Mixed Signals: When Biden lost it & Nate Silver’s election forecast

Nayeema Raza: Do you know the thing of a successful podcast, Ben?

Ben Smith: What’s that?

Nayeema: You have to say something you feel deeply uncomfortable with at the end of it.

Ben: Max Linsky told me that. I know. Oof. I feel like I’m already on like three and we’ve barely gotten rolling.

Nayeema: Oh, excellent.

[MUSIC]

Ben: I’m Ben Smith.

Nayeema: I’m Nayeema Raza.

Ben: And this is Mixed Signals from Semafor Media.

Nayeema: Today we’re diving headfirst into election season. It is here. How do you feel about it, Ben?

Ben: I mean, I’ve been covering these for 20 years and I have to admit I kind of love it.

Nayeema: You’re like a media monster. It’s like, happy when there’s a crisis. Oh, more coverage.

Ben: Yeah, that’s a real occupational hazard.

Nayeema: It is. We’re going to run down the presidential debate, which is also known as, I guess, a wellness check on the geriatric Trump-Biden rematch. And we’ll focus in on one of the stranger conspiracies of American politics that there’s this real feedback loop between political coverage and particular polling, and the actual behavior of campaigns or voters. They’re not mirrors, they’re drivers of what happens. And as part of this, we’re going to turn to Nate Silver, who’s the founder of FiveThirtyEight and now prints under his substack Silver Bulletin. Nate is seen as the guru of this world that when he put out his new predictive model on Wednesday, it prompted a premature victory lap from one of Nate’s apparent devotees, Donald Trump.

Donald Trump: We’re leading every swing state. And one just came out, I guess, Nate Silver of the New York Times, that we have close to a 66% chance of winning, which is a pretty good chance, I guess, when you think about it. Right?

Reporter: But Mr. President, you got time for-

Nayeema: That’s tape of Donald Trump calling into a Black barbershop in Atlanta where Trump supporters, including some Black members of Congress were gathered. And thanks to Semafor reporter Kadia Goba who was in the room, we actually have that tape.

Ben: Classic Trump in many different ways, including saying that Nate Silver worked at the New York Times.

Nayeema: Not factual anymore.

Ben: No, but obsessed with prestige and the most prestigious media brand he can drag into this thing.

Nayeema: We’re going to have to ask Nate Silver about that when we talk to him. And then we will take a look at Blind Spots, the stories you’re not seeing in your feed of viral debate clips.

[MUSIC]

Nayeema: All right, here we are. It is 11 PM Eastern Time, almost, and we have just finished watching the debate. Ben, you’re here with a new haircut. Did you get a debate haircut?

Ben: Got to be ready.

Nayeema: Got to be ready. And then Max Tani is here. Hi Max.

Max Tani: Ben’s haircut is the biggest news of the night for sure.

Nayeema: It is. Let’s set the stage a little bit. Where did we all watch the debates? Maybe Max, you can go first.

Max: I watched it here in the basketball stadium at Georgia Tech, which is where the spin room is.

Nayeema: Is it the most exciting thing that’s ever happened in that basketball court, or by a long mile not?

Max: I think there’s a pretty good basketball team that plays here occasionally, but actually it’s pretty exciting. There are a lot of people here. There’s 800 credentialed journalists and they’re all kind of going down onto the floor, so it’s starting to feel a little hectic and crazy.

Nayeema: June madness they call it. And Ben, where did you watch?

Ben: I watched at home with Family on my YouTube TV.

Nayeema: Oh, Ben.

Ben: That’s all I got. Nayeema, where did you watch it?

Nayeema: I watched it, probably a strange room. I’m in the house of a friend who’s a tech founder. I watched it in an election watch party with about 50 people I would say, but it’s probably about 60:40 Biden Trump in this mix, or maybe it’s one of the few rooms that I guess has that kind of diversity. There are a lot of tech founders. There are a lot of maybe liberal creatives and executives around in New York, and then there’s beautiful women, many of whom happen to be one of the first two categories as well.

Ben: This was like Frank Luntz, but only talking to billionaires. I love it.

Nayeema: No one’s disclosing the incomes of the people in this room, but it’s fascinating to watch. During the intermission, everybody started speaking about their aging parents or grandparents. So let’s figure out, we all watched this debate, played out for 90 minutes. Everybody talked about hate watching it. It was a geriatric wrestling match, whatever they wanted to call it. But there’s a tradition where somebody decides who won the debate and they decide it at a certain moment in time. And Ben, I think you have actually done that honor before in 2012.

Ben: Yeah, this was back when Twitter was new and the notion that that things were accelerating was still also new. And early in, about 45 minutes into a presidential debate, I published an article on why Romney had beaten Obama in the first debate, which got me widely denounced. But it in a way was my meta commentary on the acceleration of these debates. I mean this one I think was much more accelerated though.

Max: This one, you could have published that story within five minutes and I don’t think anybody would argue with you.

Nayeema: When was the moment you think it was decided? And who decided who won the debate?

Max: I think paying attention to the media coverage sitting here in a room full of 800 members of the media, I think it was decided very, very quickly. The moment that got the loudest audible kind of chuckle or reaction in my section of media with several dozen people was when Trump said, I don’t know what he said there at the end of one of his answers, and I think that really summed up a lot of people in media feelings about it.

Nayeema: Yeah, the moment that we’re talking about happened around 9:22 PM, where Donald Trump said ...

Donald Trump: I really don’t know what he said at the end of that sentence. I don’t think he knows what he said either. Look.

Ben: I think the thing with this sort of debate is that the journalists watching certainly I think within that first five, 10 minutes, which Trump basically summed up there, but included Biden just kind of glitching for a few seconds and forgetting what he was saying, staring down and broadly just looking as old as Democrats feared he would look, as hesitant after a week of rehearsals as anybody could have feared. And I think the thing is then maybe the debate is still going and these two guys are arguing about their golf games, but what has already happened then is there’s headlines in every publication. The stories are being written, the narrative has taken off, and the narrative is that we’re heading into a huge crisis for the Democratic Party. This is a massive new crisis for them.

Nayeema: I have to say the question of who won this debate is almost not a good question. It’s like, who lost this debate less?

Ben: Yeah, I mean it really pretty remarkable. I watched a lot of debates where it was not this clear and where the narrative did not lock up in the first 10 minutes. And this wasn’t just the media watching it, this was anybody who was watching it. And it was in a moment when we’d all been, I think spent a lot of time getting scolded by the White House for daring to raise questions about his age. So I think there is, along with just everybody having watched a debate in which an old man struggled to hold his own there, there’s also a sort of, we’re sick of getting yelled at about writing stories about his age and you’re seeing some of that too.

