ideological quadrant of the electorate.....socially liberal and economically conservative. They have an oversized footprint in the chattering class and in corporate boardrooms so we overestimate their electoral muscle before elections. I don't like her chances of picking off more late deciders than Trump does.
I checked it out and you're right that the final round of 2022 Ohio Senate polling wasn't far off. I had remembered Ryan running closer to Vance but those polls must have been earlier in the cycle.
Can agree with your take on the Trump voter, but I doubt very many people fitting your description are still undecided, they’re all in for Trump and probably have been since Biden was the nominee.
But for Harris, what about Haley voters? First time / infrequent voters who are still deciding whether they’re going to make the effort to actually vote? Dead-end Biden supporters who are still upset he is not the nominee (I doubt there are many of them but possible). I can think of a lot of folks who will probably end up supporting Harris who may not be ready yet to admit it, but still don’t see those people for Trump.
I know focus groups are unscientific but this seems to be what both Longwell and Luntz’s work suggests.
And that for most undecideds it’s a question of Harris or the couch, rather than Harris or Trump
Considering Trump's grew his 2020 vote by nearly 18% from 2016, I'm not at all persuaded that he's the third-place option to the couch for significant numbers of undecideds.
But in 2020 he was not a convicted felon, he had not attempted a coup, he was not responsible for the end of Roe v Wade. Trump picking up votes from 2016 makes sense when we look back at it. Plenty of people who didn’t vote for him in 2016 saw what he did and liked it, so wanted to keep him as president. Fortunately, more people were turned off by what he did. There’s also the incumbency factor, which may be over stated but was probably responsible for some of that gain.
But what has he done since that is going to attract new voters? Potentially Dobbs could be rewarded, but considering how far away Republicans are running from abortion that seems highly unlikely. Folks who miss the Trump economy, sure, but as I said I think they’re already for him.
Do you know if there is any analysis on the composition of undecided voters? I strongly suspect the demographics would favor Harris but admit I haven’t seen any data.
I'm not aware of any expert analysis on the composition of undecided voters. Aside from focus groups, I suspect it's a guessing game just like poll modeling generally.
Considering the incumbency factor, the cult of personality cementing itself and COVID not really being blamed on him (since it was a huge global issue) I’m not surprised he picked up votes compared to 2016
Of course, Biden grew Hillary's 2016 raw vote total proportionally 31% more than Trump grew his, despite starting from a higher base. So thinking Trump's the first choice isn't a strong bet, either.
In absolute terms Biden grew Hillary's raw vote total by 37% more than Trump grew his.
Here is a Republican pollster who tried to asked when ( ealry, couple weeks, couple days, day of election) LV decided on their candidate. Grain of salt but now is the time Harris can make her case to those that have positive approval of her. Trump’s largest cohort are voters that decide on the day of the election.
https://x.com/cygnal/status/1844443830491414962
That is why i think right now a lot of R sponsored polls are coming out now to try to flood the field with a false IMO narrative that Trump has momentum. People that have a favorable opinion of Harris are deciding now and these polls and stories are trying to blunt that with the undecided voters.
I've always assumed the Haley vote would ultimately break toward Trump even if Harris got a decent percentage of them, but it's a deceiving metric because it's unclear how many of the Haley voters were just Democrats and left-tilting independents trolling Trump. As for your other groups, it's hard for me to reconcile statistically significant numbers of voters who still haven't seen enough to disqualify Trump but will see enough in the next 25 days.
It's always been a reasonable thesis that people would head to the polls and the gravity of the situation would become real to them, rendering them incapable of filling in that circle for Trump. But since that didn't happen in the past and his tactics have been normalized for nearly a decade, it's even tougher to imagine late-breaking voters having crises of conscience preventing them from voting for him this time.
