Good Or Bad Use Of Polling (Taylor's Version)
46:08 Share

You're a podcast listener, and this is a podcast ad. Reach great listeners like yourself with podcast advertising from Lips and 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 Lips and Ads. Go to Lipsandads.com now. That's L-I-B-S-Y-N-ads.com. Toby. I see Toby. I know. I want him to come on my lap so he doesn't cause a ruckus during this.

Hey, baby. Little angel. Oh, there he is. Although, won't he just like cause more of a ruckus if he's in your lap? Back right into the microphone. Now he's got the zoomies. Can you just take Toby's collar off? And then you still hear him like squeak. What if he gets lost? Okay, he'll just have to be in the background.

Hello and welcome to the FiveThirtyEight Politics Podcast. I'm Galen Druk. In the midst of news about the possible indictment of a former president, some of the biggest news in the country has gone unmentioned on this podcast. Yes, of course, I'm talking about pop icon Taylor Swift's Arrows Tour.

When 14 million people tried to buy pre-sale tickets at the same time last year, Ticketmaster broke to the point that Congress felt the need to intervene. And on her opening night, Swift broke Madonna's record for best attended female concert in U.S. history. 69,000 people gathered in Glendale, Arizona.

Pollsters have taken note and have tried to use surveys to determine who exactly Taylor's fans are and which of her 10 albums is best regarded. And so today, we're going to ask one of our favorite questions about one of America's favorite musicians, good or bad use of polling.

Then, womp womp, we are going to take a hard turn back into electoral politics. As the informal primary contest between former President Trump and Florida Governor Ron DeSantis heats up, it appears that Republican primary voters have cooled a little on DeSantis.

But we're going to ask a slightly different question today. Who do Democrats want the GOP nominee to be? And what does that tell us about how they're thinking about 2024? Here with me to discuss our politics reporter, Alex Samuels. Hey, Alex. Hey, Galen. Also here with us is senior elections analyst, Nathaniel Rakich. Welcome to the podcast, Nathaniel. Thanks, Galen. It's great to be here. It's great to have you. And also with us is politics and tech reporter, Kayleigh Rogers. Welcome, Kayleigh. Hi. Hi.

Just two points of order before we get going. One, of course, we are still on indictment watch. If it happens today after we've recorded this podcast, my apologies. We're probably still going to publish this podcast and then we'll talk about that, you know, possible indictment later in the week.

probably on Tuesday, but we'll just generally be on watch again. So tune in if you hear the news and we will have an emergency podcast for you. I'll also mention again that we have a live podcast coming up on April 19th in New York City at the Bell House. You can get your tickets at 538.com slash live show. That's April 19th at the Bell House in Brooklyn in New York City. It's at 730 and get your tickets at 538.com slash live show.

Okay, let's talk about Taylor Swift. Hard news today on the FiveThirtyEight Politics podcast. But in reality, the polling that these pollsters conducted on Taylor Swift can teach us something about polling in general. So Nathaniel, begin by putting this all in context for us. How does Swift rank amongst America's favorite musicians?

Long story short, she is very popular, but not the most popular musician in the country. So Morning Consult asked about 23 different musicians. Why 23 and not 20 or 24? I can't tell you. But basically they asked if people were an avid fan, a casual fan, or not a fan of these musicians. And Taylor Swift ranked 6th.

Which, honestly, I was surprised that she ranked that low. But first place was Adele with 61% of people saying they were fans of her. Then it was Lady Gaga, Bruno Mars, Rihanna, Katy Perry, and then Taylor Swift. Wow. Harsh. Yeah. 53% of Americans said that they were an avid or casual fan of her, whereas 47% said they weren't a fan.

Okay. Is anyone else surprised by this? I'd expect Katy Perry to be number five. Yeah, Katy Perry beating Taylor Swift is very interesting there. What's going on? Analysis, please. I will say that is within the margin of error. Katy Perry has 54% and Taylor Swift has 53%. So let's, you know, let's pump the brakes a little bit. I mean, we're going to get into this, but like asking people to define themselves as an avid fan just feels fraught.

on its own versus a casual fan versus not a fan. You know, there's sort of different stratospheres of actual fandom. And I feel like someone might say not to be the police over fandom, but someone might say they're an avid fan when they're maybe more of a casual fan for any of these artists. Hmm. Okay. So one of the most

interesting, I would say, aspects of this polling was how, in particular, Morning Consul tried to define the demographic characteristics of a Taylor Swift fan. And the headline that came from that report was, the Taylor Swift fandom is white, suburban, and leans Democratic. Now, you may be thinking to yourself, tell me something I don't know. But when you actually look at the data that they're using to come to this conclusion, it's