Nayeema: The idea of where you call it also, I mean Twitter is a place and it’s certainly a place where the media pays a lot of attention to. I was also following on Instagram, TikTok, and the betting markets, the betting markets of Avada and also Polymarket. You started seeing as early as the first few minutes, as soon as Biden walked in and then Trump walked in, that’s when it started dropping. The race started much higher. It was like odds were 225 up for Joe Biden. It ended at 275.

Ben: Did you see something different on those places than we were seeing on Twitter? That sounds like kind of a reaction to Twitter. Was any of the social media moving in different directions? Was anybody saying, “Wow, Biden’s great.”

Nayeema: I didn’t see any of Biden’s great, but I did see some kill shots of Trump. The muted mic shot at 9:21 PM which was a minute earlier than Biden’s, kind of what some people were turning on social media as freezing, or what Trump called out as unable to finish his sentence. And then I think 10 minutes earlier than that when Biden stumbled over trillionaires instead of billionaires. There were a couple of missteps in the same minute. That’s when I heard the most audible gasp, and that’s when I think people kind of made up their minds across the room.

Ben: Yeah, it is interesting though. I think that in so many political contexts what people think and the sort of more populous TikTok, Instagram, what’s happening out there matters a lot. I think in moments like this, the elite consensus matters a lot, and what you had was a sort of crystallization of that elite consensus tonight.

Nayeema: One of the things in the betting markets was the odds on there being another candidate, and that also I think is reflected in the media, but over the course of the hour and a half of debate, it went from about 10% odds of there being another candidate other than Biden to almost 20% odds. So doubling.

Max: The reason why the elite consensus matters in this case is because those will be the people who will be able to potentially pressure Biden to get out if that’s still an option, if he still believes that that’s an option after this debate. And I think that what you’re seeing for the first time is rather than it just being a handful of small commentators here and there, people like Ezra Klein or David Axelrod who’ve taken a lot of flack for calling on Biden to potentially step aside or for the Democrats to consider someone else. I think what you’re really seeing is that becoming a more mainstream opinion.

Ben: Yeah, the big question is going to be, over the next few days, do former presidents, do members of the US Senate, people like that, come out publicly? And I’m not sure they will. I think the train is pretty far out of the station. But actually it’s kind of interesting, because I think that I’ve written Twitter’s obituary a lot of times, but I do think you see this really, really centralized political conversation that is just propelling a national narrative here.

Nayeema: Twitter does matter. In earthquakes and in presidential debates, this is when we pay attention to Twitter. What is actually fucking happening in the country? Now here’s a media strategy question. There’s people espousing a version of reality in which Biden is not the candidate come November, but there’s also the question of what the media strategy should be. And in my mind, if Biden’s going to keep running, isn’t there a point of him saying, “Look,” not him, but the media narrative becoming, “You’re not voting on the person, you’re voting on this ship.” And so do you want to be on board with Jake Sullivan, John Finer, Pete Buttigieg, or do you want to be on a little cruise with Trump, Rudy Giuliani, and MTG [Marjorie Taylor Green]?

Max: It’s just not how anybody has voted for president in the last several elections. It would defy the way that the polls look right now, and it would defy all recent voting history for presidents. And Americans don’t like being told that they actually don’t know who’s in charge. Biden’s team has worked very hard to push back against those of conspiracy theories.

Ben: Nayeema, I like the idea in some technocratic future, but the morning the press secretary comes out and says that is the afternoon the president resigns.

Nayeema: But I do think that people, they want any reason to vote for Biden, but they do not have one. Many kind of double haters, young people in this country, are looking for that reason. But the debate highlights for me were I didn’t have sex with a porn star, Manchurian candidate, and who’s going to carry whose bags at the golf course.

Ben: Just makes you proud to be American.

Nayeema: It’s good television, bad presidential debate. All right, so last thing, predictions. Will there be a second matchup between these two men? And who would win the golf game if played? Max, you go first.

Max: I think that Trump obviously spends a lot more time playing golf, so I think he would have an easier time in that regard. And I don’t think that there’ll be a second one. I know that many news executives are already, including some of those at ABC, are concerned about this, about the possibility that this debate would go in one certain direction and one of the candidates would want to cancel that. There’s precedent for that, and so that’s easy money for me.

Ben: I actually think there is going to be another debate because Biden is now trapped in a situation where to withdraw from another debate is to withdraw from the race, and he may just have to be wheeled up to the next one too. Brutal.

Nayeema: So it seems like you’re not voting for him for the golf game either.

Ben: I don’t know. Trump notoriously cheats at golf. Who knows?

Nayeema: I’m going to have to take the craziest bet here. I mean maybe it’s like I drank the Kool-Aid in the room, which is that there will be a second debate, but it will be between Donald Trump and somebody else.

Ben: I love it. I love that kind of New York room, Nayeema.

Nayeema: I will take odds on this.

Ben: Go breathe some fresh air.

Nayeema: Let’s take a quick break and when we’re back we will play a conversation that we had with a gambler, Nate Silver, and a person who had Donald Trump up before this debate. I’m sure the model will only have him up even more beyond this debate.

[MUSIC]

Nayeema: All right, as people are hearing this, the debate is over, but the polls are just kicking off and we will be drowning in them till November, and then we’ll be analyzing them till the end of our lives. A man that we have to blame for that is Nate Silver. He came up as a kind of groundbreaking baseball statistician, a kind of Moneyball character. He also plays a lot of poker. And during the 2008 presidential campaign, he was revealed to have a side hustle in his baseball days. This was as the anonymous blogger Poblano on the progressive site Daily Kos. And Nate then went on to build FiveThirtyEight, which he licensed to the New York Times for a while, and then sold I’m sure for a sizable fortune to ESPN/ABC.

Ben: When I was at BuzzFeed, we actually talked to him about buying it.

Nayeema: What was the price tag?

Ben: It never quite got that far, but I think when it became clear to him that it would’ve been BuzzFeed equity, he decided to go in a different direction.

Nayeema: Seems like he made a good choice.

Ben: Wise man.

Nayeema: Nate has since left FiveThirtyEight and started his new substack Silver Bulletin, but he’s kind of the king of this world.

Ben: Yeah. Nate, I think, kind of invented this modern conversation about polling back, as you said, when he was an anonymous blogger. And the thing that is easy to forget back then is that was a year, 2007-2008, when Obama was really consistently leading in the polls, the national candidate. But there was just Democrats were panicked every day that these polls were concealing an undercurrent of racism in which voters were going to show up and vote against Obama even though they would tell pollsters they were going to vote for him. Something called the Bradley Effect, a urban legend actually called the Bradley Effect.