I feel like I’ve seen an analysis showing no appreciable difference in Haley votes in open vs closed primaries which would indicate that the effect of trolling is minimal but it’s been months so not sure where. But agree with Henrick’s comment above that for a lot of undecided voters it’s going to be Harris or no vote. But again, will admit that is not based on any actual data
IIRC you’re correct about there being little material difference between closed and open for Haley numbers
Now, I think that Harris winning 40% of Haley supporters is probably on the bullish side, but even 20-25% of them flipping would be a major problem for Trump considering he generally trails with Indies in most polls, too. Only question is how much the electorate Is wind up being and that’s what’s very hard to properly model
That's definitely a valid thought.
I hadn't even considered dead-end Biden voters. You expressed skepticism of there being many, and I'd see your bet and raise it to... pretty much zero. Interesting thought though
Haha yeah, I was just remembering the conversations on DK when following the announcement Biden wouldn’t run and there were quite a few folks upset, saying they wouldn’t vote for his replacement, etc. I’m sure the vast majority of folks will end up voting for Harris, but I could see some holding out in polling. Enough to actually affect the polls? Probably not. But at the same time they seem just as, if not more, likely than someone who hasn’t made up their mind on Trump. I mean there’s at least a grudge for motivation there. The only reason I can’t think for having an unformed opinion on Trump is a potato brain.
Not many, but in a really close race, just a few thousand in a state like NC could make a difference. And I could definitely imagine some people who were cool with voting for an old white man but have to bring themselves to vote for a Black woman.
THAT, I think, is the bigger issue
So going back to your previous post, not zero.
Also, voters angry about support for Israel who realize Trump would hurt them and people they love here in the U.S.
That's a very coherent post, but do you really think there isn't a significant number of Haley voters and other principled Republicans and Republican-leaning independents who are wrestling with the question of whether to vote for a Democrat for President for the first time?
I think there is. The unresolved question is how many of these people there are proportional to the ongoing collapse that seems to be happening among downscale voters. There could well be more than I think, but the dot diagram of economically conservative and socially liberal voter allocation leads me to believe there's fewer of these types of voters than conventional wisdom suggests. Before Trump, it strikes me that Mitt Romney was counting on the same demographic that Harris is now to pull him over the finish line but there simply weren't enough of them.
I have to say, a ~5-point win by Moreno would be shocking but not that surprising. I have to believe that's possible.
Fair. It’s possible I’d put it on the very end of the bell curve with about equal probability of a 7 point win for Brown. I think the current polls are probably pretty accurate and Brown ends up holding on by a point or two.
I might have to think about it a little more but initial thought is an Osborne win in NE is about as likely as a decisive (5 pt) Moreno win in OH. That’s probably a little bit too optimistic but not by much.
I can see it sure, but imo if Brown were to lose it would be by a razor thin margin(5 seems kinda high to me)
You could easily make the opposite argument - Trump has been around for 3 goes now - if people were going to support him they already would. The undecided voter could just be a Trump hater who is skeptical of Democrats/liberals/women/non-white people/Kamala Harris in particular who they still likely dont know much about.
There's no proof either way (the consistently shifting partisan divide in early voting has made judging anything off of it completely useless and polls arent proof).
To put it another way - Trump in two attempts hasnst topped 47% nationally, and theres no reason to expect him to this time.
One should expect this result - Trump will get between 47 and 50 percent in the swing states (PA, WI, MI, NV, AZ, NC, GA) and Harris will get between 48.5 and 50.5% in those same states. The actual outcome - who knows - each candidate could win all of those states or none or anything in between in almost any combination. Harris is more likely to win, but its by far not a given.
Moreover, consider the fact that Trump underperformed polls, often significantly, in just about every Republican primary this year. This alone indicates massive Republican discontent with their MAGA candidate.
It also indicates that something is seriously wrong with polling. Pollsters have adjusted for their perceived misses in 2016 and 2020 – but have, IMHO, failed to adjust to the post-Dobbs reality where it is Democrats who have overperformed in the Midterms Elections and almost every special election since Dobbs.
Case in point: Women account for approx. 55% of the Early Vote!
You should prepare yourself for Brown winning.