If The 2024 Election Were Held Today, Would Trump Win? 01:03:42

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

Okay, I will. Yeah. I'm going to try just not to touch anything besides myself. Not like that. Oh my God. You know, when you said informal, I wasn't quite sure how informal. Yeah. Can that be the lead in song, please? Yeah, I know. I was thinking, can we replace the theme music today when I touch myself? Oh, Jesus Christ. When I think about you.

Hello and welcome to the FiveThirtyEight Politics Podcast. I'm Galen Druk. Stop me if you've heard this one before. Two election modelers and a polling editor walk into a bar. I mean, a podcast. The host asks, who's going to win the 2024 election?

Corny jokes aside, today I've assembled a crew from different parts of the election data world to try to make sense of what we know one year out from the 2024 election. And before I go any further, let me introduce them. So here with me is Director of Data Analytics, Elliot Morris. Welcome, Elliot. Hey, Galen. I'm happy to be here for this most ambitious crossover episode of all time.

It is very ambitious. Also here with us is editor for news surveys at the New York Times, Ruth Gelnick. Welcome back to the podcast, Ruth. Thanks for having me. So great to have you. And here with us for the first time ever is partner at the election modeling website, splitticket.org, Lakshya Jain. Welcome to the podcast. Thanks for having me, guys.

So to set the scene a little, over the past week, we got a lot of new polls meant to gauge the mood of the electorate exactly one year before the presidential election. Most of them showed a similar picture. Not only does former President Trump outperform Biden in the swing states, he leads on average in national polls as well.

The suggestion being that if the election were held today and Trump and Biden were in fact the nominees, that Trump might be favored to win not just the Electoral College, but the national popular vote too. There's a catch though. Well, there are a couple catches. So perhaps most importantly, the election is not today. Also, we got other data last week from actual elections showing Democrats performing relatively well.

So one year out, where does this all leave us? We're going to try to answer that question today. We're also going to play a game. Also, Tim Scott dropped out of the Republican primary, so we'll get to that too. But we got a lot to cover, so let's dive in. Elliot, kick us off here. Historically, how useful or predictive is polling one year before an election?

It's not there you go easy answer. It's just not historically the average error for a poll taken 360 days before an election the predictive margin of error Let's call it is like 25 percentage points on margin Which is needless to say completely useless for predicting the national popular vote Yeah, Canada tough by 10 points or they're down by 10 points. I

That's not going to be a useful prediction for me to tell you. There is a potential wrinkle here, which is that as polarization has increased in the electorate and as voting preferences has become more stable and our elections have gotten closer,

maybe those early polls tell us more, are a stronger signal. I looked at this before our last presidential election model and found between 1940 and 2000, you get that 25 percentage point average margin of error. Between 2000 and 2016, the margin of error on average for a poll taken 360 days before the election is a lot more like 10 points than 25. So maybe it's more reliable now. But like,

The candidates are tied. So a 10 point margin of error isn't really buying you a whole lot more than a 25 point margin of error.

So we got some data there. We also have some caveats. Are there any other caveats that people want to shout out or disagree with the data for some reason? Is there a reason that even the past 20 years shouldn't apply to our experience today? I mean, I think Elliot's exactly right. I think that, you know, it is hard to tell this early what can be predictive. I think that, you know,

We, as Elliot said, polarization is on the rise and the group of voters that we're pulling from to really make these decisions is getting smaller and smaller. So it's interesting to sort of represent who that group is when you're out. But, you know, a lot can change in a year. And that's a reasonable expectation to think a lot can change in a year. Elliot and Ruth are absolutely correct that it's a long time to go from now. And also that the pool of potential voters to swing and switch sides is huge.

but I also think that in general, when people respond to polls and we've seen this over and over a year on out, a lot of them aren't necessarily really thinking about the exact contours of the race because the campaigns haven't happened yet. And,

and it's not really framed as the choice that the election usually ends up being. And so people are responding to a almost a different reality in their minds than the one that tends to manifest on the ballot box a year later. It's why we see, you know, in Ohio, for example, every time an August or July poll comes around, Democrats tend to be ahead. And then as the election nears, you

you get this coalescing of partisans, so to say. I mean, we saw this with Tim Ryan last August. So I think for all of those reasons, I'm a little bit skeptical when I see like a bunch of polls saying that Biden is only leading 18 to 29 voters by one point when they voted for him by 25 last time. So...

All right. Starting to get specific there and specifically about the New York Times polling that Ruth put out last week. Other polls too. Other polls. It's not just New York Times. And other polls. We're going to get into the whole argument of why polls would not be predictive one year out. But put that aside for a second.

Is there a reason to think that this cycle in particular, if the nominees are in fact Trump and Biden, that the polling could be more stable or predictive because both candidates basically have 100% name recognition? You could even argue that both candidates are running in some ways as incumbents because Americans know what it's like to live under the presidencies of both candidates and can be specific about what their lives were like

under the experiences of both candidates, which they do in the polling that we saw from The Times and others in the past week, basically talking about the economy and foreign entanglements and things like that. So I understand the historical data. This is also a potentially historical election in the sense of who the actual candidates are. Should we reassess those priors? You kind of got right ahead of me. So good job, Galen. I was going to say there's

You can take me through the story of the election in multiple ways at this point. One is that everyone knows who Joe Biden and Donald Trump are, and so the preferences are even harder than polarization would suggest, and the polls are more reliable.

However, there are some events that we know that are going to take place next year that might make those polls not reliable, and they have to do with the people who are on the ballot. So there's multiple trials against Donald Trump. I don't think that potential convictions exist.

of federal felony indictments are priced into the polls at this point. I also don't think that the polls are pricing in a fall for Joe Biden or a different medical health emergency that could happen as a result of his age. And same for Donald Trump. He's also old, so that's not priced in either. So that's a little bit wishy-washy, but I think that angle to whether or not polls are reliable right now could go either direction.

