This interview was originally published on Audible.com.
Note: Text has been edited and does not match audio exactly.
Katie O'Connor: Hi, listeners. I'm Audible Editor Katie O'Connor, and today I'm honored to be speaking with Nate Silver, statistician, author, poker player, and founder of FiveThirtyEight. Nate now publishes on his blog, Silver Bulletin—great name by the way—and is releasing a new book called On the Edge, a sprawling and ambitious examination of how successful gamblers and risk-takers think. Welcome, Nate.
Nate Silver: Hey, Katie. How are you?
KO: I'm doing all right. So, not a lot going on at the moment for you, huh? [laughs]
NS: Oh, my gosh. I had a weekend, I don't know when this is airing, but I had a weekend in upstate New York that I thought would be my kind of last relaxation weekend of the year till November, and then the president chooses to step aside and Kamala Harris wins the nomination in 24 hours, basically, so yeah. Good for book sales, I guess, and newsletter subscriptions, but bad for my sleep schedule.
KO: A little bit, yeah. So, to set the stage for our listeners, we are speaking in the week that Joe Biden has announced that he's dropping out of the 2024 presidential race and has thrown his endorsement behind his VP, Kamala Harris, and also the night after President Biden finally spoke to the nation. What do you make of how quickly the Democratic Party has rallied behind Kamala?
NS: Look, part of the book is about making good decisions under difficult and risky circumstances, and I think they made the best decision under the circumstances. Biden had fallen behind about four points in our polling average nationally, maybe more like five or six points in states like Pennsylvania. It's not ideal circumstance. I think Democrats would've been better off if Biden had chosen earlier not to run for another term, but, look, when people face high-stakes decisions, they sometimes behave pretty rationally, and Democrats I think have improved their chances of defeating Trump with Harris instead of Biden.
KO: In the opening of On the Edge, you shared that you find election years to actually be quite alienating. You've been publicly working on elections with your renowned election model since the 2008 presidential election. Has this election cycle felt similarly alienating for you or has it felt different in any way?
NS: I mean, it's a very important election for the country. I think I've had a little bit more fun with it. I like being back independent on my newsletter, Silver Bulletin. The fact that we now in Kamala Harris have a relatively younger candidate by political standards running, there's less of a sense of inevitable doom than there has been in the past. We are not in the middle of a pandemic as we were four years ago. I think it's okay in politics to treat politics with a little bit of irreverence and humor and appreciation for the whimsical nature of all the crazy things that have happened, and I'm trying to have a better attitude than I did I think when I wrote that in the book.
KO: I like that, a little bit of evolution there for you.
NS: Yeah.
KO: So, you're a poker player and you actually used to be a professional poker player before you launched FiveThirtyEight. And part one of On the Edge is about gambling and your first two chapters are dedicated to poker. So, just at its base level, I'm curious about what most appeals to you about the game itself?
NS: It's the combination of being able to be really analytical and competitive. I mean, I would never deny that I'm a competitive person. And it's also the mix of kind of math skills and people skills. It really is right down the middle in terms of a game. On the one hand, you can apply game theory and all these advanced statistical techniques to poker. On the other hand, it is a game played person against person. Things like physical tells do matter, things like performance under pressure do matter. And I just find that I get along really wonderfully with poker players for some reason. They're like my chosen people, basically.
KO: Your tribe.
NS: Yeah.
KO: I think that's a nice segue. I want to talk about your mental landscape for this book for a second before I ask my next couple of questions. You call it “the river,” an ecosystem of people and ideas, and people that live in the river possess many traits, some of which you just mentioned, including they're analytical, they're great at decoupling or blocking out context, they're competitive, and they're risk-tolerant. And then on the other side, you have “the village” that's competing with the river for power and influence, and these are the people that are in government, in the media, in academia with sort of left-of-center politics, who are not as good at that decoupling, who are definitely more risk-averse. Did your brain sort of always split people into these two factions before you started working on On the Edge, or did that become more pronounced while you were doing the research for this book?