The conclusion is actually questionable. Nathaniel, you wrote about this actually for the website. So again, we're going to go to you first on this. Is that a good or bad use of polling? So I think that headline specifically is a bad use of polling. The way they did this was a little strange. So they, you know, I think normally we're used to looking at like

there are two ways to look at this, right? So you can, you can say like what percent of white people of suburban people are Taylor Swift fans, or you can look at the kind of racial demographic breakdown of Swift fans itself. And usually when we talk about like polls and crosstabs, we're looking at like, you know, what percent of, of white people and suburban people answered a certain way, but morning consult based their headline on their breakdown of just looking at Swift fans as the denominator. And so,

Yes, most Taylor Swift fans are white, but that's because most Americans are white. Yes, most Taylor Swift fans are suburban, and that's because most Americans are suburban. So that didn't really tell us much of anything. I think you need to look at the actual traditional polling crosstabs in order to tease out some interesting patterns there.

To Nathaniel's point, I agree. You know, yes, maybe Taylor Swift fans tend to skew white, but they're not that much whiter than the country as a whole, as Nathaniel got at. I was confused whether Morning Consult asked about Latinos, because if they did, they were not included in that poll. And that just seems like a very large demographic group to exclude. And I'm not sure what the rationale was behind that.

But even their numbers for black and Asian listeners were pretty on par with their respective census numbers. Yeah, just to add in the data here. So when they asked about race, it was 74% white, 13% black, 9% Asian, 4% other. So if they didn't break out

Hispanic Americans, then that 74% white number basically matches what the census shows, which is about 70%. Black Americans are about 12 to 13% of the population. Asian Americans are about 5% of the population. So if they're indeed 9%, you know, maybe Asians are more overrepresented than any other group. That's an interesting conclusion in and of itself. But they also asked about, you know, gender, age, income, urban, suburban, political party. Did we learn anything from those categorizations?

Yeah. So some of these were interesting and some of them were less interesting. I think probably the most interesting was, you know, this is a politics podcast and the partisan breakdown of these numbers. So, for example, if you look at Democrats and I'm kind of going by the traditional crosstab approach.

situation now. So 62% of Democrats said that they were fans of Taylor Swift, whereas only 48% of Republicans did. And this is interesting because I actually found a poll from back in 2010 asking about Taylor Swift, not straight up like, are you a fan of her or not? But it was actually kind of an interesting poll. It was from CBS News, 60 Minutes and Vanity Fair. And it asked, which of the following music performers would you like to have dinner with? Would you most like to have dinner with?

Paul McCartney, Lady Gaga, Susan Boyle, remember her, Taylor Swift, or Jay-Z. Hold on, hold on. Paul McCartney, Susan Boyle? Yes. It was 2010, Galen. It was a different time. She's a very good singer. Oh, to say nothing about Susan Boyle's capacity as a singer, more just like...

What? Okay, anyway. Yeah, well, it's actually very interesting, right? Because, you know, also Jay-Z is on this list. So some interesting, you know, it was definitely a moment in time.

but so among, um, all respondents for this poll, um, most people wanted to have dinner with Paul McCartney, 33%. Taylor Swift was in second with 22%. But among Republican respondents, Taylor Swift was actually in first place, although it was within the margin of error. She had 30% of Republicans said they wanted to have dinner with her. 28% wanted to have dinner with Paul McCartney. So this is interesting because of course, back in 2010, she was still more of a country singer, right? And, and of course, country music is, um,

Listeners are disproportionately Republican. But since I think around 2018, when she actually endorsed the Democratic candidate for Senate in Tennessee, her home state, Swift has become more identified with liberal causes. And I think you've probably seen things flip in that regard. And plus, obviously, she's become kind of more of a mainstream pop performer. And so it's interesting that you have seen this kind of shift in her fandom from being predominantly Republican to being predominantly Democratic today.

I mean, look, my problem with this polling overall is, you know, these are subsets of a subset of a subset, even within her fandom. You know, it's only looking at Americans, first of all. This is a global performer who has fans all over the world. Oh. And then within that, you know, they're doing... Oh, the Canadian is sticking up for the non-American Taylor Swift fans. Okay, you know. Yeah.