Yeah. And I think in that sense, right, like you could absolutely convince me that like these polls matter a lot more than they would have historically mattered. Like, I mean, four years ago, we kind of saw, um,

The New York Times released an excellent set of polls that ended up being eerily predictive of what the actual results would be. But I hesitate to extrapolate from a sample size of N equals one, which is basically what that was. We don't really know if that was, you know,

indicative of something else or whether the New York Times just got a really good sample that time. And this time around, as Elliot said, because there's so much that could come through, right? Like Donald Trump's indictments and criminal trials, we saw in the Times survey itself that a lot of voters switched their preferences upon learning like, hey, yeah, if he gets convicted, I would probably not vote for him. That could change the entire complexion of the race because it's an event that hasn't happened yet.

Sorry, I want to jump in. Ruth, do you think those people are expressive responding? There's my very specific question. The, you know, if Trump is indicted or is convicted and sentenced to jail time. So this is something that we think about in polling and we do see in polling from time to time, right, where people are sort of following their partisan cues and telling us what they think.

think we want to hear or what they think they're supposed to say in response to that, if they are doing, you know, I mean, what's right, what's partisan. You know, for example, we notice expressive responding when like a political leader says something that's breaking with his party's tradition. And then you start to see people from that party follow. They might just be expressively responding. Or in this case, per Elliot's question, you know, if

We're asking, oh, if Trump is convicted and sentenced to prison, and they're like, well, you're telling me that's bad. So maybe I should switch over and vote for Biden. We found that around 6% of voters sort of made that switch when we gave them that option. We really wrestled with the question wording because it can be pretty leading. I mean, there's really no other way to ask it, but we had to kind of spell out this scenario. It feels a little bit early to know for sure the total impact of that because, again,

It's so hypothetical and which trial and which wraps up first. So this was really kind of like our first volley into understanding what people might do. The relatively small share tells me that I think it's real. The impact is hard to say, right? Like it's a very specific scenario. What if that doesn't wrap up before the election? Like there's a lot that's uncertain there, but it was certainly worth poking. Yeah.

Ruth, I do want to get you to weigh in on one more thing, which is that Lakshya mentioned that poll from November 2019. And it's a poll that looms large in the elections analysis universe because it was released when Elizabeth Warren was increasing in her support in the polls. It

almost single-handedly caused this sort of freak out about Elizabeth Warren's electability because it showed that she wouldn't win an election against Trump while Biden would. And it was a poll of the swing states, much like you did this year or just last week. And the margins that Biden was winning by in those swing states

ended up being extremely close to the margins by which he won those exact swing states one year later. And what I should also mention is that the difference between November 2019 and November 2020 is that there was a global pandemic. Talk about unknowns, unpredictable things that could shake up a race.

All of that happens, you know, a health crisis, an economic disaster, all kinds of things happen. But one year later, the polls are exactly the same. So, I mean, not exactly the same, but basically exactly the same. So Ruth, talk to me about how that makes you think about polling predictability. Yeah, so to Lakshya's point, you know, we did recognize that was an N equals one and we don't want to over-interpret from that at the same time. I think somebody I know said that we were the most accurate pollster in America. Maybe it was 530.

It's kind of a trusted website. Nice. So, you know. Nice.

But in all seriousness, we entered into thinking about this series of polls that we released a couple weeks ago with that 2019 poll in mind, right? Like that was fairly accurate. It had a good beat on the race. And so as we started down this process, we sort of thought of that as parallel. How can we do something similar? How can we try to predict, you know, what might happen in the race? And as we got deeper into it, you know, we started to sort of think about how it's really different. And that's what we ultimately found here is that this

rematch between two candidates that are deeply unpopular, that everybody knows that people's opinions are so baked in about. It couldn't really be about electability in the same way that it was in 2019 with Warren and the Democratic state of the race. And that's the other thing is that race, the Democratic side was really uncertain at that moment. And that poll really clarified it for people on the Republican side. It's really not uncertain. Donald Trump is leading by huge margins and right. Like the fight for second place is still shaping up.

But it's still second place on the Democratic side. You know, somebody could challenge Biden. It's possible they see this bowl and they could challenge Biden. But our goal sort of shifted from the electability question to really testing these well-known, deeply disliked candidates weaknesses. And I think that is at least what I find most instructive one year out is what are the things that people dislike about these candidates? How does that impact them or not?

And one of my sort of biggest takeaways from this poll is are the things that we think are Biden's weaknesses, right? Like people, how they feel about his age, his mental acuity, the economy, those are hurting him really dramatically at this point in the race. It's still early to say what that means. And for Trump, the things that are sort of

generally thought of as his weaknesses are not hurting him to the same degree. And yes, if we test this very specific scenario of conviction and sentenced to jail time, we do see some, you know, some attrition and support on a fairly small scale. So it really became less about predicting the race and more about testing their sort of deep weaknesses to see how that impacted people's choices.

I want to nail down what you're saying here is different about 2019. On one hand, in 2019, Biden was thought of relatively favorable. He was, I think, barely above water with the voters then, while Donald Trump was significantly underwater in terms of favorability. Is it different because you're saying now, since both candidates are deeply unpopular, there's so much disaffection that that could cause more volatility in the polls?

Oh, I don't think it will cause more volatility in the polls. It's hard to say. I mean, it's really hard to say how this impacts one year from now. It's almost like that wasn't our goal this time around, was to use this as a prediction of the future. It was more to say, one year out from the race, what do things look like? I actually think it could cause less volatility in the polls because people's views are very baked in and the pot that these candidates are choosing from is getting smaller and smaller. One thing that's really interesting, you mentioned that Biden was

less unpopular then in, in 2019, 2020 cycle, 9% of people viewed both candidates unfavorably. Now that number is 20%. Like it's really grown. It's almost doubled. Um, so you do have this increasing group who just really dislikes both candidates and it's not that it makes the race more volatile, but you know, it's, it makes it easier to poke at their weaknesses to see what, what may make people change if anything.