And Nate’s job that cycle is just every day to say, “Don’t worry about it, the polls are correct, Obama’s going to win.” And there was just nothing more addictive to Democrats than that, just waking up in the morning, hitting refresh on whichever of Nate’s blogs he was on at that moment and being like, “Is Obama going to win?” “Yes.” And then Democrats could go about their days. And so that made him this really iconic figure in 2008, and then again in 2012 actually. And then to some degree again in 2016 when he went into the election saying, “Hillary Clinton’s probably going to win.”

Nayeema: I think in that case it was like 25, 27% that Donald Trump would win the election, and 70+% that it was going to be Hillary Clinton in 2016. He has this probabilistic model rather than the kind of polling itself. It’s not a snapshot in time, instead it’s looking across the aggregates and it’s giving a percentage chance that this person is going to win the election.

Ben: Thing that’s always driven me a little crazy about Nate is after that election, not wrongly, Trump won and everybody said, “The polls were wrong, you guys were wrong. You’re all idiots.” And Nate said, “No. Look, I said Trump had a pretty decent chance of winning, and he won.” And so, all right, I don’t know what you’re supposed to do with that.

Nayeema: That’s the thing with being a statistician, it has a built-in hedge so you can always be right, but you’re like, “But I was less right than the other thing I could have said.”

Ben: Yeah, but he really is, I mean he really is king of this world, and so much that he has a whole legion of imitators and rivals. I think some of them are also named Nate, which is a little confusing.

Nayeema: Yes, you have referred to some as Nate One and Nate Two, but we’re not going to say which one’s which.

Ben: I mean Nate is. This is Nate One.

Nayeema: Nate Silver is definitely. Poor Nate Cohn, great journalist at the New York Times. But these days, Nate Silver is a big deal, and in the mouth of the former president Donald Trump being celebrated by him, because Silver Bulletin, which is Nate Silver’s substack has just unveiled this brand spanking new probabilistic model which shows a greater than 60% chance of Donald Trump winning the race.

By the time people are listening to this, we’ll have just had this debate. I think that Nate thinks that debates are just like a flash point in time and don’t really matter to what’s going to happen, and so we really want to spin forward this conversation. What’s the horse race we’re going to see from now till November? What’s the signal? What’s the noise? What actually matters? And how right or less wrong is he going to be this time around in 2024?

Ben: Hi Nate, welcome to Mixed Signals.

Nate Silver: Hey Ben, how are you?

Nayeema: I’m so excited to be here, because Ben tells me you and he have been fighting on Twitter since its advent basically.

Nate: I feel like we’re pretty sympathic. We’ve both been through four or five different media hype cycles, I think. Right?

Ben: We used to fight more.

Nayeema: All right, well, I’m happy to be here to moderate. Congratulations on the model by the way.

Nate: Thank you. I appreciate it.

Nayeema: Has Donald Trump called you yet to sing your praises for showing that he has a slightly better chance of winning this election?

Nate: I mean there was a spam call earlier. Right? I mean it had some Kansas City area code. I don’t think he has any properties out there. I don’t think he has any casinos out there, so I don’t think it was him, but I’ll have to check my voicemail.

Nayeema: Trump Golf Resort, Kansas City is coming. So you just released this model. There’s a big debate that by the time people are hearing this has happened, will the debate matter? Will the debate change your model?

Nate: Look, the way I put it is that the debate, one of the two remaining debates, is one of the only two times that there could be something big that happens. I mean, clearly debates don’t make that much impact, but if you have a one point race in Pennsylvania, Wisconsin, and Michigan, a one point impact is a big deal. I mean, look, it is interesting that if Biden does really badly, having this debate be earlier, which their campaign wanted, it does create this potential bedwetting metacycle.

Nayeema: It’s a really bad metaphor for this particular debate.

Ben: Bedwetting metacycle. That is so gross.

Nate: This potential cycle, we’ll just say that, where because the Democratic convention is not until August, Biden could say tomorrow that, “I am choosing after all not to run for a second term, and Kamala Harris is my recommendation for the nomination.” And the superdelegates in Chicago will determine the different Democratic nominee. He does have that option. Optionality is important in kind of a gambling decision-making context. Any option that you can potentially take, if Biden does really, really badly in the debate, then at that point the right rational play, if Democrats care as much as they claim, this is a conditional, this is not a prediction, Democrats do have the option to try and persuade him that, “Joe, it’s too far gone, it’s time to step aside.”

Nayeema: I know that you’re embroiled a little bit with the Biden administration, or you’re getting heat for putting out your critiques of age and calling attention to this. Have you heard personally from people in the administration?

Nate: No, I don’t. No. I think, and campaigns have never been very often to talk to me because I just tend to be, it doesn’t really affect the model and if you know me, I’m like, it’s 50:50 they’re going to piss me off and-

Ben: Does this sound like a man whose opinion is going to be changed by being worked? I think not.

Nate: But for the book, I do a lot of reporting in the book. I talk to some of the best VCs in Silicon Valley, and some of the best poker players in the world. Because that’s my world. I don’t get along with these politics types. And I’m leaning into that deliberately in saying this is a perspective that’s independent, and not that I’m not subject to my own cognitive biases. This year, it’s a little weird that the candidate I think’s going to win is not the candidate I want to win. But I am looking at this as a way that someone who’s making bets for a living is looking at it.

Nayeema: We’re going to dive into your model in this conversation, but first we really want to address the distrust that surrounds kind of the entire election. And in some cases there’s a sense of conspiracy around predictions and polling, the media, election coverage writ large. The most insidious version of this conspiracy is kind of that there’s no underlying reality. That rather than being a mirror, and this is not specific to you, but the entire industry. That there’s no reality that rather than being a mirror of what voters preferences are, that these models are like an invisible hand whereby media mediates these preferences and democracy overall. They tell people when to freak out, they tell campaigns when to pay attention. They drive people’s expectations of who’s a winner and a loser, and they create this kind of pendulum shift where if you think your candidate is losing, you’ll show up, but if they’re winning by a wide margin, you’ll calm down.

Nate: Yeah. I think that’s people who spend way too much time on Twitter. And you see evidence of, I think, and this is now one consequence of Democrats being the party of college-educated elites, maybe not the financial elite most of, is that they are in an echo chamber. Democrats think that the New York Times being mean to Biden is the biggest reason that he’s losing. Democrats think that because they hear a lot in their social circles about Israel-Palestine, obviously a tragic situation, but if you poll regular voters and especially swing voters who are disengaged, Israel-Palestine in the Quinnipiac poll this week, it’s literally the lowest priority on the list, I think except for Trump’s criminal convictions maybe.