Just a little self-serving there, but okay. But it's true. It's true. She has fans all over the world. So we're already only looking at a slice of it by focusing on American polling. Kaylee knows this all too well. I do. Don't get me started with the puns. We made it this far into the podcast and we only got one Taylor Swift pun out? No, I made one earlier. Oh, I missed it. It must have been an Easter egg. Was it the number 13? It was an Easter egg.

Good Lord. Sorry, Kaylee. We'll let you actually speak now. Oh, good one. I hear it. I hear it. Okay. Focusing only on Americans, it's already a subset of her fans. And then within these polls, it's American adults at large. So then they have to find within that the subset of people who either self-identify or in some other way indicate that they are fans. Again, someone just saying they're a fan, I don't know how much that actually...

with the YouGov survey, first they asked if you were familiar with songs by Taylor Swift, and then among people who had at least some familiarity, which I don't know where they cut that off, was people who said not very familiar? Did they have some familiarity? I don't know. They asked them about if they loved or liked each of her albums, and then if they loved or liked at least one of them, then they asked for their favorite albums. So we're talking about like a really slim range

actual proportion of people. And so I feel like you just can't draw a lot of conclusions of that, especially with the self-reporting. You know, the people who say they like or love whatever album you're tossing at them, like, when was the last time they listened to it? I gotta ask. Personally? I'm sorry, are you gatekeeping Taylor Swift? She is gatekeeping. Hardcore.

I'm just saying, if you want the data, you should go to the best source, which is the actual fandom. And every year, the Taylor Swift subreddit does a census, collects data. For 2022, they had respondents, more than 11,000 respondents.

They went through the data and removed the trolls, which is important to know. I'm curious, what's the scientific method for removing trolls from data? We'll have to ask the subreddit moderators about that question. But interestingly, the data was different than what we find with these public surveys.

Gatekeeping aside, I think that's a fair point. We always have to think about who's actually responding to a poll. In many cases, we do rely on self-response. But when we're talking about elections, we try to verify past voting behavior, location, geography, all that kind of stuff. And in these polls, we don't. We just asked, morning council just asked, are you suburban? Are you urban? Whatever. Alex, did you have any thoughts on any of these counter organizations before we do move on?

My understanding was that millennials are pretty high in the survey, at least compared to their proportion of the U.S. electorate. So millennials, I think, are around like 22 percent of the U.S. electorate. But then in the poll, it was 45 percent.

I was kind of surprised by that. Maybe as someone who's not super familiar with Swift and her music, I would just assume that more Gen Zers, millennials too, but I mean, I figured Gen Z would have higher numbers. I was kind of surprised that millennial was that high. I think it's because Taylor Swift is a millennial. So a lot of her fan base has kind of grown up with her.

And are, you know, roughly the same age. So I wasn't surprised. I feel like Gen Z has their own artists. You know, this is kind of like their older sister or their potentially mom's artist rather than Gen Z's. Wait, I don't think Gen Z is that young compared to millennials. Yeah. No, but I mean some, sorry, I mean like some boomers and Gen Xers are also fans of Taylor Swift.

Yeah, in fact, Gen Z was the least likely to be Taylor Swift fans. So only 50% – it's not a huge difference, but only 50% of Gen Zers identified as Taylor Swift fans, 51% of boomers, 52% of Gen Xers, and 59% of millennials. But yeah, overall, it definitely seems like her sweet spot is in Gen Z.

the people who are in their 30s now. But when you look at the question from the perspective of the people who say they're fans, what percent are millennials? It was 45% in this morning consult polling, where it was like 11% were Gen Z, 21% Gen X, 23% boomers. And when I looked at this, honestly, my thought was, the way that they headlined this was, the Taylor Swift fandom is white suburban and leans Democratic. Really, the way they should have headlined it is, the Taylor Swift fandom

fandom is millennial because we already went over race. But when you look at like urban suburban rural, it's crazy how accurately her fandom actually tracks with the geography of the country. It's like almost one for one. So 55% of Americans live in suburban counties. 53% of her fans are suburban. 26% of her fans are urban in America. It's 31% of Americans who live in urban counties. Um,

Again, 21% of her fandom is rural, 14% of Americans live in rural counties. So actually most overrepresented here is people who live in rural counties. And then when you look at the income, households making under $50,000 are the most overrepresented in her fandom. Like 50% of her fandom makes less than $50,000.