It's interesting because we're in a situation where broadly the American electorate hates both people running. And I think we can use that word pretty accurately here. I mean, these are very strong disapproval ratings we're seeing for both Trump and Biden. I think Biden is at 38% of the 538 averages. And it's not like Trump's favorability is much better there either.

But at the same time, their nominations are more or less the most likely events to happen and by a significant margin. And what I've repeatedly heard from people is that like, you know, they can't really believe that this is the nomination matchup that we're going to get. But it really is heading to that. And I think...

What's going to be interesting here is that 2019, 2020 was defined by, you know, in a sense, you can almost think of it as like a referendum on Donald Trump. Did you like Donald Trump or not? Because Joe Biden was thought of as essentially a ham sandwich candidate, a very generic Democrat that broadly the electorate found unobjectionable. Yeah, there were some questions about his age and whatnot, but nothing like what we see today.

And Donald Trump was the incumbent president in the headlines every single day. Contrast that to today, where you're going to have a lot of comparisons on Biden's record versus on what Trump did and what Trump is going through right now. And the choice is going to get much more difficult for voters, I think, in some ways, because it's

Who do you dislike more? That speaks to one of the most interesting stats, in my opinion, from our poll. We asked this question if people if Biden's policies helped them or hurt them, if Trump's policies helped them or hurt them. And you had the majority of people, the vast majority of people saying that Biden policies hurt them and Trump's policies help them. So it's that kind of, you know, mixed memories. But they do feel like Biden's policies have really hurt them and they don't feel that way about Trump's policies.

Yeah. I mean, this is just experience out in the field and your poll is more scientific than this. But when I was in Simi Valley for the second Republican debate and Simi Valley is, you know, on a razor's edge in terms of its partisanship, like Biden won the city by less than a percentage point, I think half a percentage point or something like that. You hear constantly about prices under Trump and, you know,

foreign entanglements under Trump. And that's why I say he's running as an incumbent in some ways, because people have very specific things that they can talk about when they think about life under Trump, which could be, you know, a challenge for Biden and may not make that sort of historical advantage for incumbents the same this time around, because...

Biden isn't really the only incumbent in a way if it ends up being Trump versus Biden. So I want to get a little more nerdy here. Let's talk about national polling.

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

We don't have a rolling average on the 2024 general election for good reason. There aren't actual clear, it's not clear who the candidates are at this point. So hold your horses. But if you want to do a rough average of the hypothetical matchup of Trump versus Biden, it looks like Trump is leading by about one point on average. Now, of course, winning the national popular vote does not mean you win. And for the past two elections, Republicans have had an advantage there.

What does the advantage look like in the Electoral College today? Do Republicans still, is it still clear that Republicans have an advantage? And if so, how big is it? Yeah, I mean, this is certainly why we polled battleground states or swing states, whatever you want to call them, instead of national for this particular poll, because we know that that makes a difference in the Electoral College. Americans don't vote nationally. They vote by state. And then we, you know, tally that up through the Electoral College. And right now,

based on what we found in our poll and what others, there was another swing state poll that came out right around the same time, that Republicans definitely have an advantage, an electoral college advantage, if things stay exactly the same. I do want to talk a little bit more about that point too, Ruth. What I noticed is in your guys' surveys, and I think Nate Cohn wrote about this too, that Donald Trump's electoral college edge has been significantly dulled if you look at the polling though, because

He's making gains among. So we mentioned earlier, I think that like this time around, voters are really just not happy with whoever is there. But there's different degrees of discontent. And what we see with Biden is the biggest degree of slippage in polling seems to be just like nonwhite voters, young voters who are upset. But.

But he's holding up really surprisingly well with older voters who are way less likely to question his age for reasons I think are pretty obvious and with white voters in general and especially white educated voters. And so why?

What I keep seeing in all these polls is that Biden's doing, yes, he's lost ground basically everywhere, but he's doing better in Wisconsin, Michigan, in Pennsylvania, and presumably also in states like New Hampshire, which have similar demographics. Whereas you go to Georgia and he slid quite a bit because of his decreased standing with African-American voters in Arizona.

where he's got some weaknesses with Latinos. And then you look at some safe state polls or in Nevada, I think Nevada is the best example. Nevada, his standing has catastrophically plummeted. What that leads me to think about is that

National polling can sometimes be a little bit weird because we still live under the electoral college system. And here, if you hold your own with white voters, you will pretty much always be competitive in a time of high polarization in the electoral college because there is a path for Democrats to win the presidency by just holding the Rust Belt and losing the Sun Belt and everything.

You could even lose the popular vote while doing it. I mean, Obama's tipping point state was three points to the left of the nation. Yeah, though, based on our polling right now, holding the Rust Belt is not a guarantee, right? Like it's I mean, obviously, you know, Biden was winning Wisconsin in our poll or was ahead in Wisconsin. Shouldn't say winning because nobody's winning anything. There's no actual election right now. Right.

But one thing that we saw is the more diverse the state across the six states we polled, the better Trump was doing. And conversely, the whiter the state, the better Biden was doing. So in Wisconsin, he's up to in our poll within the margin of error. So not actually up in any meaningful way.