NS: Oh, for sure it was more organic, I think. The initial trip for the book was the first trip I took by plane since the pandemic, where I flew down to Florida for a gigantic poker tournament in April 2021, which was still in the midst of what was technically the pandemic and people were playing with masks and glass dividers at the tables and things like that. And yet the tournament did record numbers because there are crazy people like me who feel like they need some challenge, need some risk in their lives, I suppose, and had felt cooped up.
"The theme for me is people, at least in the United States, are losing trust in institutions, in political institutions and media and scientific institutions and business institutions, so we kind of, I think, are forced now to confront risk and make decisions for ourselves without having a lot of handholding."
So, it kind of radiated out from there where, like, I didn't realize how common between these different categories—the way poker players think is pretty similar to how people who work in finance think or how people who are trying to figure out crypto or AI or topics like that think, and it's very different from this politics world that I do sometimes feel alienated by. It's very individualistic, sometimes to a fault, where people become kind of annoying and contrarian and things like that and don't know their own limits. But to discover that they are kind of a tribe of people that think this way and are complicated people and flawed people in a lot of ways, and in some ways they're kind of ruled by their demons but are also sometimes very successful and think differently—that emerged organically. I talked to 200 people for the book. I spent three years reporting and writing it, so that wasn't something I expected going in.
KO: This poker tournament that you played in Miami in 2021, when you're talking about it in the book, you're also talking about how our society is getting more extreme. You say, quote, that, "People are becoming more bifurcated in their risk tolerance and it's impacting everything from who we hang out with to how we vote, and that COVID itself has made these risk preferences public." So, again, you say at this tournament and then in part two of your book, I believe it's in the chapter on Sam Bankman-Fried and cryptocurrency, you share that Bitcoin actually peaked with COVID, which I find so fascinating, but it also makes total sense. Do you think that this bifurcation would've happened in this moment without COVID? Do you think that COVID was sort of the ultimate catalyst, or is there something else that would've gotten us there?
NS: I think COVID accelerated a lot of things, including people feeling like they were left to their own devices a lot. Advice on things like mask-wearing changed various times throughout the pandemic. You had to kind of chart your own course, and none of the choices were comfortable. No matter what you were doing, you were either taking a risk with your health or taking a risk with maybe your livelihood and your freedom and your happiness.
KO: Sanity.
NS: Sanity, yeah. I think COVID, and people almost don't want to talk about it now, I think it was, and should've been, I guess, a traumatizing experience for a lot of people. Millions of people died around the world, life was disrupted in a way that it hasn't been since World War II, and, again, the theme for me is people, at least in the United States, are losing trust in institutions, in political institutions and media and scientific institutions and business institutions, so we kind of, I think, are forced now to confront risk and make decisions for ourselves without having a lot of handholding. And people react to that very differently. Some people kind of have a fight-or-flight reaction, some people freeze and kind of lock up and can't do very much at all. So, I mean, it was a world, a moment, when I think we realized the world was a world where we're not going to be coddled and we have to make decisions and we have to assess risk whether we like to or not.
KO: And I'm sure kind of contributed to the impetus of the "why now" for you with this book, like this is the moment then to write this type of story. But as you said, you did spend three years working on On the Edge and had a front row seat to a lot of history while you did so. You mentioned you conducted over 200 interviews. So, COVID, the Bitcoin bubble, OpenAI. How did these moments change the landscape of your book as you were working on it?
NS: I think I got very lucky to choose this topic. I hadn't expected AI to become such a big conversation. When I first talked to Sam Bankman-Fried, I hadn't expected him to become one of the most infamous people I guess in the history of the modern world, or at least the modern economic and finance world. One of the great things about writing a book is that you get a little bit of a time capsule of yourself, so you actually get to live—I was kind of in Miami at the peak of the first Bitcoin bubble. I was in the Bahamas talking to Sam Bankman-Fried. I talked to Sam Altman, the CEO of OpenAI, in I think it was August 2022, which is right when they were internally developing ChatGPT 3.5 and version 4 and were very excited about it, but he was a little bit more unguarded. I was there for the big explosion of sports betting, if you're a sports betting fan, in New York and other states. So, I think I really lucked out, but it does mean that it's an active book. I'm traveling a lot in the book, I'm talking to people, and like you said, it kind of gives people a front row seat to everything, I hope.