So if you look at Leans Democratic is relatively low income and very millennial, what do you get? Oh, surprise. Her fans are young people. That's what this poll told us. That's all this poll told us.

Well, the interesting thing about income, Galen, I thought that was actually one of the most misleading parts of the poll, and thankfully it wasn't in the headline. But if you look again at the crosstabs, of the people who make $100,000 or more, 61% are Taylor Swift fans, whereas the under 50K group, only 51% are. So actually her fans are disproportionately wealthy, which you would not get if you just looked at the breakdown, as you mentioned, of her fans. And you have to be to be able to afford to go to her shows. But I'm sure.

It's a good point, Kaylee. Okay, lest listeners thought we were done with this topic, we have one more important controversial area to get to, which is her fans' favorite album. Depending on which data you look at between Morning Console and YouGov, it is either 1989 Fearless Lover or Red. That's the polling version of a shruggy emoji. What's going on here? Like, can we hashtag trust the polls?

I mean, there's a couple of issues here. As I mentioned already, there's the fact that this is a subset of a subset that's actually being asked this question. So you don't have a lot of actual people responding to it.

Also, the difference between each of the... They were all kind of fairly evenly distributed as far as which of the albums is their favorite. For the YouGov poll, Lover won with 15%. 1989 was 13%. And then Debut, Red, and I Don't Know, which...

I'm sorry, this is supposed to be fans who love Taylor Swift and they don't know which is their favorite album. What if you're torn? What if you can't choose? All had 11%. So these are all pretty closely within one another and pretty close, I believe, to the margin of error. 2.4% was the margin of error. So it gets really muddy when you actually look at the breakdown. It's not like there was any kind of clear winner there.

Yeah, if you compare the YouGov numbers and the Morning Consult numbers among avid fans, the numbers are within the margin of error for pretty much every album, except for Lover, which is interesting. YouGov said 15% of people who said they liked one of her albums picked Lover as their favorite, but in Morning Consult it was only 5%. So I'm not sure what to make of that. But yeah, in general, the margin of error can account for a lot of this.

But Morning Console and Kaylee, I think maybe you got at this, they asked about the Taylor's versions of Fearless and Red. And like, again, I say this as someone who doesn't fully understand, but...

Are there like any big differences between the first version versus Taylor's version? Like if they're the same songs, I don't understand why they would break that apart as two different answers. Should we add a little bit of context here? Because of for like legal reasons and sort of reclaiming the narrative and money behind these albums, Taylor Swift tried to acquire the rights to her early albums. She wasn't able to, or at least it got complicated. So she re-recorded some of those albums and

And in this poll, the Morning Consult one separated out the original albums and the ones that she re-recorded. So it'll say like favorite album. Is it Red or Red Taylor's version? The one that was re-recorded. Do we think it's fair to break out original album versus re-recorded one? Having listened to them, they sound basically exactly the same.

I mean, there are some differences. There are vault tracks that she's added to the re-releases. I think the real information you're getting from that is that true fans would probably pick, if given the option, Taylor's version because they support her as an artist and want her to have that right to her masters, which she doesn't have with the original albums.

Many fans actually probably prefer, in certain cases, the original because we have a history with it and a fondness for those original recordings, especially some of her older stuff. She was younger. Her voice was just different. And these updated versions are just different than that. But there's sort of a political...

view among fans that you ought to be listening, you ought to be streaming and buying and paying for Taylor's version of the songs and albums when they're available in order to support her in this effort that she's making. And sort of casual fans maybe wouldn't understand that nuance or really care about it.

Yeah, and the YouGov poll asked about those albums together, which personally would have been my preference. But you do have to remember that when comparing the two. So for example, YouGov, 10% said they liked Fearless the best, one of the two versions of Fearless. And in Morning Console, it was 12% for the original, but then 7% said Fearless Taylor's version. So actually it's 10% versus 19%, which is a more significant difference, though still within the margin of error. So taking all of this under consideration,

Can anyone look at this data and make some kind of claim as to which album is Taylor Swift's preferred across all 10?

The thing with these surveys is when they came out, there was a lot of buzz among Swifty, Twitter. Just surprise. A lot of these results were not intuitive or what we would expect as far as the understood ranking. Because say it with me, Kaylee, Taylor Swift's best album is 3, 2, 1, 1989. However, when you look at the actual source, so for example, the Taylor Swift subreddit does an annual census every year.