But right now, as it stands, Michigan and Wisconsin, while they are whiter, older states that will be better for Biden, they're still not places where he has the right kind of traction he needs to succeed. But you are absolutely right that the electoral college advantage has dulled significantly as Biden has sort of, you know, done better with that group of older, whiter, more stable voters who I might add are much more likely to vote than younger voters who are sort of

at least curious about Trump at this stage in the race, but they're just a lot less likely to vote. Also a lot less likely to respond. Yeah. And if they're disaffected, they might not turn out next year, right? That's right. I think Lakshya earlier mentioned...

this dynamic where in not just the New York Times-Siano poll, but in other polls, young people are either tied or going towards Trump, which would imply like a 30-point swing since the last election, which we can be skeptical about, right? It's not that it's necessarily wrong. I think we can just be skeptical about it. One thing that I think is getting lost in the national conversation here, that swing is not necessarily entirely due to increased margins for Trump, but

a large proportion of young people just saying they're not going to vote or they're unsure. Right. So if you transform that to a margin, it looks like a huge gain for Trump. But in fact, I mean, it could just be a

sort of residual level of uncertainty among young people, and maybe they don't turn out, and that has the effect of increasing support for Trump, but it's sort of like a different mechanic. We're trying to infer, right, right now, the 2024 election outcome, not just nationally at the states, using polls, which we've agreed is fraught, maybe to a different

degree now than in the past. There's this great website called splitdicket.org, which in 2022 or after the 2022 election calculated the swing using the midterm election results. At the House level, they did this really cool model where they imputed the results in uncontested congressional districts. So you get like an apples to apples comparison.

And even there, you see a decreasing of the electoral college edge, huge swings against Democrats in the places where they have lopsided margins, places like, uh,

California, New York, to some extent even Vermont, New Hampshire, Connecticut, right? So that decreases their overall popular vote tally, but they did better in the state-level House vote in Wisconsin, in Michigan, in Pennsylvania, even a little bit in Ohio, which probably won't be competitive, but for what it's worth. Which points to an electoral college edge of like two points or one point, right? If you work out that math. So even there, I think we're probably living under an environment where

The propensity to get anti-majoritarian or counter-majoritarian, whatever you want to call it, outcomes is probably lower than it was in the past. But hey, we're still making predictions about the future. So we'll see. There's two things I wanted to touch on here briefly, Elliot, in connection to that. So when we did this, that national house vote aggregation, we actually found something really interesting. So in

In 2020, the electoral college bias was roughly about four points. And that was the tipping point state was Wisconsin. Biden won it by half a point. He won the national vote by about four and a half points.

This time around, the tipping point state was actually once again Wisconsin, but this time the national House vote was R plus 1.6. And Democrats in aggregate lost Wisconsin by about 1.8 at the House level. So that electoral college edge nearly entirely evaporated. I mean, 0.2% here, there is really nothing. And what we saw, to your point about safe states swinging away, and I think the Times also wrote this point,

Battleground states saw really, really competitive elections and people saw their choices as much more of a high stakes vote. Whereas in California or in New York, I mean, I don't really think anyone believed that Lee Zeldin was going to overturn New York's 2024 election results. Whereas you get to Pennsylvania, you get to Michigan, you get to Wisconsin, and those fears start having more heightened levels of concerns and

Probably rightfully so, given what they saw in 2020. And so it makes sense that when you start to look at the environments, even in turnouts in actual elections and performances in actual elections, that you see battlegrounds consistently being close and Democrats not really sliding there at all.

I want to put some numbers to the state-level polls that we're talking about. The New York Times obviously released a set this past week. We have a rough average from about 20 polls in each of the states since the midterm elections. And if you take those averages...

Trump leads in Georgia by three and a half points, leads in Arizona by two points. Biden leads in Wisconsin by two, Pennsylvania tied. Trump leads in Nevada by two and a half, and Michigan tied. So to be clear, we're talking about an environment where the Republican advantage in the Electoral College may have faded, but we're also talking about an environment where

If the election were held today, Trump would likely still win the Electoral College and perhaps maybe the national popular vote as well. Again, all of the caveats that we just mentioned. And underneath the conversation that we're having is all of this polling that we've gotten about voters of color and young voters. We talked a little about the young voters in particular. I want to talk about voters of color. So here's a quote from The Times. Biden barely leads at all among non-white voters under 45.

even though the same voters reported backing Mr. Biden by almost 40 points in the last election. As a result, the candidates are essentially tied amongst voters 18 to 29. We already talked about that. Now, Black voters in particular, the Times reads, long a bulwark for Democrats and for Biden, are now registering 22% support in these states for Mr. Trump, a level unseen in presidential politics for a Republican in modern times.

What Democrats will say is, we've seen this before. You know, when we're further out from an election, people, the choice isn't clear, the argument that Lakshya made earlier on. And the...

that the sort of partisan impulses can lessen. And there have always been more conservative Black Americans, more conservative Latino Americans. But once the partisanship kicks in, as we get closer to an election, they sort of like fall in line is almost like a crass way that they would describe what happens at the end of an election. Is there any reason to believe that things are different this time around?

Yeah. So I think I would sort of separate Black voters and Latino voters because I really see those trends very differently. Latino voters, this is part of a slow trend that has been occurring for the last three or four election cycles. If you look back at exit polls or Pew Research Center's validated voter study, you see that the numbers that we got for Hispanic voters, and we were looking at swing states, those are national. But

You look at those Hispanic voter numbers, and it's actually not that far off what we found in our poll compared to, for example, the 2022 or 2020 numbers that Republicans showed with Latino voters. So, you know, I think that there's reason no voter group is a monolith. And I think there's reason to believe that there is growth among Latino voters and interest in

Republican Party. And interestingly, I mean, I haven't we don't have the sample size to look at that now. But when I looked in the past, there's a really big college on college split there with a lot of the growth coming from non non-college Hispanic voters. Among black voters, I see it really differently because this is more of an abrupt jump. Not that there hasn't been some growth over the past few cycles, but

but not as sort of steady as it is among Latino voters. And that's one where it's hard to say if that's something that we will see stay through the election or if that's just expressing discontent right now. It's really hard to know. When I talk to voters and when we ask them about this over and over in the poll, the thing we heard primarily from black voters and particularly black men or black voters under 45 is,

is really, it's all about the economy, right? Like, can we channel the Clinton campaign? It's the economy, stupid. Like, that is what these voters are caring about.

