Transcript: What El-Sayed’s Narrow Win Tells Us About the Democrats

Azari: Yeah, I agree with that. I think, again, it’s important to be cautious in how we interpret election results, and I think if we’re looking at these results in Michigan, what we’re finding is, this is a coalition, or a part of the party, that’s ascendant, not that’s dominant. And if you’re Gretchen Whitmer, or you’re one of these more establishment figures, I think you’re not experiencing this as a complete defeat, right? We’re in a moment where public sentiment is pretty negative, and yet the establishment candidate is still getting roughly half the vote.

But at the same time, I think we are seeing some rules having the potential to be rewritten. I don’t want to infer too much from any one election data point. One of the things that I have seen online was just the spending differential between the two candidates, which is quite dramatic. And—

Bacon: Total, like, sixty million to ten million, something in that range, or thereabouts spending, yes. For Stevens, like a big ratio for Stevens. Yeah, go ahead.

Azari: Yeah, she had a huge advantage. And so I’m seeing some of that, and I think, again, we ought to really think about the importance of campaigns, especially in a primary where you don’t have partisan cues, and the importance of what seemed to me like actually really strong emphasis on the economic part of the message. And how that is—one thing I also am keeping an eye on, that’s still in progress, is how the left wing of the Democratic Party is integrating the cultural messages, especially around Israel and Gaza, with the economic messages. And there are some actual efforts to tie those together, about the spending of public money and tax money.

But also there’s, like, both parties always have to be in this constant negotiation between those two dimensions of their party, and the different parts of the coalition that they attract and repel. And that’s just—again, I don’t think we got any definitive data points here, but I think this is additional data to understand that dynamic process.

Bacon: So the polls coming into this is tricky, because there are a lot of polls that are released by each campaign, and it’s not clear there were a lot of objective, quote-unquote, polls. That said, most of the surveys coming to this showed a double-digit lead for Abdul, which led a lot of people, including me, to think he was going to win, and maybe win pretty resoundingly. So I’m always going to want to read polls, because I think that’s better than, like, look, counting streets, counting yard signs. But that said, how should we feel? We’ve had a lot of polling misses that have been major, 2016 being the big one. How should we feel about polls right now?

You guys are political scientists. You guys worked at FiveThirtyEight with me. I blockume you guys are going to keep reading polls, but maybe we shouldn’t. So how should we feel about polls today? Reblockure me that polls are worth reading. Or tell me that maybe I should read fewer polls per hour, if I would just phrase it. Either one of you.

Masket: I’ll jump in on that and just say that, OK, this is a big polling miss. And yeah, Julia was referring also to the polls 10 years ago between Bernie Sanders and Hillary Clinton, where they missed wildly. They predicted a strong victory by her. You get a strong victory by him. I would say generally primaries are much harder to poll in than general elections. It looks like Michigan primaries in particular are harder to poll.

But also, it’s a state with open primaries. Anyone can vote in these, so it becomes much harder to figure out who the likely voters are. That can switch around a lot at the last minute, depending on some campaign spending or whatever last-minute news story activates people to vote. And also, suddenly this race, just in the last few weeks, gets national attention, which it was not, until fairly recently.

Bacon: They’ll drive up turnout, probably.

Masket: Yeah. So there’s a number of things that can change at the last minute. But this is a point that Elliott Morris made online this morning, that if all our electability calculations are, like, two- or three-point differences, and we have a poll that’s been missing by 10 to 15 points—we should take all of that with a huge chunk of sand. Or salt, rather. Sand, salt. Go ahead, Julia.

Bacon: Yes. Sand, salt.

Azari: Yeah. So I think I agree that it’s always better to have some data and some understanding of the data than to just make stuff up. I think that my sense, given all the different theories around why the polls have been off at different elections, has mostly been about the likely voter screen. And as Seth points out, that’s especially kind of a wild card in a primary election. So I would not stop reading the polls, but I do think it’s something to pay attention to, that we seem to have polls that consistently overcount.

The thing about this is that they weren’t wrong in a bunch of different directions. They were generally wrong in the same direction.

And that indicates there are a number of maybe fairly predictable things that happen, that certain people were over-polled. I guess there’s a possibility there were, like, Shy Stevens voters. I don’t know why that would’ve been the case. I’m really skeptical of that theory for the 2016 general election, but in the interest of exploring all things.