KO: These conversations that you had with Sam Bankman-Fried were wild. You mentioned that you also got to speak with Sam Altman. Is there one interview in particular that really stands out in your mind or perhaps someone that you would want to have as a dinner guest now, after the fact?
NS: Some of the interviews that were most fascinating were with people that are not in the quantitative risk-taking category, because most of the book is about poker players and venture capital and hedge fund guys that are kind of working with numbers and spreadsheets. But I also talked to some physical risk-takers, I call them. One of the people who was most impressive in the book is a woman named Kathryn Sullivan, who was one of the first female astronauts in the first class along with Sally Ride that NASA allowed women to become astronauts, and she's amazing. She was studying oceanography in Nova Scotia and was one of like eight people chosen out of 10,000 candidates because she's extremely levelheaded and extremely cool under pressure. And so hearing from people like her, understanding that in outer space not everything's going to go to plan, and how that parallels or does not parallel people under pressure of financial stress or a poker tournament. She was extremely interesting and impressive.
"Can you handle stress? Can you handle big moments? The good news is that people can learn."
KO: Yeah, I did appreciate that section on physical risk-takers too because it is similar parts of your brain, but they are different types of risk at the end of the day.
NS: Yeah. I think realizing that physical risk or financial risk, people are having a stress response, which is actually helpful. It means we're kind of paying more attention to our environment. Athletes use the term "in the zone" sometimes, you know, your heart rate goes up and your pupils probably dilate and you're having a nerve response. One basic lesson in the book, in poker tournaments some decisions are thousands of times more important than others. So, can you handle stress? Can you handle big moments? The good news is that people can learn. As you kind of get more practice in this alternative operating system when you are under stress, you can gradually get better, but that's the way life is, right? I mean, we like everything to be meticulously planned out and plodding along carefully, but what happens is usually we have a few big high-stakes decisions in our lives once every couple years, and how we perform in those situations outweighs a lot of smaller things, so getting to speak with experts who have experienced this in different ways was fascinating and fulfilling.
KO: You talk about this earlier on in the book, this idea that a bigger chemical response can correlate with a better ability to handle risk, and you also use it as sort of an argument against using AI in those moments because AI can't mimic that type of chemical response. You definitely have your reservations about AI on the whole. There are some nerves there, for lack of a better word, but ultimately, you don't want to hit the stop button, as you say, on this technology. You are pro–moving forward. Can you talk a little bit about your concerns with this industry, but ultimately why you are in favor of that forward progress?
NS: Part of it is that, like you were saying, Katie, physical feedback is important. Human beings have evolved for hundreds of thousands of years based on reading clues about our environment that we learn through our body and our psychology, so I think it's a very kind of constrained form of intelligence that we're talking about with AI. AI can solve math equations or do lots of things with language. It still can't fold clothes, right? Still can't do your laundry for you. Maybe that'll come along in a few years. My concerns are mainly that the technology is progressing fast. Maybe more importantly, that we don't entirely understand what large language models are doing. The inner workings are kind of a mystery. They work way better than people would've expected five or 10 years ago, and that's a little bit dangerous.
Look, technology in general has greatly improved the human condition. You can debate more recent technologies like social media, for example, and I think it's kind of a cop-out to not try to learn how to use these technologies responsibly. So, I think it's too soon to permanently hit the pause button for things like that. I think these industries need to be regulated and monitored carefully.
One theme in the book is that even though in some ways the river is my kind of people, left to its own devices then you can have crazy things happen. You can have a Sam Bankman-Fried-type situation. I definitely would not put Sam Altman in that category, but you need people who have some ability to provide adult supervision.
KO: Something that I did want to bring up with you, actually, was in your chapter on Silicon Valley, you sort of have this aside where you said you had considered making autism a theme in this part of the book, looking at those who obviously presented as on the spectrum. But you're like, "Oh, some of their risk-taking behaviors don't necessarily align." And I was thinking, so, two of my kids are neurodivergent and I was thinking about sensory processing differences and how if you are on the spectrum, there's a 90 percent chance that you do have some sort of sensory processing difference as well. And sensory seekers, in particular, I think fall into that sort of risk-taking category. I don't know, it's just this thought I was having that maybe it is a larger spectrum conversation, but with this other neurodivergent quality of being a sensory seeker that contributes to sort of that riverian side of the brain.