If you kind of look in reality, first of all, it’s not that surprising that Trump is beating a candidate like Biden, as flawed as Trump is as a candidate himself. But look, Biden is 81 years old, is asking to be president until he’s 86. This has been a consistent concern in polls, overwhelming concern even among independents and Democrats for a year or so now. We had very high inflation, which has abated. That’s good for Biden. There are issues like immigration which is higher partly because of pandemic-related fluctuations. By the way, the most important reason is that we know what happened in 2016, which is that Clinton did win the popular vote and lost the Electoral College. Our model actually has Biden as a very slight favorite to win the popular vote again, but almost a 2:1 underdog to win the Electoral College because there’s that persistent gap of about three points, and that means that you are kind of fighting very uphill. It’s like getting 15 yards for first down instead of 10, if you’re the much better offense. Sorry for the sports analogy here.

Ben: Before we get back to sports, and sorry, I have to ask you about media. Obviously, really this is about us. And I feel like you sort of skipped over Nayeema’s question a little, which is there is obviously a kind of a feedback loop between polling and real politics. The most obvious thing that comes to mind is the way in which the 2019 poll really showing Warren doing terribly against Trump, I think, did help shift votes and push her out of the race. And I’m curious how you think about that feedback loop.

Nate: I think in multi-candidate primaries, then it’s more rational for voters to switch horses. If you’re indifferent between Bernie and Warren and one of them seems like they can beat Biden and the other can’t, then it’s rational to do that. In a two-way race, I’m much less convinced of that. I mean the empirical evidence of ... Because you hear it both ways. You’re like, “Oh, it’s going to suppress Democratic turnout because it makes them depressed.” Or you hear, “Oh, it’s going to motivate Democrats because they’re not complacent any longer.” Look, coverage of polling is a subspecies of media coverage in general. I think having a model that has set rules that is ... I mean what it needs to be objective is a complicated philosophical question actually, which mirrors ironically, I think, some of the debates and media about media objectivity overall.

Ben: For sure.

Nayeema: Yes.

Nate: But to have a process that you decided on ahead of time, this is 98% the same model that had Democrats winning in 2008, and ’12 and ‘16 and ’20. It’s been a long time now. You plug in the same data to the same model and Trump has a pretty persistent lead in polls, and the polls have been very steady. And so yeah, I think Democrats in general really over-index on blaming the media for everything now.

Nayeema: Why do this? Why put this out into the world and say, “This is what Trump ...” What’s the impact of it, if not influence?

Nate: I don’t care. The impact is that you now have more accurate information that you can use to make better decisions in your life. Maybe if you don’t like Trump and you want to apply for some foreign visa in Canada or something now, then you know that you probably should take that pretty seriously. Or maybe you’re trying to hedge risk in the stock market because you think Trump will cause inflation. Or maybe you’re just on an offshore betting site. My audience is not the political audience. This is what I’ve realized. It’s like I do not like the political class of people. I like poker players. I like Wall Street, and Silicon Valley.

Nayeema: These days some of them don’t like you either, Nate.

Nate: Yeah, and it’s mutual because I’m cut from a different cloth from them.

Ben: Though didn’t you like them better when they liked you? There was this era when Democrats would wake up every morning and say, “Oh my God, is Obama still winning?” And they would go to you and you would say, “Yes.” And they loved you.

Nate: Empirically, I never love them back.

Nayeema: But back then I remember when you were hired at the Times, there was a bit of kerfuffle. You had said that you voted for Obama, and you showed your card. So at the time you were part of not necessarily the class, but you were in that party for the moment. So how does it feel to go from being a darling to being at odds with?

Nate: I think if I do my job well, it’s inevitable. And journalists sometimes say, “Well, if both sides are angry at you, then you’re doing your job right.” Which is a cliche that I don’t entirely or mostly don’t agree with. However, if you’re in the specific enterprise of making election forecasts, look, I would be a total hack if all of a sudden my model said, “Oh yeah, forget the polls, something fundamental has changed. And now ignore the fact that Joe Biden is down six points in this Quinnipiac poll that comes out, or four points in the Georgia poll, or whatever else. Ignore the fact that nothing has moved the numbers.” That’s what would terrify me if I were in the White House. It’s like he’s not that far behind. It’s just like nothing has changed the race and voters have dealt with the same candidates now for five years, literally. Trump now three times in a row. So what’s going to change this?

I mean fortunately for the White House, they’re close enough that if you have another one of these systematic polling errors, they could win. It wouldn’t have to be that large an error, but it’s a narrow path. The election that reminds me of a lot is 2012 where the national polls were tight, as they are this year, but Obama had consistently more robust polling in the swing states. They were very consistent.

Ben: The other thing in 2012 was there was a little cottage industry of, as you recall, I think unskewing the polls, right?

Nate: Yeah.

Ben: There was a set of Republicans who just felt like this can’t be true, and we’re going to kind of go into the math and tweak a little here, tweak a little there, which I think people like you and me mocked. And I believe after the election, some of the big un-skewers sort of apologized maybe to you personally. But how do you see Democrats responding? Is it a mirror image of that? Is it something different?

Nate: Yeah, I think that epistemology, to use a fancy term, of Democratic poll watchers is a little bit ... I was one of these people who because of that experience in 2012, I thought naively that Republicans uniquely suffer from epistemic closure, and now it’s clear that Democrats do just as much. If anything, the unskewing is because their audience is more college-educated, it’s more elaborate versions of lying to yourself. But the polling overall was very accurate in 2022.

So after 2022, it’s like this is about as well as polls can do. And if people still blame the polls and make up this whole narrative ... Whereas in contrast, in 2016 and 2020, the polls did have a clearer skew, and it was a skew against Trump. 2020 in particular, Biden won. People don’t blame the polls as much in 2016. In 2020, Biden was up eight points in some of these states like Wisconsin, ended up winning by a point or so. That was a bad polling area and it didn’t favor Democrats. And so it’s absolutely right that the error distribution on polls is larger than people assume, but you can’t predict which direction it’s going to go. I mean the media was guilty in 2022 of assuming that, oh, we’re going to have another Democratic skew, and we didn’t.