Their 2022 results had more than 11,000 respondents. For them, they did two versions. They said before Midnight's came out, what was your favorite album? After Midnight's has come out, what was your favorite album?

Before Midnight's Folklore was the winner by a long shot, 26% of respondents said Folklore was their favorite. 17% said Evermore was their favorite. Those two albums ranked really lowly on both the Morning Consult and the YouGov polls. To me, that's the true fan favorites are the work that she did during the pandemic.

I don't know, Kaylee. I think, you know, this does get to an interesting question of, right, who is a real fan and who gets to decide this. But, you know, give me a scientific survey any day over people who are on Reddit choosing to spend their time ranking Taylor Swift albums. Yeah, I feel like that group is not going to be super representative. Okay, so if you look at the polling that was actually done, Nathaniel, I can't tell what the answer is. Can you?

No, I mean, I think you do have a tie at the top roughly within the margin of error between the likes of 1989, Red, the debut album, and Fearless. And I think the tie should go to the preferences of the people on this podcast. So 1989. 1989 it is.

Obviously, we have talked longer about Taylor Swift than probably most listeners of this podcast would expect. But, you know, we got to indulge sometimes. Did we learn anything about the polling industry in general from this exercise, Nathaniel? Bring it home. Give me an excuse for having spent 20 plus minutes on Taylor Swift on the FiveThirtyEightPolitics podcast. Yeah.

Yeah, I mean, I think that people who listen to this podcast probably know all the lessons that we can learn from this, but pretty basic stuff like that margins of error matter and the fact that 1989 was first in the Morning Console poll, but Lover was first in the YouGov poll is easily explainable by the fact that you get different samples and that

polls are going to be off by a few percentage points. It also matters who you ask about. So these kind of self-identified avid fans in Morning Console versus, as Kaylee mentioned, the people who say that they like one of her albums in YouGov versus the people in the Taylor Swift subreddit, right? All these people are different populations and they're going to give different answers.

And then finally, the question of, I think, whether you ask about the Taylor's version separately can make a difference. And that just kind of gets to the question of how you ask a polling question matters and can get you a different result. And it doesn't mean that any of these polls are inaccurate or wrong.

You know, and that one is better than the other. It's just a different way of asking different methodologies, different audiences that they are surveying. And you should kind of consider all of them and keep them in mind when kind of making your judgment, whether it's about Taylor Swift's best album or whether it's about who's winning a political election. Amen, Nathaniel. And with that, let's talk about political polls.

who Democrats would prefer between Donald Trump and Ron DeSantis in 2024.

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.

Earlier this month, FiveThirtyEight's own Nate Silver took to Twitter to poll Democrats on who they want the Republican nominee in 2024 to be. After more than seven years of Trump being absolutely anathema to Democrats, would they pick anyone but him?

Well, 230,000 people or bots voted and the results came back showing that quote unquote Democrats chose Trump over DeSantis by a four to three ratio. The poll attracted some attention, including from Elon Musk himself, who wrote slightly surprised by this result. More importantly, it got some attention from you, dear listeners. You were curious to know if this was a good or bad use of polling and you let me know about it. So we're going to talk about it today.

Nate himself suggested, you know, that people keep in mind that this was a self-selecting Twitter poll. In fact, there was nothing we could even, you can even do on Twitter to make sure that it's actual Democrats responding or Americans even responding or people who are eligible to vote or actual human beings. But we wanted to know if there are other ways to get to the bottom of this question. And so that's what we're going to do. Nate is tied up today, but Alex, you took the lead on getting to the bottom of this.

What did you find when we maybe put Twitter aside for a second? Who do Democrats want the Republican nominee to be in 2024? And like, does it actually matter? Like, why does this even matter?

To try and get to an answer to Nate's question, I looked at a series of polls from The Economist and YouGov. And late last year, they conducted at least three surveys that asked U.S. adults the same question. And that question was, if you had to choose, who would you rather see as the Republican nominee for president in 2024? And in all three surveys among Democrats, DeSantis had an edge over Trump, particularly when respondents were only given...