And we see that black voters are voting for economic issues over social issues again and again. So for me, I sort of see those things differently. And for black voters, it's hard to say what will happen down the line. But I do think if, you know, sort of economic circumstances don't change or the perception of the economic circumstances not changing, there's reason to think that a lot of the black men that are leaning towards the Republican Party might stay that way.

I would add, though, that, you know, Ruth, I definitely don't doubt what the Times found. I think those are... You better not. I don't know. Kind of sounds like you do. Feel free, please. Yes. Those are sentiments that are being seen in, you know, a bunch of surveys elsewhere. But I also want to contrast that to election results that we have seen. And in 20... So in 2022 and in, you know, 2023...

We saw that Democrats, when they ran actual campaigns, which means not Louisiana, when they ran actual campaigns, they put up really good numbers with black voters in a lot of places. I think Mississippi, we had Brandon Presley at split ticket. When we calculated it, just doing some rough calculations and eyeballing, we had Brandon Presley winning 95% of the black voter, 96. He outperformed Biden very likely.

To me, that's interesting because at a time of super high polarization, yes, there's state-specific factors, but if black voters were – and if voters really are that disenchanted with everything at large –

You would expect to see somewhere some significant grounds of slippage also show in election results, I think. But we're not really seeing that, so to say. Catalyst had a What Happened in 2022 article, and there we saw the share that Democrats got from Black voters dip from 90 to 88.

And that's not really, you know, in line with what a lot of the polls say. The question I ask is, are Black voters moving away from Democrats or are Black voters upset with Biden and interested in Trump? And those are two really different questions, right? Like to look at election results, you might see Black voters generally sticking with the Democratic Party. But a lot of what we heard in our poll and when we talked to voters was it was sort of Biden specific about the economy, not the Democratic Party writ large, which I think is really interesting. Yeah, actually, Ruth,

One of those interesting nuggets I thought in your polling was comparing a Trump Biden matchup to generic Republican or generic Democrat or Nikki Haley. And what you see is that Nikki Haley performs significantly less well against voters of color compared with Trump. So it is a candidate specific thing. It's not just, oh, you know, voters of color are upset and now just want to vote for Republicans.

And I think there could be a couple of things going on here as well, that we're talking about a presidential election, not an off-year election or a midterm. And the kinds of voters that you're talking about, and not just voters of color, but all voters who aren't really sure who they're voting for, are deciding whether or not they might vote, are pretty marginal.

By definition, they don't feel super close to one party or the other. They may see both parties as kind of, it's all just politics. They're all kind of the same. Both parties are bad. That kind of attitude...

who aren't showing up to a gubernatorial election in Mississippi in 2023, or for that matter, a midterm election in Wisconsin. And so I think the question here is how different is the electorate? And does that mean that we can't use the elections of the past two years to inform what we think will happen in 2024? Come at me though. Bring me some counterpoints. Bring me some counterpoints. You know,

I hear all of your points, Galen, but also I step back and I look at this. It kind of sounds like you don't. Well, I hear them. It doesn't mean I agree with them. See, shots fired in the last. I just look at the bigger picture here in the last six,

seven to eight years, we've seen polarization, partisanship accelerate significantly in the electorate. In 2022, we saw like 97% of Trump voters vote for Republican candidates and 94% of Biden voters vote for Democratic candidates. So there's not that much crossover happening anyways. Turnout in the 22 midterms was not 2014 levels. Okay, it wasn't 2018 levels, but it was also, you know, not bad for a midterm. And so for me to think that like,

you know, non-white voters in particular are going to see 40 point slides to the right.

would indicate that basically almost every one of those voters that didn't vote in 2022, either they hate Biden and they like Democrats and they're fine voting for Democrats in general. But that would run counter to the stratification of partisanship that we have seen over the last six or seven years. So you get a little bit of an interesting thing there where the argument becomes that they're going to cross over a lot for Trump, but vote Democratic down ballot.

That's possible. I've seen really weird things happen. But again, it would run counter to a lot of what we've seen in the Trump era. And the second thing would be that maybe they just are just going to vote Republican en masse. And I think I would be more willing to subscribe to that theory if

it didn't imply that basically all of the surge voters were just disenchanted Republican curious voters who stayed home because they had an opportunity to make their point against the Democrats and against Biden, and they didn't show up then. So I think that might be a little of an exaggeration because obviously the other trend that we've seen in the Trump era is that Democrats do quite well in midterm and off year elections and that elections get awfully close when Trump is actually on the ballot. And, you know,

I think writing off that possibility this time around, especially when Trump is currently leading in state-level polling and national polling, also feels hard to believe. But let's get Ruth and Elliot in here. Who are you picking? Who's your favorite son? Is it Galen or Lakshya? Ruth. Split the baby? No, I mean, I think, you know, I think my opinion... You all made some good points. Yeah, exactly. Everyone gets a gold star. That's my line. Okay.

No, I think my opinion comes a little bit closer to Galen's here. You know, I think the Trump era really does scramble a lot of the traditional things that we think about partisan alignment and particularly when he's on the ballot. And like we certainly see that from the polling perspective, right, where our polling is extremely accurate when Trump is not on the ballot. When Trump is on the ballot, we have a little bit of a trickier time.

But CNN did a good national poll that came out a couple days ago that had, you know, let's take young people as our example, right? They had similar

high numbers for Trump among young people, but they also had a generic congressional ballot sort of buried down there in the bottom. And Democrats do remarkably well among young people for that generic ballot. So those are people who are making that choice to think about splitting their ticket. Caveat being it's one year out and split ticket voting per your website's name has gone down pretty dramatically. At the same time, I think people are

feeling that frustration, feeling that anger towards Biden. Is it expressive responding at this stage? Are they just trying to vent some of that frustration? Quite possibly. Like, I think we have to, you know, live with that being a possible reality. But I do think people are considering these options in terms of, you know, splitting their tickets.