Nayeema: Right. I want to make sure that ... By the way, we’ve gone deep down in this conversation, I just want to pull up for a second, because yes, the polls were wrong in 2016. As you point out, also wrong in 2020 though Democrats don’t like to talk about that as much because they were wrong in the quote “right” way for them, I think as you were saying. But now in 2024, you have just released this new presidential election model which shows Trump with a 66% chance of winning the election, and that made a lot of news, which I’m sure pleased you. It surprised a lot of people. I’m curious if it surprised you, and I’m curious if you say the model is like 98% what you used to have in FiveThirtyEight, which FiveThirtyEight right now, no longer owned by you under ABC. That poll at FiveThirtyEight is showing Biden up and you’re saying this is 98% the same model?

Nate: No, it’s not because, yeah, I should just clarify. I owned all the IP for FiveThirtyEight because I had a good lawyer when I signed with Disney.

Ben: This is one of the great media coups, isn’t it?

Nayeema: Yeah.

Ben: For real.

Nate: So I kept all the IP and they hired a new guy to create a model that has nothing to do with mine.

Nayeema: Got it. Okay.

Nate: So they own the trade name FiveThirtyEight. They own the Fivey Fox mascot, I think. But this is the same model that has been the most accurate model the way that we judge it, which is whether the probabilities align. That you’re supposed to in some cosmic multiverse, Biden is supposed to win this election 35% of the time, despite being an underdog. And 0.001% of the time RFK Jr. sometimes wins, or whoever else, and Trump wins 65%.

Nayeema: Right. So if there were a hundred such elections, then Trump would win 66 of them, and Biden would win 34 of them?

Nate: Right. Yeah. This is the gambler’s mentality. In 2016, our model, or my model I should say, to avoid confusion now, my model had Trump with a 30% chance and that was much higher than the betting market odds. So if you were using our model, you would actually put a pretty large wager on Trump, even though you expect to lose that 70% of the time, the odds you’re getting are so good that it’s what I would call plus EV, or positive expected value.

And I know it seems totally foreign to people in the politics world to think this way, but that’s some world that I come from. I was a professional poker player before I started covering politics. I’m an experienced sports better. I just wrote a whole book about gambling, and it’s actually a pretty big world of people who think like me and not like the political class. And they subscribe to the newsletter and like the podcast and things like that. And it’s very freeing not to have to coddle people and say, “I mean, look, the fact is that Biden is in a fair amount of trouble.” And I think people are being too polite about it.

And by the way, this is also a case where the decision to run again was certainly not obvious. I mean, a lot of Democrats in Congress assumed that Biden wouldn’t run again. Or he never pledged to be a one-termer, he did not do that, but he kind of said, “I’m a bridge.” And made that implications sometimes. So this is a decision where Democrats would’ve had the option. And now it would be kind of insane to be fair, right?

Nayeema: Is it impossible? Is it impossible in your ...

Nate: No, it’s not. Until August, whenever the final date of the convention is. No, it’s not impossible. And look, to pull the emergency lever is obviously an insane thing to do, but the question is it less insane than walking into an election where you’re a 2:1 underdog? And by the way, I want Joe Biden to win. I said that in the article. I’m not a Trump guy at all, right? But I have experience. You have to do this in poker. In poker, if you take a bad beat, you can’t then mope and lose the next hand and tilt off. You have to actually kind of understand that our emotions, especially for smart people who are very, very good at rationalizing, they’re the worst about it.

Nayeema: And now liberated in your new reality, no longer bound by the ethical standards of the New York Times or ABC newsrooms, how much money do you have on the race?

Nate: I have none personally. I do a consulting deal with a financial firm, and they may be making bets based on their conversations with me.

Ben: Why aren’t you? Is that a lack of confidence?

Nate: Because I don’t need the money, and because I would want to hedge my risk if anything. It’s better for me. Financially, it’s better for me if the favorite in the model probably wins. And so I’m not looking to pile onto that. I think the financial implications are significant of that.

Ben: But maybe you should bet the ... Actually, right, you should bet the opposite.

Nate: Yeah. That’s a decent argument.

Ben: For sure.

Nayeema: You should bet that. You should hedge.

Ben: Yeah, it’s interesting, right? There’s often I feel like there is that funny, the polls were wrong thing, that is, it’s always sort of over served, isn’t it? I wonder why. I actually don’t really have a theory on why that is. I’m curious if you do the sort of desire to think that the polls were wrong.

Nate: I mean, I think it’s partly because when we’re confronted with objective facts, and I’m aware of an objectivity is a complicated concept, but-

Ben: No, but I mean retrospectively there’s this sort of myth that the 2016 polling, for instance, was wildly off.

Nate: Yeah, 2016 was not great, but not terrible,

Nayeema: But it’s a binary outcome. What does it mean to be not terrible? It means nothing to voters who are voting one way or another to be less than right wrong.

Nate: Take this from the opposite point of view. Biden is behind in all the swing states. He’s been behind for months. It’s hopeless. No, it’s not, because the polls can be wrong. He has actually a 35% chance. Not bad at all, right? That’s a pretty good chance. It’s an NBA, a three-pointer or something, right?

Ben: I like this new sort of liberated version of you who’s talking to people who understand statistics, and is tired of people who don’t understand statistics. But if I could ask one other sort of possibly annoying question for you. I do think the other thing that, at least what I do when I look at polls, is immediately sort of look into the cross tabs and then often say like, “Ah, this doesn’t make any sense.” What’s your remedy for that?

Nate: I think you should look at cross tabs as being about as accurate as astrology, but there’s a reason for it.

Ben: Are you into astrology, Nayeema?

Nayeema: I actually believe the world is divided into people with 12 types based on the month of their birth. But we should also explain for a moment, cross tabs are looking at very specific voter demographics. So Biden down seven points with Gen Z, or Trump getting up with 35% with Black voters, et cetera.

Nate: Yeah, so what happens is that you might have a poll that has a thousand respondents, which has a margin of error of three and a half percent, but then you slice and dice it literally into 10 or 12 cross tabs. African-American women might be 50 people in that poll, if they’re representative of the population. And that has a margin of error of 18 points or something like that. Just trying to estimate it offhand. And if you have 20 random sequences anchored around some truth, but if you’re basically drawing a bunch of lottery tickets with small sample sizes, you’re going to have some that create weird patterns. And so this is for ... You know the term midwit? Now I’m really getting a little spicy.

Ben: Do it. Do it.

Nayeema: We love spice.

Ben: I don’t even know that term.

Nate: Midwit is for people who are educated up to a point, but actually just overcomplicate things.

Nayeema: People who think they know more than they know?