Trump and DeSantis as the options. Now, it wasn't necessarily overwhelming support for DeSantis over Trump. In fact, a plurality of Democrats in all three surveys said they were not sure. But I did think it was interesting that still between the two, DeSantis had an edge. Is there any other way we can go about getting to the bottom of this? You know,

I guess in political years, last December was like quite a bit ago. DeSantis was also more popular amongst Republicans even back in December than he appears to be now in primary polling. Was there other data you could look at?

Well, we have our own average of hypothetical head-to-head matchups of both Trump and DeSantis against Biden. And what we found is that DeSantis has a slightly better chance than Trump in a general election. So the question then becomes, you know, if Democrats do prefer DeSantis, is it that they want their party to face a tougher fight next year? And we kind of explored whether that was the case and why it might be that Democrats prefer DeSantis, at least in this

So what's the point of asking this question? Like, why? I mean, unfortunately, Nate's not here. But like, why does it matter who Democrats want to be the nominee in 2024? Yeah.

It doesn't really, but it's an interesting question, right? It's a question that I have wondered, and I thought Alex's article was interesting, right? Like, do Democrats view DeSantis as, you know, like Trump 2.0? Like, you know, a Trump who will continue, or yeah, a version of Trump who will continue to carry out his policies and maybe be more effective at it? Or do they view him as

a black box, which seems to be more of the case. So do they view him as more electable than Trump, less electable? It's just kind of an interesting question, but it is academic. Democrats will largely not have a say here. They obviously cannot make their Republican electorate vote a certain way unless they live in an open primary state and want to take advantage of that, which is very rare. So yes, they're not going to have a ton of say in the process, but I did think it was an interesting question.

Yeah, we okay to that point, Alex, like the little piece of this that might not be academic, say there is no competitive primary on the Democratic side, and the only competitive primary is on the Republican side. Historically, have we seen Democrats in places like New Hampshire go and vote in the Republican primary for the candidate of their choice?

This idea of party rating is not a common practice based on the research that I have and the understanding that I have of it. There was this pretty interesting 2019 study by a researcher at Marquette University who looked at voting behavior in Wisconsin.

And what he found using data from both 2016 and 2020 was that an identical share of Republicans and Democrats, and it was just 2% each, said they planned to vote in the other party's primary. So even if there is this tiny share of people who were indeed party-rating, they canceled each other out. But there was also no evidence based off his research that these people weren't voting in good faith. Like it's possible that some share of self-identified Democrats voted

genuinely preferred one of the Republican primary candidates and vice versa. Yeah, I think you can also, there's little bits of data in Colorado, for example, once they changed it to be, basically, if you're an independent, you can choose which party's primary you're going to vote in that year. The percent of independents started to go up in that state, which does suggest that there was some amount of either party that was like, well, I want to take advantage of this ability to vote. But

That could also be just people that are kind of in the middle and maybe typically leaned one way, but are like, sometimes I would rather have more say on the other party and it's not necessarily nefarious in trying to pick the least electable candidate or something like that.

The other really, I feel like it's such a small, like the type of person that is politically engaged and motivated enough to vote in a primary already is kind of narrow. And then to also be so engaged and motivated that you're willing to vote in the other party's primary to try to like sabotage it in some way. It's just such a small percentage. It seems really unlikely that it would have a significant sway in any given race.

Yeah, our colleague Jeffrey Skelly looked into this in the context of Liz Cheney's primary in Wyoming and basically asked, like, you know, if Democrats join, you know, vote in the Republican primary, can they save her? And he found that the impact would be minimal, even if it happened.

Um, so like generally the it's like, I think Rush Limbaugh called it operation chaos. Um, this kind of cross party sabotaging efforts. It's really, you know, a, a, a non-factor, uh, in these primary races, I think in, in a state like New Hampshire, New Hampshire is kind of the canonical open primary state where you can see like independence come in and influence one party's primaries, but it's still generally like independence, right? And New Hampshire, of course, is a state with a lot of independent voters.

It's people who would, I think, in some universe be open to voting for that party if that party's nominee was the person that they wanted. But, you know, just anecdotally, if you talk to self-identified Democrats, I found that they have no interest in voting in Republican primaries because they just like find it.

very hard for them to fill in a bubble next to Donald Trump's name or Ron DeSantis' name or whoever because it's just it goes counter again to their identity. Like people, you know, these days identify with parties more strongly than they ever have potentially. And, you know, and that can be a strong deterrent.

Yeah, I think the idea of the other party brigading the opposite party's primaries is kind of a political boogeyman. It just like isn't really something we need to be worried about.