NS: Yeah, I like that term, neurodivergent, maybe more than autistic or Asperger's just because it's a little bit ambiguous sometimes, and if you look at the kind of clinical definition of autism spectrum, there are a lot of components and some people have some components but not others, right? Clearly, people who are in the river are risk-taking, which violates some of the precepts about autism where people are seen as needing a lot of order. The people in the book are definitely unusual people, in terms of both being very analytical and very focused and driven, but also being risk-taking. Some people describe having kind of almost a genetic predisposition to take risks. Despite being very calculating, I think some of them are kind of ruled by their demons in different ways, or in some cases are ruled by traumatic experiences in childhood. I mean, you definitely meet people in the book who I think you would classify as being on the spectrum, but like I said, I like the term neurodivergent more because it encompasses a wider spectrum, kind of pun somewhat intended, of personality types.
"Usually we have a few big high-stakes decisions in our lives once every couple years, and how we perform in those situations outweighs a lot of smaller things, so getting to speak with experts who have experienced this in different ways was fascinating and fulfilling."
KO: So, you also recently launched a podcast called Risky Business—another great name, by the way—covering politics and poker, some news, some personal interest stories, with multihyphenate author, journalist, psychologist, and poker champion Maria Konnikova. You are promoting this book. You have Silver Bulletin. What does a typical day look like for you?
NS: I guess my routine is kind of not having a routine. I travel a lot. I have weeks when I'm out playing poker tournaments, I have weeks when I'm writing or promoting a book. And so that gives me the capacity to do a lot more work, I think, because it's a different story every day and every week. Being adaptable is pretty important. The one constant is that I've spent a lot of time just writing over the past four years, writing the newsletter several times a week and writing for the book, and I have found that writing is something that's, like any hobby I suppose, almost like physical training. When you write every day then you're able to write more efficiently and more effectively on compressed timescales, and so that's kind of really been the theme is discovering my inner, inner writer more than my math side, I guess, recently.
KO: Oh, I love that. I do want to come back to your election model work again. I found this point very interesting in the book that after the 2016 election, you shared that people were angry with you because your model had Secretary Clinton's chances of winning at 71 percent and Trump's at 29 percent, which was admittedly higher than other major forecasters. But you explained that your job is to handicap the race, it is to tell you how you should bet, and that 30 percent was actually a pretty good chance. So, while you had the politicos angry with you, people from the river were thanking you for the money that they had made using FiveThirtyEight's model to bet on the election.
This is so fascinating to me, and when you turn your election model back on next week with Kamala Harris integrated, I will certainly be looking at it with that in mind. But you do have so many laymen looking at your election model and, of course, people in the village majorly relying on it, perhaps without that betting mindset. Do you feel any sort of social responsibility as a data journalist to be a bit more blatant about this for non-riverians?
NS: Part of why I've moved the model back to the newsletter, where the probabilities are paywalled, technically—people tend to leak them on Twitter and things like that, but the idea is that this is a product that I think should be targeted at more of a knowledgeable audience, because people don't encounter probabilities, or they do for things like weather that are less emotionally driven. I think people are gradually coming to understand the probabilistic nature of forecasting a little bit better. If you're a Democrat or a Republican you've won some races and lost some races, obviously, in recent years. I think people might understand with Kamala Harris that it's inherently a difficult circumstance and a little bit of a long shot bid, but that her chances are probably better than Biden. I mean, she's already polling a little bit better than him, she's raising a lot more money, and in some ways, it was a very I think mature decision by the Democratic Party to understand that like, "This is not a great circumstance, but we have to play the percentages a little bit more."
"People, I think, when the chips are down, actually do behave pretty rationally when they have real decisions to make."