Nate: Yeah. Like the Dunning-Kreiger or whatever. Am I pronouncing that right? I’m probably Dunning-Kreiger myself. But look at the top line of the poll, and look at the quality of the outfit doing the poll, and look at their track record. The New York Times/Siena college polls have been excellent polls. They got a lot of things right in recent midterms, like 2018 and 2022 that a lot of pollsters didn’t. The fact that they show Biden losing all these swing states and yet show Democrats doing well among these high turnout voters that vote in congressional elections. I mean, that’s a pretty dispositive piece of evidence, I think. Yeah, look, if it were a matter of like, oh, the crappy pollsters show Trump winning and the good pollsters show Biden winning, that would be, I think, a credible kind of unskewing type argument. But that’s not what we see. I mean if anything, the better polls show Biden doing worse.

Nayeema: I have to ask why do we need all of these polls done at every single minute? I think I read somewhere there were 1500 polls that we will have between now and November. It was a couple months back, so maybe there’s 1300, great glory. But is there a time when we actually know the answer?

Nate: No, we don’t. Okay, so there’s one thing. It’s not like we’re in a zoom to ... Trump is 65% now. Let’s say nothing more changes and he maintains this two point lead in Pennsylvania, and whatever. It’s not going to go to a hundred percent by election day. But no, I agree though. I am sympathetic to the notion that we don’t need as many polls as we have. I think it’s probably right. I mean because if you look at it, the irony is if you look at the more robust averages, I mean the New York Times has one now, or the Washington Post has one. I think ours is pretty good. They’re actually very steady. So it’s almost like you don’t need all these polls. Now look, when Trump gets convicted on felony charges, or you have a debate, or he picks his VP, or after the conventions. But yeah, I mean if we just had a big set of polls seven times a year, you have a designated polling weekend or something, that would be perfectly fine and would give you most of the information that you need.

Nayeema: Or maybe we need to be more French. By that, I don’t mean have a snap election that might put Marine Le Pen in power, but to have a bit of a blackout period of campaign advertising coverage the way France does where there’s just a weekend right before the election.

Ben: Give everybody a second to think.

Nate: Yeah, maybe. It wouldn’t be the worst thing. And I think it probably would violate the First Amendment here. But the games that pollsters play in that last weekend anyway, I tend to think that’s when there’s a lot of ass covering, and a lot of what we call purding, where they don’t want to be seen as the outlier.

Ben: So you’re saying these things aren’t totally objective?

Nate: No, I think science is hard. I mean, look at the replication crisis. Look in academia how most scientific studies are bad. I think polling is actually a relatively rigorous technique, but one reason why pollsters did have a good 2022 is that they figured out that the classic technique of just literally manually dialing people on the phone book, and hoping to get a representative sample, doesn’t work in a world where 8% of people respond to surveys. So instead, these are basically little mini models, and my model kind of averages these models. And modeling is tricky. Modeling requires domain knowledge, it requires making a lot of subjective choices. And that’s why you have The Economist forecast has Trump at 72% or something, and the new FiveThirtyEight has him at 49%, and I have him at 66, or 65, or whatever. These are all reasonable people making these forecasts, and you have dozens of choices to make with the way you build a model.

What you don’t want to have happen is when a pollster declines to report a result because they self-censor. And that’s, I think, been made worse by the mobs on Twitter where a lot of people are like, “Oh, this poll’s terrible. Of course Biden’s not losing by 12 points in Wisconsin,” or whatever. And so that can create suppression of useful evidence, which has had effect some time.

Ben: Yeah.

Nayeema: It’s like a free speech ... That’s so interesting.

Ben: It’s interesting. I think we came in asking about the way in which the discourse and the public conversation is driven by polling, but actually in a way it’s the opposite, right? Pollsters are at risk of being driven by the discourse.

Nate: For sure. Yeah, because it’s a pretty brave thing to do. I mean, look, my former employer, ABC News had a poll in Wisconsin in late October 2020 that had Biden up 17 points in Wisconsin. And they were brave to publish that because they probably knew it was going to be wrong. The average in Wisconsin did overrate Biden, but had him up by six or whatever. 17 is a big outlier. But they believed in the kind of process of polling, and now you’re seeing less of that, and that could lead to some big surprises sooner or later. Maybe Trump wins by 10 points though. I mean, that’s the thing. The surprises don’t always go in the nice way that you’re hoping for as a reader.

Nayeema: If your model still has Trump with better odds close to election night, and he ends up winning, will you be celebrating being right, or sad that the person you voted for didn’t win the election?

Nate: In that sense, my risks are well-hedged, right? And if you want a more complicated answer, it might depend on things like which party controls Congress, who Trump’s vice president is.

Nayeema: I wanted a less complicated answer, Nate.

Nate: I’ll take the fifth, because I don’t know. I don’t know how I’ll feel. And it depends too, on the narrative of the election. I mean, it could be that by the time we do the election, that Biden’s back in front in the model or something, right?

Nayeema: Yeah, I gave you a hypothetical.

Ben: Right? There’s a real lose-lose for you here.

Nayeema: Yeah.

Ben: Biden’s in front, and Trump wins, and you got to move to Canada. It’s just a shame, honestly.

Nate: No, I’ve already won, I think. I’ve already won by building a successful newsletter product, and having an interesting time intellectually, trying to figure out how to forecast this election. I’m doing this because of the rewards along the way. If I were just looking forward to election day, most of those outcomes are bad for me. Most of the time people are mad at me either because Trump wins and I said he’d win, they’ll still be mad at me. Or because Biden wins and I was probabilistically wrong. We can debate what that means. So it’s going to end in tears no matter what. But along the way, I collect-

Nayeema: You collect $200 when you pass Go. Yeah?

Nate: I collect a subscriber base who does understand what I’m thinking. Seriously, that’s why I’m doing it. We got offers to go to some network or something and do this. And I am only doing this because I believe in the subscriber newsletter model. Right? Because that is a way to kind of self-identify the people that are going to get you.

Ben: Well, you’re talking to your people. I mean, I remember when we were talking when I was at BuzzFeed and it was a different moment for you and you were like, “Look, I just got to maximize my return in this particular moment of my life.” So I’m kind of glad that you’re maximizing your satisfaction here. However strange that is, Nate, honestly.

Nayeema: It really is strange.

Nate: In some ways the gamblers are more normal. They kind of have very normie politics a lot of the time, or at least they’re kind of eccentrically normal, or something like that. Vegas, that’s a real cross-section of America.

Nayeema: For sure it is. That’s true. All right, well thank you so much Nate for being with us.

Ben: Really thanks for doing this, Nate.

Nate: Thank you. It was really fun guys.

[MUSIC]

Nayeema: Nate One. What a character.

Ben: Yeah, man. Man knows his own mind.

Nayeema: He does not seem like he is at all upset about upsetting anyone.

Ben: No. I mean he’s always sort of loved trolling in a way, but I do think he now, to some degree, thrives on annoying people who he believes don’t understand probability.