So I wouldn’t stop reading polls, but I would kind of have a sense that it’s especially worth paying attention to when the polls are all wrong in the same direction, that something systematic may be happening.

I guess the last thing I’d point out is that people who really do this—which is not me, remember, I was at FiveThirtyEight as the politics and history person—people who really do this, they’ll point out that response rates are just really low.

So that’s another kind of obvious culprit. So when thinking about how polls are wrong, I think we need to be kind of systematic and cautious, and not wildly swing into crazy explanations.

Bacon: So we haven’t fully engaged electability, so let’s do that now. So I can be honest with my own politics. I would’ve voted for Abdul over Stevens, because I’m more ideologically left than center, for lack of a better way to say that. That said, I’m open to the idea of an electability that maybe Joe Biden would’ve been a better electoral candidate than Senator Warren. I’m open to the idea that might be the case.

And so let’s talk about this. In this case, I think a good faith argument might be, it might be that the kinds of voters that the Democrats have to win in a general election—and the kinds of voters they struggle with are rural voters in small towns, voters without college degrees. And the data we have available to us suggests Haley Stevens won more of those people, based on the map showing her winning the rural counties, than Abdul did.

And so if the pool of people you need to win the general election looks more like Stevens’ voters than Abdul’s, isn’t Stevens potentially the better candidate for the general election? Let’s discuss that. I may be wrong. I just want to discuss that openly, though. Either one of you.

Azari: So I’ll say three things about electability. Look, if I can narrow down my electability thoughts to—

Bacon: No, that’s good. Yes.

Azari: Yeah. I think there are three things. One is that I think the Democrats’ vision of what electability is, was basically trained on Bill Clinton in 1992 and has never been fully updated. And so it’s very heavily about a very specific kind of moderation, along with some implied demographics.

Bacon: Some of which I just named, by the way, in my question. I’m not denying, yes.

Azari: And I noticed this with my students when we were talking about—we were talking about the Texas race, I think. And so I’m pointing this out because we don’t live in Texas, and we don’t necessarily know a ton about Texas. They’re just getting national news, and they’re saying, The electable candidate is Talarico. And I’m like, “The electable candidate is no one. This is Texas, and this is a Democratic primary.” The difference between the two candidates versus the broader political spectrum. The second thing is that—

Bacon: They mean Talarico is white and Crockett is Black, basically. Is that what they were saying, with code words? Probably. Or is it—OK, got it.

Azari: Yeah, probably. And it was—you know what I mean? We didn’t have a chance to follow up.

Bacon: OK.

Azari: Yeah. The second thing is, I think electability really speaks to the fundamental problem of our system, which is the weak parties and strong partisanship. So it tells you, on the one hand, voters, it’s now your job to coordinate. The party can’t do it, so now you need to figure out what millions of other people are thinking on the fly, while doing your regular life. So weak parties, and then also strong partisanship, because what it tells you is that the stakes are really high. And that you need to be thinking about this for those reasons.

And the third thing is, I have no idea who I would’ve voted for in the Michigan primary, because by the end of this primary, both candidates seemed so irritating, and I was so irritated with both of them. And I think that maybe is an indicative data point about the nature of primary campaigns. OK, that was my three points. I’ll let Seth talk now.

Masket: I have 29 points about electability here. Let me think about this. So—

Bacon: But think about the question I laid out a little bit, which I think was a useful question to bring about. The blockumption is, the pool of people you need to win a general election looks like the people Stevens won. I think there’s a reason, and I think that might have some validity. Go ahead.

Masket: Yeah. But also I think Julia’s point about, we’re trained on the Bill Clinton era—

Bacon: Yeah, I agree with that too.

Masket: I think has some relevance here. It’s interesting, because electability usually doesn’t rely on a ton of data. To the extent it did this year, where we have polling matchups, I think those are probably not very useful, considering the polling misses this year. So we tend to fall back on some basic beliefs, or myths or stereotypes that we have about who the swing voters are in a given state, which tends to fall back on the rural whites. Maybe, right? But that’s not the only model for what a general election can look like.