Also, it's important to remember throughout all of the data that we look at for the coming year that when we say, oh, like split ticket voting has declined or, you know, it's small, it's marginal, whatever. Yes, as we become more partisan, all the numbers that we're talking about have become smaller.

the margins by which somebody wins the presidency are smaller. So that means that those little differences may honestly still be proportional, but are still very important. You know, split ticket voting made the difference in states like

New Hampshire, for example, in 2022, right? Voters are still thoughtful enough for you to see a 10-point gap in Pennsylvania as well, places where people were voting based on candidate quality.

So I think, yes, split ticket voting has gone down. But as polarization has gone up in general, the smaller numbers of split ticket voters may still have the same bearing on the election overall. I think a point worth emphasizing is that even on Election Day 2024, it is going to be impossible for us to identify the population of likely voters. That is an unmeasurable group. There's no population benchmark.

It is ephemeral. It does not exist. It comes into existence for a snapshot of like a few hours and then fizzles away. Can we put that on a banner somewhere? There's no way for us to know really like the truth about how that group voted in the past. And it's very hard for us to make guesses about what that's going to look like.

in the future. So I think it comes down to a rationalization here. Yes, in the post-Biden 2022-2023 elections, you haven't seen as much movement among Black voters in election outcomes as the polls anticipated. I think the final poll, for example, in Kentucky had Cameron winning about 18 percentage points of Black voters. It looks like that's probably closer to 11, for example. But

The group of ephemeral likely voter 2024 black voters is a group of lower turnout, like lower engagement people that we have, that we doubly have not observed since 2022. So to me, I think, and I hate to be the wishy-washy guy, but right, it seems totally plausible that there's been this shift among people that we haven't observed. At the same time, I think it's probably not as much as polls are anticipating.

Cool. We all agree. Great. We all agree. Excellent. Excellent.

Does it seem at this point like there is a clear path to somebody else winning the nomination? Or is that still a sort of pipe dream for Trump skeptical Republicans? No. Controversial opinion right there. I don't think there's a clear path. I think you can look at the tails and you see, you know, two or three candidates left and

When Iowa comes around and Trump doing 10 points worse than the polling average and the race tightens and it goes past Super Tuesday, but you get the same outcome.

I could also see going past Super Tuesday and the conviction start rolling in and you get a couple of really good polls for Nikki Haley on electability. And suddenly Republicans think maybe we can dump this guy and it's more important to us to beat Joe Biden than it is to stand by our man. Maybe. But these are tail outcomes, right? I don't think these are the likeliest things. Maybe a 75% chance, if I'm just guessing, of Trump winning a nomination today based on

the historic movement we've never gotten ourselves in trouble with those 75% chance I've never heard that before

No, I mean, I agree with Elliot. And if anything, I think the chances are even a little higher. I think he's being conservative. You know, even if you coalesced all of the other Republican candidates and they coalesced around one person, let's say Haley, it still, you know, wouldn't be enough to outpace Trump's support. Now, of course, there are tail outcomes that where, you know, there's plenty likelihood for them to happen. But the most likely outcome for sure is Trump getting the nomination, even with

you know, convictions rolling in. Because we know from our own polling that most Trump supporters, it doesn't bother them, like the vast majority of Trump supporters and Republicans writ large. All right. Final question here of the one year out data extravaganza.

is, we already mentioned that both candidates are historically unpopular. According to our averages, Biden is 15 points underwater, Trump is 14 points underwater. As an example of a comparable election, maybe 2016 was a race where both candidates were quite unpopular. Part of the reason that Trump won is because people who disliked both candidates ended up breaking towards Trump.

But what we also saw in 2016 was the highest levels of third party candidate voting or just leaving the top of the ballot blank in some time. It was either since 1992 or the second time that Ross Perot was on the ballot.

We have a lot of third party or independent candidates running in this race, and we may get more. So we have RFK Jr. We have Jill Stein running on behalf of the Green Party. We have Cornel West also running an independent campaign. Of course, folks will have heard that Joe Manchin just announced that he is retiring. There's some speculation that he will also retire.

try to make an independent or third party run. No labels have said that they're looking for somebody to put forward. Like, I don't want to be the Pollyanna-ish sort of like host here who's like, ooh, could this be the year that a, you know, a third party candidate actually wins the presidency? But how much does all of this matter? How does this all fit together? The number of people who are making third party or independent runs and the number of disaffected voters who dislike both candidates?

I can only speak to our data because I've been able to dig into it. But we found that even though RFK Jr. did very well in our poll, the vast majority of his supporters were planning to vote for another candidate. 40% said that they were definitely voting for Trump or definitely voting for Biden. And two thirds were definitely or probably voting for one of those candidates. So

You know, I think it's reasonable to think that a lot of that, I mean, and partially it's how we're asking these questions. We ask the main ballot test with just the two candidates. And then later we ask with, with RFK Jr. in the ballot. So sometimes people are letting off some steam, letting off some of that frustration, but

And so it's reasonable to skim off, let's say, 40 to 50 percent of that support right away from people who are just kind of venting their frustration that they're angry with both candidates. Having said that, I think it's interesting you talk about Perot and we think about Humphrey and other candidates in the past that have been successful as third party candidates with really high margins. And those were times where partisan ties weren't as strong as they were now and polarization wasn't as strong as it was now. And I think there's a lot of reason to believe that people are not going to

you know, jump to a third party candidate for worry that they would hurt their own party, which they feel tied to very strongly. Kennedy right now is hurting both candidates equally. If somebody like Manchin were to jump in that to the race, you might wonder if that would hurt Democrats more than Republicans. Yeah, I mean, I guess we don't have enough

of a history in the high polarization era to know how tactically partisans are thinking. But my very hard prior is that any one of these candidates can't get above 20%. If you have a bunch of them running, maybe you have a scenario where Joe Manchin gets 20, Kennedy gets 15, and Cornelius gets five or whatever, and you have both the other candidates at 30 or 35, and you have a very low percentage plurality presidential election winner,