But yeah, look, it's difficult, I mean, people are firing from a very either emotional place with elections, understandably, because they're a very high-stakes thing, this election in particular, or from a very partisan lens where they're trying to weaponize every piece of information and every piece of political news. People should be prepared for another 30 percent chance to happen at some point. Maybe this time it'll be a 30 percent chance that a Democrat wins and they win and people will still be mad at me, but that's okay. I'm used to it at this point.
KO: You know how to let it roll right off you, right?
NS: I thought about not publishing a model this time, but I do think it provides useful information. I mean, you saw the roll that polls and forecasts played in Democrats' decisions about what to do about Joe Biden, where you saw Biden underperforming, all the Democratic Senate candidates at some point underperforming Harris. People, I think, when the chips are down, actually do behave pretty rationally when they have real decisions to make, and I think they've improved their chances.
KO: One theory that you share is that in financial and career decisions in particular, most people don't often undertake enough risk. If someone is not a riverian, necessarily, but they are risk-curious, what do you recommend their first step be?
NS: Steven Levitt, the U of C economist who wrote Freakonomics, has conducted studies where literally he'd have people flip coins and make decisions on that basis. For example, not just decisions like where to go to dinner. I'll do that sometimes, be like, "Do I want Italian food or sushi?" I'll just flip a coin or something. But in terms of decisions like, "Should I leave an unhappy relationship? Or should I leave a job where I feel stagnant? Should I move to another part of the country?" And he found that when people flip the coin and it says "make a change," that they wind up being happier on average.
People have been trained for thousands of years, more than thousands—hundreds of thousands of years—in an environment where there was no abundance. Instead, we were just trying to survive against violent predators and things like that, and the average lifespan was 40 years. And so our instincts sometimes can be very risk-averse, especially if you're younger and have choices to make. Oftentimes quitting a bad situation is the boldest course of action. You'll find people in the book, the kind of 10 percent of people, the Elon Musks of the world, or Sam Bankman-Fried, who take way too much risk. But I think most people would be better served by being willing to make changes in their life because it's just easier to say, "Well, this is a bad situation but there’s nothing acute about it. I can wake up tomorrow and muddle through again." It takes courage sometimes to make changes, and people are usually happier when they do.
KO: Yeah, and I think it's interesting, too, because some people have such a hard time with change but, to your point, when that change ultimately comes, it can just lead to a whole new outlook, a whole new level of happiness and satisfaction. Another major lesson that you bestow in On the Edge is reciprocity, put yourself in your rival's shoes, and also to play the long game. What long game or long games are you playing right now?
NS: I mean, part of the decision to release a model even though I know that sometimes people might misinterpret it is partly to kind of set myself up for the future. I do run a newsletter, and a model is a compelling product, or upsell, if you will, for people. It's a little bit like a cycle where once every four years, like the World Cup or the Olympics or something, then I understand that every four years I get to be in the spotlight a little bit more. And trying to set myself up in a position where I can pursue creative work, I'm 46 right now, for hopefully a long time to come, my working life. To go big, I guess, this year, to put myself out there and take some risks and have the book come out and the newsletter come out and the podcast come out. I'm operating often on four or five hours of sleep, but that's made with kind of the long term in mind where, if it goes well, then I can have hopefully a happy and creative second half of my life.
KO: I like the idea of that creativity in a second act, but I think certainly there is so much creativity in what you're doing already, and I do hope, though, that sooner rather than later you get a little bit extra sleep, although probably unlikely in the next couple of months.
NS: Yeah. As you get older, you'll have more days you're like, "Oh, I woke up at 5:30. I'm going to go back to bed." And then you never get back to bed, you start surfing the internet, checking your email or writing a newsletter post. But it's okay. I'll sleep on November 6, the day of the election. I'll take a nice 12-hour rest, I guess.
KO: I like that optimism that we'll be done and dusted on November 6. I'll take it.
NS: To be fair, yeah, probably November 8 or 9 or something like that.
KO: Thank you, Nate, for your time today. I really appreciate it in what has been I'm sure a crazy week in an already crazy year. Listeners, you can get On the Edge by Nate Silver right now on Audible.
NS: Cool. Thanks so much, Katie.