There’s been some longstanding belief that ideology imposes a cost on votes. If you’re too far to the left or too far to the right, people won’t vote for you. That’s a model. I think there’s lots of evidence for that in the 20th century. More recent scholarship on this is finding that those beliefs matter less and less to vote shares in November. There was just a recent paper I was looking at, Adam Bonica and Jake Grumbach, where they tried to look within both parties: like, does being a further left candidate or a further right candidate cost you votes in November? And they really had a hard time finding any effect at all, and in fact, sometimes the more extreme candidates actually get kind of an enthusiasm boost in turnout.

So a lot of this just may be a wash. And thanks to that, and thanks to polarization and a lot of other things, you’re going to get a Democratic vote no matter who the Democrat is. And this is a pretty good year to be a Democrat. Now, we’ll get an interesting test of that this year. But in the end, we still won’t know, like, how Haley Stevens would have done.

Bacon: Julia, I can ask you this. I can’t believe I haven’t had this kind of thing before. How should we discuss—my actual concern is that a person named Abdul El-Sayed, who is Muslim, may lose some voters in Michigan who maybe don’t love Muslims, and who maybe don’t describe themselves as racist, but who are, like I said, not outright bigots, but who might prefer a white guy like Mike Rogers. That’s my—I actually think a lot of this is coded around. The actual thing is not that he supports Medicare for All, which most people are able to support, but that his name and background and so on.

And I know Barack Hussein Obama won two elections, but that was an unusual situation—even he was not Muslim, which he would tell you himself all the time, because he seemed to think that might be a downside. And so that’s my—how do we talk about that? Which is the actual issue, I think, is that a lot of people, particularly people who are being honest, worry that maybe Michigan is not ready for a Muslim candidate who also is progressive and married to a Muslim woman.

Azari: I think there’s sort of two ways to think about this, as far as the Michigan electorate, which is: are you going to win the election on the swing voters who somehow have made it to the year 2026 not having a strong party preference, but nevertheless cast ballots? These people exist, and they decide the fate of the world.

Bacon: Yes. The fate of the world, literally true right now. Yes. Quite literally.

Azari: Yes. Who might—who are like, “I vote for the person and not the party.” So my disdain for this perspective has no limits. But that person versus—you’re a young person who’s unhappy with the economy. You’re probably going to vote Democratic if you vote, but you often don’t vote. And there’s political science research around this, that actually the big swing constituency is people who do or don’t vote and who are in the sort of demographics that are not super consistent. Added to that in Michigan is a substantial Arab and Muslim population. And so—

Bacon: Who might be more excited about the candidate instead of less. Which—yeah, that’s the point. Yes.

Azari: So that’s one way to think about it. And then the second way to think about it—I am making the two- and three-part comments, that’s happening—the other way to think about that is, there’s sort of limited data about, we’ll call them non-traditional candidates, right? Women, ethnic or religious which is tricky. There’s people who’ve done experimental work on this, and there are people who have done observational work with the candidates that we do have. And the trick is, there is a penalty.

There’s evidence that Obama paid an electoral penalty for his race in 2008. There’s evidence that ***ism affected evaluations of Sarah Palin in 2008. So we do see that, and we nevertheless see that Obama won twice, and we nevertheless see that Hillary Clinton won the popular vote, and we don’t have a lot of examples of high-level Muslim candidates. We have some survey data that suggests, of course, this is not the most popular group, and that there is some penalty.

So there’s sort of two separate dimensions there. We know that non-traditional candidates pay some electoral penalty, but we don’t know if that’s going to make the difference for whether or not they get elected. And we also know that if as a country we keep in those confines, then we’ll never break out of them, right? And we’ll have a country where a lot of people aren’t really seeing themselves represented politically, and that’s—I think that’s not sustainable. And the work that I’ve done in my recent book, which I unfortunately don’t have to wave around conveniently, suggests that’s not really sustainable. So those are the ways that I think about it.

Bacon: Two final questions. We’re going to bridge a little bit away from Abdul and Stevens to talk about—so a lot of other primaries last night, the sort of anti-establishment, progressive, whatever term we’re using here. There was a DSA candidate who beat an in***bent in Michigan’s Thirteenth District. Let me not mix them—hold on one second, I might be getting my numbers wrong, because I’m a little tired. So there were a couple primaries last night.