You could also have a scenario where these candidates have strong home bases of support and that throws the election to the House of Representatives, whatever. But I don't see a scenario. And that's the chaos scenario. I don't see a scenario where one of them can actually win. And just to put a couple of numbers backing what Ruth found in the New York Times polling, in our average, we have Biden and Trump around 45% each. We have

Robert F. Kennedy Jr. at 12 and Cornel West at 5%. So there's a lot of talk about these people being popular, maybe being able to win the presidency. They're a sign of mass discontent with the major party nominees or whatever spin you want to put on it. They're still really far behind. I agree with Elliot and Ruth entirely here. I think

So there's a possibility that one of them or the combination of them could get a surprisingly high vote share compared to the past, especially compared to 2020. I'd be shocked if they didn't. But I think that speaks to discontent among the electorate. But also at the same time, when you look at the track record of third party candidates, I mean, I think 92 per row got 18 percent. And that was under one of the more low polarization times that we've had.

In 2016, Gary Johnson actually hit like 13% in polls sometimes and he ended up getting like 3%.

And I tend to think that what happens, right, is we just see over and over a year on out, people are just angry and they don't like either of these people. So they're going to say, I don't want these guys. But as time comes around, they're more likely to vote for one of their guys. And so, as Ruth said, you can really just skim off a lot of RFK support right now is just anger. And I think it would just crash like historically every third party run has done.

Can we do like a Sesame Street style? Like word of the day is expressive responding. Benching frustration. Today's podcast is sponsored by expressive response. Yeah, but it really does sound like the answer to

several of the questions that I've asked here, both on third party and on what we see in the swing states and in national polling is that it's too early to tell in part because everyone's just angry and hasn't had to make, hasn't had to come to terms yet with the fact of American elections are a choice between two parties. Will things change when that sets in?

that's the big question that we cannot answer today and we will have a year left to try to. With that, Ruth and Elliot, stick around. Alexia, we're going to let you go. I know you have a big product launch to get to. So we wish you could play a game with us, but next time. Thanks, guys.

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lipson Ads. Choose from hundreds of top podcasts offering host endorsements, or run a reproduced ad like this one across thousands of shows to reach your target audience with Lipson Ads. Go to LipsonAds.com now. That's L-I-B-S-Y-N-Ads.com.

All right, to wrap things up, we are going to play a one year out data installment of Guess What Americans Think. Folks are likely familiar with this game. Basically, I'm going to put out a polling question and y'all are going to have to choose the

The percentage of folks who answered it in a certain way, I mean, some of the answers are different, but you'll get it as we go on. The Price is Right rules, close without going over? No, you can go over. It's just closest. And I'm going to have you all write down your answer on a piece of paper and then show it to me so that you don't

affect each other's responses. We want this to be a rigorous, we want rigorous responses here. We don't, you know, can't go like polling people sitting next to their friends. I'll ask my dog. She's next to me. She'll know. First question is, according to an ABC News Ipsos poll, and this was a set of polling that came out about a week ago, a year before the presidential election, what percentage of Americans believe the country is headed in the wrong direction?

Three, two, one, reveal. 76. 80%.

Okay, this was, I'm wondering if this was unfair and that Elliot actually looked at all of this polling. Yeah, don't you guys work for ABC News? That's not what happened. Because it's, wait, because it is 76%. Elliot, is that what happened? No, I just guessed. I'm being really truthful. You're being totally honest? I'm being honest. I'm impressed. I bow down.

I would not cheat against Ruth. Wow. Because that's not even the average. The average is closer to 66. We're going to move on to the next question. By what margin do Americans trust Republicans more than Democrats on the economy?

Three, two, one, reveal. 40%. 22. Whoa. 40% or 40 points? Well, 40 points. Okay. 40 percentage points. Okay. Wow. Okay. Okay.

So according to this ABC poll, it's 10 points. Oh, wow. But 29% said they were undecided. You didn't have to choose one or the other. Oh, you could say. Well, see, you should have revealed that at the beginning. Yeah, response options matter, Galen. That's right. You're tilting the scales here. Well, I don't know who I was tilting them towards. Okay, that's two for Elliot. Okay.

Next question. Who's getting smoked now? Me. By what margin do Americans trust Democrats more than Republicans on the issue of abortion? And again, you can say neither. You have a neither option. Okay. Three, two, one. Reveal. 14. 20 percentage points. All right. And Ruth is on the board. It is actually 27 percentage points by which...

voters or Americans trust Democrats more on the issue of abortion. So it is two to one. You're making that comeback. Next question. What percentage of Hispanic Americans have a favorable view of Biden? This is again, according to ABC, I'm supposed to pull, but this is the last one from that poll. Then we're moving on. I feel like we shouldn't be thinking hard about it. So yeah.

I feel like the more I think, I'm going to get in trouble. All right. Three, two, one. Reveal. 38. 40%. Whoa. Okay. With some colorful drawings from my Fabulous. It better not be 39. It's 33%. Ah. So, Elliot, you got that one barely. Yeah.

And in that same poll, 49% of Black Americans had a favorable view of Biden. So some really low numbers there amongst groups of voters who historically have voted for Democrats. All right, next question. Elliot3Ruth1, according to FiveThirtyEight's averages, who has a higher net favorability amongst Republicans? Tim Scott, Nikki Haley, or Vivek Ramaswamy?

I think I'm wrong on this one, but... Well, then maybe you should change your mind. Well, I don't want to second guess myself. All right. Three, two, one. Reveal. Nikki Haley. Haley.

Is it Vivek? It is Tim Scott. See, I thought about Scott, but I thought that the high DK might get in his way. No one gets that. It's still three to one. All right, next question. According to a Quinnipiac poll, what percentage of Republican primary voters say they are firmly set on their choice for the Republican nomination, no matter what happens leading up to the Republican primary? For their primary choice or their general choice? Primary. Three, two, one.