Okay, so then now that we've answered that question, like it could have some impact around the margins and say New Hampshire or something like that, but not too much. Then what do we make of this? Like why in the actual YouGov poll that you mentioned, Alex, Democrats preferring DeSantis? What do we chalk that up to? Is it because they see him as genuinely more moderate and are thinking like,

okay, if you put a gun to my head and I had to choose Donald Trump or DeSantis to be president, I would choose DeSantis. Or is it because they think he's easier to beat or what? Yeah, so the poll didn't really get into the whys. So that's where my reporting tried to get it in and guess those things. I would assume that Democrats now maybe aren't aware of DeSantis' current formidability against Biden and maybe think,

of DeSantis as more beatable than Trump, who scared everyone by pulling off that 2016 impossible win. So they might think DeSantis is a better alternative. Now, as one expert told me, if you are a voter who thinks that Trump is uniquely anti-democratic or uniquely a threat to American democracy,

then sure, maybe DeSantis is the alternative that you prefer, especially if he's the only other person that pollsters are asking voters about. So I think at this stage, if Democrats really do prefer DeSantis, it could be more about their dislike to Trump versus anything special that DeSantis is doing at this point. Yeah, and I don't think that we can really...

draw great conclusions by what Democrats think of a candidate, even if we get that why, even if they were to say, well, I think DeSantis is easier to beat. That's why I prefer him. That doesn't mean that DeSantis is actually easier to beat. The Democratic perspective on that is going to be very different than the type of the Republican or maybe some independents who would actually be voting for one or the other candidate.

Yeah, I was going to ask, should we take Democratic preferences for a nominee as a negative indicator of who the nominee will actually be? Like, if Democrats prefer DeSantis, should we sort of think of that as a suggestion that it's less likely to be DeSantis?

I guess I'm just thinking, as Alex mentioned, in 2016, there was a certain segment of Democrats that were kind of on board with Trump being the nominee because they thought, well, this guy's never going to win. So make him the nominee. Great. Now we're a shoo-in with Clinton. And obviously that didn't turn out to be true.

Yeah, it's an interesting question, Galen. I feel like maybe before 2016, that may have been true because I don't have data on this, but I would imagine the Democrats preferred John McCain over George W. Bush in 2000. I would imagine they preferred somebody like John Huntsman in 2012.

And obviously those people didn't end up winning because obviously Republican electorates tend to go for more conservative candidates and a Democratic electorate would tend to go for the more liberal one. But I do think that Trump...

somewhat shook up the the kind of the clear left right divide within the party right it's not clear who's more conservative trump or desantis um like it's not just the left right it's the you know the trumpy versus not trumpy scale the kind of norm breaking versus non-norm breaking scale so i'm not sure it it really tells us much of anything these days

Hmm. When Nate did his Twitter poll and found the opposite result of what the YouGov polling from late last year suggested, and, you know, this four to three ratio in favor of Trump, he theorized that possibly Democrats may think that Trump is easier to beat. That's why they could prefer him or that Democrats think DeSantis would be more effective if elected.

Or maybe in addition to all of these, that DeSantis is more conservative on some issues, for example, entitlement cuts and gay and trans rights. He also theorized that DeSantis is very online, you know, picking fights that generate a lot of heat and that this is a Twitter poll so that perhaps online, you know, Twitter users, Twitter voters may be more keyed into that and might have more of a dislike for DeSantis as a result.

You know, it's like it's hard to evaluate that analysis because, again, we don't know if that if that result is correct in that. Oh, you know, Democrats have shifted their thinking between December and March and that now Democrats would rather face Trump in a general election. So let's maybe ask a question that seems maybe more relevant, maybe more answerable, which is how do Republican voters feel about all of this right now?