In the Seventh District, the Working Families candidate won. His name is Will Lawrence, and he defeated two kind of more centrist candidates. In the Thirteenth District, the DSA candidate defeated a Democratic in***bent. So two places where the sort of more progressive candidate won. In Missouri, Cori Bush, progressive, who had been in Congress before, was the leader of the Ceasefire movement in Congress, actually lost to Wesley Bell. Cori Bush was the progressive, lost to Wesley Bell. So you had mixed results. So I think last night is worth noting. Sometimes progressives won, sometimes centrists won. I think the point Julia made earlier is, we’re not seeing one clear-cut story here, that things are going in both directions. Any other primary results you all want to talk about, or we want to mention? If not, we can move on.

Azari: Yeah, that was all last night. There were those two referenda in—what was it, Missouri and Kansas?—on judicial issues. Those weren’t primaries, but it was conservative states rejecting the conservative perspective.

Masket: Those are important too, yes.

Bacon: I will finish with Wisconsin, where two things I did not love happened on Tuesday. The first was Governor Evers, asked about Francesca Hong, the person leading the polls right now at least, predicted that she would not win the general election, which I don’t know that is overly helpful. And then, if he wants her to win—which hopefully he does, I know he’s for another candidate right now—but I’m not sure how useful that comment is.

And then I guess Hong was interviewed, and was asked about some—I think it was a tweet, but maybe it was something else—where she had questioned the virtues of the Thanksgiving holiday, and she didn’t necessarily walk that back. And while I might think the creation of the Thanksgiving holiday is maybe not perfect, I’m not sure the Wisconsin general election is the perfect time to litigate that. Julia, discuss what you think is happening in Wisconsin.

Azari: So I think the important thing to know about Wisconsin is that, as far as I can tell, talking to people close to politics, people who study politics here, is that no one has any idea what the hell is going on. And no one has. So as near as I can tell, just to give the state of the race: Hong came in as this sort of unlikely DSA candidate. She’s in the state legislature. The rest of the field refused to consolidate. The two major candidates who were blockumed to be front-runners crashed and burned in different ways. The state—

Bacon: Mandela Barnes and Sara Rodriguez?

Azari: Sarah Rodriguez. Yeah. Who, the former and the sitting lieutenant governor. The sitting governor—who has declined to run for re-election, he’s not term limited—declined to endorse either of them, his two lieutenant governors, and then endorsed David Crowley, the county executive in Milwaukee, who re-entered the race when Rodriguez dropped out. So they unconsolidated the field.

Also, during this period from, like, March to June or something along those lines, there was very little polling. So no one really knew what was going on. And then my colleagues in the law school at Marquette released a poll last week that suggests that Hong is pretty much running away with it. That David Crowley, as the Milwaukee County Executive—he was the one who was endorsed by the sitting governor—that’s not really bearing out in the results.

There are two other candidates, Joel Brennan and Kaldar Royce, who also have some support, but not enough. So no one knows what’s going on, and Hong is really benefiting from this sort of party lack of consolidation. And then Hong has these weird liabilities, like these weird tweets. And so I think there’s a kind of—

Bacon: Are we calling them—you and just people—are we calling them weird, or are we calling them ideologically consistent and morally courageous, but elector—are they weird, or are they morally courageous? This is a serious question.

Azari: Is it morally courageous? I think it’s weird, honestly.

Bacon: OK, all right. I’m just going to ask you too. OK.

Azari: All right. I think Thanksgiving is a weird hill to die on. Although—I should note, Hong works in the food industry, and this was somehow in the context of the food industry. But yeah. There’s a political scientist at University of Wisconsin, Nathan Kalmoe, who wrote a piece yesterday for a very good local Wisconsin site called The Recombobulation Area—named after an area in our airport. And it really lays out the case that electability is not really a concern in a year where Democrats have a kind of generic advantage.

I think that’s really the question. And then the finer point question with Hong is, are people going to vote based on, did this person say they are morally opposed to Thanksgiving? Or are people voting based on this sort of sense that seems to be driving Hong, which is that she is talking about economic issues and about doing something different in a period where the state has been in a kind of partisan stalemate, has been hit very hard by—basically whatever happens economically, it hits the upper Midwest really hard.

That is a question. I don’t know the answer to that question. I don’t have a lot of evidence that this opposition to Thanksgiving is like a core element of what she’s running on. It struck me more as, I’m riffing on food politics, but I might be wrong.