62. 80%. It's 46%. Whoa. 52% say they might change their candidate choice depending on what happens leading up to the Republican primary. We were both off. That's four for you. I'm curious, yeah, why you thought it was so high. Because people are pretty baked into their choices. I feel like it was higher among other polls, probably. Yeah. It feels like it's definitely been higher. Like some chance of changing your vote has been pretty low. Yeah, Galen, you picked an outlier to make Ruth look bad. Yeah.

Slammy. I'm making myself look bad all by myself. I'm only technically, I'm only getting technical points for this. Yeah, Elliot's been very far off. What percentage of voters say they are more motivated to vote than in past presidential elections? This is across everybody, not just Republicans. More motivated? Yeah. And the options are more, less, about the same, not sure? About the same?

More, less, just as motivated as usual. All right, reveal. 27. 15%. This is wildly wrong again. This is an interesting bunch of polls. It's 48%. No.

Of Americans who say they're more motivated than in past presidential elections, 9% say they're less motivated and 42% say they're just as motivated as usual. Republicans were the most enthusiastic, according to this poll, with 58% saying they're more motivated to vote in next year's presidential election.

It's funny because when we compare our polling to previous polling, enthusiasm is down. Even among Republicans who are more enthusiastic, enthusiasm is down. So what's going on there? It's just expressive response. People are like, I should be enthusiastic to vote in the presidential election because it's my civic duty. Social desirability, maybe. Yeah, social desirability. And people also forget how they felt a year ago. Like, it's hard to remember exactly how you felt. And so you might feel enthusiastic now and say, like, sure, it's more enthusiastic than I was before. But in reality, it's less.

All right, final question here. From a CBS News YouGov poll, what percentage of Americans say a Trump presidency would decrease the likelihood of the U.S. being in a war?

Decrease, increase, no change. Are those the options? Yes. All right. Three, two, one. Reveal. 30%. Whoa! Look at that. You both said 30%. That is beautiful. Not kidding. All right. Well, I'm going to give you both the opportunity to guess again. Okay.

Because it was so far off. Well, no, it's not that it's so far off. It's just that there's no distinguishing between the two. There's no distinguishing between the two. We broke at a point in time. Well, but here's the thing. Secretly, this question is worth seven points. So Ruth, you could really make a comeback here. I've revised. I've revised. Okay, hold on. I need to revise.

Three, two, one, reveal. 35. 37%. And Ruth wins the entire game. Turned it around. I can't believe I did so poorly. I know, it's embarrassing, really. 43% of Americans say a Trump presidency would decrease the likelihood of the U.S. being a war. 23% say a Biden presidency would decrease the likelihood of being in a war.

For what it's worth, I heard that from a lot of Democrats. I mean, we didn't ask this question on our poll, but when I called them back, a lot of people were like, you know, we wouldn't be in these wars if Trump was president. Ruth, how many callbacks do you do? I do a bunch. I like doing callbacks. I think they're a lot of fun. For listeners, we can call back people from our poll who agree to be recontacted and ask them more questions about why they gave the answers they gave. And I probably did, you know.

15 or 20 this cycle. They're a lot of fun. I think it's something we should do more. We should do it. And you said you heard from Democrats? Yeah, I heard from Democrats who said if Trump was in charge, we wouldn't be in these wars, which I think, one, we aren't in any wars. But also, it's kind of fascinating that this is a sentiment that spans across the aisle.

All right. Well, congratulations, Ruth, on winning the latest installment of Guess What Americans Think. Your trophy is in the mail. Final thoughts here before we wrap up on wisdom you would like to impart on the crowd one year before Christmas.

the 2024 presidential election. Maybe even just, you know, sort of advice for how to stay sane, you know, sleeping, exercise habits, really anything you may have. Read more polls, I think. No. Honestly, I mean... Refresh538.com at least seven times a day. An hour. That's a mental health suggestion. My mental health suggestion to you.

No, I mean, I think a lot can change in a year. But also, I think a lot of people who find these results to be scary or frustrating are trying to tune it out. And I think it's worth paying attention to not the top line numbers, but the underlying findings in these polls to show the mood of the electorate and what very well could happen if things don't change. Like I would say, you know, this doesn't mean a lot, but don't ignore it.

I'll say this is not how campaigns think about polls. The horse race so early is of such little utility as we said that the thing they want to do is understand the issue landscape. I think that's something that the New York Times does a really good job at and understand what issues might be vulnerable six months in ahead. There's some amount of forecasting of issues and events that might happen in the future that they are pulling about privately.

So not only, right, the recommendation, not only don't put so much stock in the horse race polling, but like there's a lot of qualitative indicators too that we might pay attention to aside from issue polling. Yes, very much agree. Yeah, I think that point about however you may feel about these polls is

It's not personal. These polls are not trying to make you angry or tell you something that you don't want to hear or hurt you. They're just trying to reflect

the world as it is. So if they're making you angry, don't get mad at the polls or the pollsters. Your emotional response should be reflected towards the world. Although I don't want people to be angry at the world either. Maybe just be open-minded to the possibility that the world is not as you thought it was. It may be that this data is reflecting the world as it is and not as you wish it would be.

Most people live in bubbles with their friends who tend to agree with them. That's something that happens more and more with increased polarization and partisanship. And so it's good to hear what other people who don't agree with you might be thinking. Yeah. Polls are just another way of talking to people. Polls are people too. Polls are people too. That's a good... They're corporate entities that are allowed to donate to political campaigns also. Yeah.

And with that, Elliot and Ruth, thank you so much for joining me today. Thanks for having me. So much fun. Thanks. My name is Galen Droop. Tony Chow is in the control room. Our producers are Shane McKeon and Cameron Chertavian. And our intern is Jayla Everett. You can get in touch by emailing us at podcasts at 538.com. You can also, of course, tweet at us with any questions or comments. If you're a fan of the show, leave us a rating or review in the Apple Podcast Store or tell someone about us. Thanks for listening. And we'll see you soon. Bye.