Nayeema: Midwits.

Ben: Midwits.

Nayeema: Are we midwits?

Ben: I believe so. But you can, if you want, just talk to the people you feel really get you/ That’s a viable financial model. In a way he was at the opposite end, which was talking to random viewers of ABC News. And it’s possible that he did not feel totally understood by those good people.

Nayeema: I don’t know, it was very interesting to me because when he described his mission, there was this public service to it. It wasn’t just, “Oh, I want to do the math really well and get the math right.” There was part of that, right? I’m doing this thing, and this is just what the thing is. You can tell me you don’t like it, but I’m telling you the sky is blue, right? So there’s a bit of that he has, and at the same time it’s like I’m just trying to give people information to make micro decisions in their lives. If you like Trump or not. If you’re going to move to Canada or move to the United States. And I thought that was really insightful, actually, this public service mission to the ordinary voter.

Ben: He really, I think has been kind of scarred through his career by, and really rejects the premise that the polls are shaping reality. I think he obviously just finds that incredibly irritating. Maybe he’ll grant that occasionally in primaries. It’s true. And in fact, I think the thing that was so interesting to me is that, to him, the risk is that the pollsters are bullied by particularly people on social media getting mad at them and shift their perceptions of reality. I think he’s built a real kind of carapace to resist what he sees as pressure on him to tell people what they want to hear.

Nayeema: Yeah, but this is part of the broader critique of cancel culture and the concern of free speech. The Elon Musk warriors of the world. This is a Silicon Valley kind of argument you hear. We’re not allowed to say the thing anymore. Do you think that there’s some truth in that, in polling or political coverage?

Ben: Oh yeah. And I think the subtler version of it is just that you want to give the paying customer what they want hear. Not so much that you’re going to be canceled, but just that the market incentives are ... There’s got to be somebody out there right now saying, “Don’t worry Joe Biden’s going to win,” because there’s a market for that.

Nayeema: That’s a really challenging perspective in, as commercial as the business of media is, in a business that’s oriented towards truth.

Ben: Yeah, to a trust.

Nayeema: Democracy dies in the darkness, et cetera. The idea of catering to your customer is not the mission of the organization.

Ben: Not of ours. I guess our belief is that you are in trust over time and if you say crazy stuff and it proves to be untrue that people will stop believing you. Although it is also just worth noting that in the literature of doomsday cults, they never lose steam after doomsday doesn’t happen. People only grow more committed. So I don’t know, but he probably sees himself as a realist on this point. I love that he refused to say whether he would be happy if he was right if Donald Trump won.

Nayeema: Yeah, I know. It was so interesting. I think he would be happy if he were right.

Ben: Depends on some of the down-ballot races.

Nayeema: It depends on the day.

Ben: Yeah.

Nayeema: Yeah, exactly. It’s a very complicated answer.

Ben: Unique American character.

Nayeema: Meanwhile, extremely important. As you know, Allie, our producer was saying she was just imagining Donald Trump refreshing the feed every minute saying, “Is it out yet? Is it out yet?”

Ben: What does Nate Silver think?

Nayeema: What does Nate Silver of the New York Times think?

Ben: He really has personally reshaped how people think about polling, think about politics in America. It’s pretty amazing.

Nayeema: But yet he is saying he’s not changing the outcomes. He’s just telling you how it’s going to be.

Ben: Yeah. And it is also true that if and when Joe Biden wins, he’ll say, “Well, I told you it was pretty likely.”

Nayeema: All right, let’s take a quick break and we’ll be back for Blind Spots.

[MUSIC]

Ben: This week on our branded segment from Think With Google, I talked with Josh Spanier about AI’s impact on the creative industries, and just how worried ad creatives ought to be about their jobs. And if it sounds like it was recorded on a beach in France, that’s because it was during the Cannes Lions Festival last week.

Josh, just down the street here, your colleague Vidhya Srinivasan, Google’s VP of ads just talked about how this business is entering an AI era. What will the rise of AI mean for people and for businesses?

Josh Spanier: So I’m thinking about it very specifically for the marketing process. We need insights. We need insights to make our marketing relevant to people like you and to people in the world, so they actually discover and use our products. So I was speaking with McDonald’s yesterday and they found these insight in part using AI that kids love going to McDonald’s with grandmothers because the grandmothers will treat them to the McFlurry that they always want, where their regular parents may not do so.

Ben: I always suspected that.

Josh Spanier: So they actually turned that into a whole campaign, really cleverly done, inspired by insights that derived from AI. We’re also looking at creativity. So we run many, many campaigns across Google Marketing, and we’re using generative AI to actually scale that across all these platforms, to actually create hundreds and hundreds of variations of ads, which make them more relevant and more engaging, and actually do so in a way which actually is scalable, but previously we couldn’t do. And it leads to better business outcomes. And actually our creative teams really enjoy it, because they can express different ideas in different ways and get to market quicker. And along the way, measurement and accountability.

Every CFO, every financial organization wants to know that the dollars you’re putting in, you’re getting dollars more out. With AI, we’re actually speeding up the ability to measure our programs, and we’re using a lot of Google Analytics and Google AI to actually prove that we actually are getting the value from these campaigns along the way. At every stage in the process, AI is actually making us smarter, it’s making us faster, and it’s actually making us more creative along the journey.

[MUSIC]

Nayeema: Today we’re doing something a little different, a little more global, and unfortunately a little less Max Tani for Blind Spots, he’s traveling at the moment. But we have been so consumed with elections in the United States that we think it’s worth casting a light at the broader, to use cable TV terms, decision 2024 around the world where more than 60 countries that represent almost half of the world’s population have or will head to the polls this year.

Ben: I’m a little skeptical of this segment, Nayeema.

Nayeema: Why Ben? Because you’re like an American isolationist, an exceptionalist, and you don’t want to know about the rest of the world?

Ben: No. Very global. I just don’t want this to be a sort of eat your spinach thing. And I’ve never bought the idea that the job of media was to lecture people about things that they were supposed to care about.

Nayeema: Don’t worry, Ben, this is going to be a, don’t eat your spinach, but kind of be aware of what might be on your menu in 2024, in November in the United States. Because sometimes outcomes around the world, they give you some indicators of where the mood is, where the vibes are, let’s say. And that was the case with Brexit, and Donald Trump in 2016.

Ben: Yeah. No, and it’s really one of the great blind spots of American media actually. This notion that you can try to explain, in this case, the rise of the populist right in an American context, when it’s such a global story.