Bacon: My impression is she made a lot of comments in 2020, in that period, about police abolition. She made a lot of the same comments—a lot of it Zohran, or else a lot of people have made those comments. It’s my impression, and you live there and I don’t, is that she’s a little more reluctant to walk them back than Zohran was, Whatever, I said that and now I’m done with it. Because Zohran—I feel like she’s more reluctant to walk those things back, which I would argue might be a great, morally correct position that may also not be useful exactly right now. Maybe that’s what I’m trying to ask about. Yeah, I don’t—I guess we don’t know her.

Azari: Yeah, I don’t either. Although she materialized at a coffee shop I was at once, so I guess I’ve seen her in person. But just randomly, one Saturday afternoon. I think what we’re seeing is—I don’t know if I would call it morally courageous, but I would call it consistent. She’s running on a sort of consistency around, I’m not going to play the electability game. I’m just going to be who I am. To use another word that is banned, it might be part of the authenticity appeal, but I do think that’s something that’s driving her. I’m just trolling Seth at this point. That is what’s driving it.

So anyway, that’s the view from the ground with Hong. I think it’s hard to translate, and it’s hard to explain, because the way the race has gone down has been so strange and so bizarre, and the logic of it has been so unclear.

Bacon: Because she’s the person who’s opposing data centers, because she came—the first one—and that also came from a position of, I’m not an insider. And I think a lot of her views are coming from, she’s not an insider, she’s not trapped by what the insiders think. And on policing, that may be a bug. And on data centers, that’s probably a feature. And it’s interesting to figure out where she ends up.

Azari: Or what people are weighing. Because the chances that we’re going to abolish the police in Wisconsin are quite small. Everyone’s got their eye on a state that borders Lake Michigan when it comes to a data center. And I think an interesting data point there is that she’s been the strongest among Democrats on that, but actually the Republican candidate has taken a pretty—

Bacon: Oh, sure.

Azari: —cautious stance on data centers as well.

Bacon: Seth, since you’re really close to Wisconsin—I know you’re not—but what are you seeing? Because I guess part of the issue is, like, Hong is having some of the same… Now Abdul just barely won. So I think now all the attention is going to be on, can Hong win this race? Are the polls off with her? I blockume the centrist establishment’s going to really try to beat her in this next week. So I’m curious what you think, what you’re seeing there.

Masket: Just to follow up on what Julia said, for one thing, just following that race from afar—I’m really glad I’m not a Wisconsin pundit. That just seems—it’s a lot of work just tracking that one race. So my hat’s off to you, Julia.

Also, like, with Hong specifically, I think there’s one thing that we’ve seen in a lot of these candidacies, these progressive or DSA candidacies this year. What seems to be mobilizing a lot of their supporters, their donors, the people who turn out to vote for them, are a couple of issues. I think Israel-Gaza is a big one, and US support for Israel. And the other is just a general sense of, we don’t think the Democratic Party is pushing back hard enough. They’re not being strident enough against Donald Trump. And that’s a very vague thing, but it’s a consistent thing, and that seems to be mobilizing people with a lot of energy.

And I think people who are looking for that—the number of people who even know that she had a tweet about Thanksgiving, no less will actually change their primary vote because of it, I think is very small. I think they’re more mobilized by, like, they see her as the one who is going to shake things up, who is going to stand up to Donald Trump, and do these heavily symbolic and more bombastic things than other Democrats will. And everything else is just kind of noise.

Bacon: All right. Julia and Seth, tell everybody where they can read you, in short form and in long form.

Masket: Oh, yeah. Let’s see. I’m doing a lot of my work on my Substack these days. It’s called The Smotus Report, S-M-O-T-U-S. That’s Seth Masket of the United States. You can find me at smotus.substack.com. I can also be tracked on Bluesky, also under the name Smotus. And I have a new book out that I hope you’ll check out if you’re interested in seeing more. It’s called The Elephants in the Room. It’s all about the Republicans’ decision to nominate Donald Trump in 2024.

Azari: I do most of my blogging on Good Politics, Bad Politics, also on Substack. It’s a group blog, so you get a couple different perspectives on there. I’m also on Bluesky, just my name, just Julia Azari. And I have a book out about a year ago from Princeton University Press—it keeps, I think, getting more relevant—called Backlash Presidents, and it is a historical deep dive into the connection between race and presidential impeachment.

Bacon: Julia, Seth, thanks for joining me. Good to see you. Bye-bye.

Azari: Thank you.

Masket: Thanks so much.