So we have seen a bit of a shift in the polls and we should absolutely caveat that it is still early. We are still probably like nine months away from, from anybody actually voting in the race. Um, but there was a period, um, basically in January where Trump and DeSantis were running pretty close. So in kind of a quick and dirty average that our colleague Jeffrey Skelly put together, um,

In multi-way polls, Trump was clocking in around 45% and DeSantis was around 35%. And in head-to-head polls, actually, DeSantis had a slight lead. Again, this was in January. But these days, Trump is on the ups and DeSantis is on the downs. So in the multi-way polls, we're looking at about 50% for Trump and DeSantis is in like the low.

high 20s. And in the head-to-head polls, Trump has retaken a clear lead. And it's interesting, you kind of see this coinciding with a period of some bad stories for DeSantis. So there was, you know, he kind of came out against aid to Ukraine, and that caused a lot of consternation with other Republican elites. There was a story the other day about like, you know, some DeSantis allies worrying that he's not ready for prime time and should he not run in 2024 at all and wait until 2028. And

And it's hard to know which of these is kind of, you know, like which came first, the chicken or the egg, right? Are these stories kind of coming out because, you know, DeSantis is slipping in the polls and people are getting cold feet? Or are people actually cooling on DeSantis because they are seeing these signs that he is indeed not ready or that he's making some kind of unforced errors? Yeah.

But I think in general, like, we should all kind of take a step back and, you know, realize that, like, we are still very early in the primary. DeSantis is still doing pretty well for a guy going up against Donald Trump, who has been basically the unquestioned leader of the party for, what, seven years?

Certainly doing much better than any other non-Trump candidate. But that said, Trump has been leading in the polls for basically the entire time except for that brief period. But Justin had to head polls that he was trailing DeSantis. So I think my perception of the race hasn't changed that much, which is that Trump is the favorite. DeSantis is a strong contender, though. He could beat Trump.

Um, you know, like there is plenty of time for him to, to, you know, rev up and, you know, maybe if something happens with Trump, with the indictment, et cetera, um, you know, there's a lot of question about obviously whether that will actually hurt him, but it could. So, so there's definitely room there. But, um, but yeah, I think that people's, you know, right now, people who are freaking out about DeSantis's chances should probably not be quite so freaked out, but also back in January when everybody was like, Trump is done, DeSantis is, you know, is, is overtaking him. That was obviously premature too.

There's still a lot of race left. So many months. As we know, all too well. No, this is the time for DeSantis Trump references. I'm sorry, not Taylor Swift references. Just to wrap up here, back in 2022, we talked about Democratic leaders fighting

trying to boost more MAGA-aligned candidates in the primary, thinking that they would be easier to beat in the general. Now, I have a hard time imagining that Democratic leaders would try to boost Trump in a Republican primary. I mean, according to very preliminary polling that we have, DeSantis does better head-to-head against Biden than Trump does. But even so, I find it

Just very difficult to imagine Democratic leaders boosting Trump in a Republican primary. When we look back at 2022, was that strategy, though, successful?

It was, yeah. So every Republican candidate who got help from Democrats in their primary ended up losing in the general election. Now, I do think that that doesn't take away kind of some of the normative questions about, you know, does this dilute Democrats' argument that they are the party of democracy? And it also doesn't mean that it would work 100% of the time, right? It ended up being a pretty good midterm for Democrats. And I think that, you know, if the Democrats

national environment hadn't been what it ended up being. Some of those certainly could have backfired. So basically a kind of a, it doesn't, a good outcome doesn't necessarily mean you had a good process. You could, there could be an element of luck in that, but it did work out. But I agree, Galen, like a presidential race is just different and

you know, like the optics, I mean, the optics were already not great for Democrats when they interfered in the Republican primaries. But for a Democrat, you know, to help Trump in a primary is just like those headlines would be terrible for the party. And also just like, I think I would personally cast a lot of doubt about whether Trump or DeSantis would be a stronger candidate. Like,

My gut is also that DeSantis would be stronger than Trump because he doesn't have this great track record in 2020 and in 2022 with his endorsements. But we got burned with that in 2016, as Alex mentioned. And so I don't think that anybody should assume that they know for sure who the most electable Republican in 2024 is going to be. Yeah, it seems like they would really be playing with fire to try to

pull that off in a presidential primary and especially when one of the candidates is Trump. He's unique. Yeah. And also we should say when we're talking about electability here, we're talking about some pretty narrow margins. It's not like, you know, DeSantis,

against Biden running away and Trump is trounced. It's like a point of difference or so. So within the margin of error, as we've said multiple times on the podcast today,

But I think that's it. Should we leave things there? Two very discordant segments. Thank you for playing along. Thanks, Galen. It's all polling and it's all politics at the end of the day. You know, whatever we have to do to explain away our Taylor Swift segment.

And with that, my name is Galen Druk. Tony Chow is in the control room and is also on video editing. You can get in touch by emailing us at podcasts at 538.com. You can also, of course, tweet 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 will see you soon. Bye.