Nayeema: Exactly. So already this year we’ve seen major elections in Mexico where they’ve elected the country’s first Jewish president, South Africa, where the ANC struggled. Cyril was able to hold onto power, but ANC struggled. India where Modi held onto power, but actually lost seats in the parliamentary elections, although he would not necessarily admit any defeat. Prime Minister Modi.

Ben: Yeah. Somewhere the Nate Silver of India is telling people, “I told you so,” because his model said that there was a 4% chance that Modi would lose this badly.

Nayeema: That’s true. And in Indonesia Prabowo Subianto won. Do you know who that is, Ben?

Ben: I don’t.

Nayeema: Prabowo is the son-in-law. He’s a country’s defense minister. And someone I know a little bit, who grew up in Indonesia when his father-in-law. Suharto was ousted from power in the late ’90s-

Ben: Is this a very elaborate way to get to a name-drop?

Nayeema: No, no. Don’t worry. If this was a name-drop, my flex would not be Prabowo Subianto.

Ben: Kind of impressive.

Nayeema: Prabowo Subianto, who was the country’s defense minister, who was for a long time on the human rights persona non grata list for his actions in East Timor, might still be actually, but he was the son-in-law of President Suharto. And it’s kind of a nostalgia in Indonesia, but also around the world for a dictatorship of the past. These countries have experimented with democracy. It hasn’t necessarily proved out that much better for them. And you saw this in places like Libya. You’ve seen this in Indonesia, where people want the sons and son-in-laws of dictators back in power.

Ben: Marcos is back in the Philippines. Yeah.

Nayeema: Yeah. There’s something happening. And then around the world, on the heels of this episode, there’s some pretty big elections happening. In France where, after the far right swing in the EU parliamentary elections, Macron has made a very risky bet. He’s called for this polarizing snap election that many think could hand the country to Marine Le Pen and the far right.

Ben: Man loves to roll the dice. I mean, incredible story actually. I actually have covered him a little, interviewed him once. And it is every reporter’s dream to cover somebody who loves to gamble as much as Macron loves to gamble. He loves to gamble more than Nate Silver loves to gamble.

Nayeema: We should introduce Nate Silver and Macron. They could go to a little betting table together in Vegas.

Ben: I was in Paris last week and visited the office of Reporters Sans Frontieres, which is a great crusading journalism organization, whose executive director Christophe Deloire tragically died recently. And there was a big wreath of flowers from Macron, and from Madame Macron, sitting there in the entryway. And I was with a very prominent French journalist who took a look at that and said, “Oh, well, when I die, I hope that the president and Monsieur Le Pen will send flowers.” And then everyone was like, “They’re not going to send flowers.”

Nayeema: They’re not going to send flowers to journalists. Another place where they’re not probably going to be sending any flowers to journalists is Iran, where presidential elections are going to be held on the day that we’re releasing this episode, Friday, after the unexpected death of Ebrahim Raisi in the helicopter crash last month. Look, ultimately, regardless of who wins, we know that the real power in that country rests with the supreme leader. It is an important election to see where the country is going. This is a country who is embattled in a lot of proxy wars, including a very important one to the US. And so that will be an important election to pay attention to. And then on July 4th, as the Americans celebrate their independence from the Brits, the UK will be holding a general election and it looks like Prime Minister Rishi Sunak and the Tories are expected to be voted out after so many years, and the Labour Party taking power for the first time since 2010.

Ben: And I think when you’re looking at sort of storylines and the opportunity to see that, to look to the future in other countries, I mean, I think that’s what Macron is looking to, and to some degree what US Democrats are looking to, which is, all right, let the right take over. Let them do something like Brexit. And in Britain really I think, proved that they were unable to govern, or at least most of the populace left thinking, “God, these clowns cannot govern,” and really get wiped out. And that’s what Macron’s hoping for in France, I think, is that if the right comes in, they will quickly show themselves unfit. I’m not sure that’s going to happen.

Nayeema: Well, I don’t know. There’s a couple ways to look at these elections if we think there’s any kind of crystal ball in them, and I don’t know that there is. But one is to say, “Look, the Brits are ahead of the curve and the pendulum shifted all the way, and now it’s coming back.” And maybe Americans will have that too, right? Pendulum shifted to Trump, and now it has come back and will it stay back? Or will it go the way of what we’re seeing in so many of these countries where the incumbents, even people as powerful as Modi are losing seats? There’s this anti-incumbent fervor.

And people are upset across the globe around many of the same things. They are pissed off about the economy. They don’t trust the media or institutions that much. There are real conversations about gender and family that are happening across the world right now, including in the United States, but all over the world. The role of men is a big issue in elections in places like India even. And then the kind of anti-outsider perspective, immigration, ethnicity, et cetera. The real populist dynamics are playing out. So a crazy global election year. Was it too much spinach, Ben?

Ben: Probably. But I do think that it’s a useful ... I don’t know, that if you’re getting sick of looking at the polls and just staring into the sort of furor of the US election, a good way to understand it is to look and read around the world, listen to podcasts around the world, things like that.

Nayeema: And pay attention to what was that thing at Semafor you wanted to plug, Ben?

Ben: Oh. And do pay attention to the Semafor Global Election Hub, which basically does this with a good amount of voice and insight.

Nayeema: That’s the best way to eat your spinach, I’m told. All right. Thank you for listening to Mixed Signals from Semafor Media. Our show is produced by Max Tani, Allison Rogers, Alan Haburchak, Sheena Ozaki, Benjy Sarlin, and Andrea López-Cruzado. With special thanks to Britta Galanis, Chad Lewis, Rachel Oppenheim, Anna Pizzino, Garrett Wiley, and Jules Zirn. Our engineer is Fernando Arruda. Our theme music is by Billy Libby. And our public editor is who? The Spinach Lobby?

Ben: Maybe it’s Alastair Campbell this week. We’re looking at Britain.

Nayeema: Oh, what? You just have such a crush on Alastair Campbell. You want it to be him.

Ben: I suppose I do. The Rest Is Politics.

Nayeema: Okay. Stop plugging other podcasts, Ben. Our public editor is the house, and the house always wins, they say. If you like Mixed Signals, please follow us wherever you get your podcasts. And if you really like us, give us a starry review.

Ben: And if you’re watching on YouTube, give us a like and subscribe to Semafor’s channel.

Nayeema: And if I were, Max Tani, would tell you to tune in for the Semafor Media newsletter. So I’ll tell you to tune in to Max Tani’s Twitter. It’s great.

Ben: And hopefully Max Tani’s Twitter will tide you over next week because Mixed Signals will be off to celebrate the 4th of July. Or in Nayeema’s case, the British elections. We’re back after that.