Search the Site

Episode Transcript

My guest today, Annie Duke, is a psychologist turned poker player, turned bestselling author. Her focus in all of these endeavors, how to make better decisions.

DUKE: I had this realization that I never had stopped doing cognitive science because the thing that I was obsessed with in graduate school, which was learning under conditions of uncertainty and uncertain systems, was really what I was doing in poker.

Welcome to People I (Mostly) Admire, with Steve Levitt.

In her latest book entitled Quit: The Power of Knowing When to Walk Away, Annie tackles one of my favorite subjects: quitting. All sorts of systematic biases have been laid bare by the field of behavioral economics. But in my opinion, not quitting soon enough is the most costly mistake people are making in their everyday lives. But I haven’t had much success in convincing people that’s the case. Hopefully Annie Duke is more persuasive than I am.

LEVITT: So Annie, a couple of times on this podcast, I’ve offered my guests one of the highest compliments I can give. I’ve called them a quitter. How would you feel if I called you a quitter?

DUKE: Well, knowing you, I would feel complimented.

LEVITT: Because you’ve just written a new book and it’s called Quit: The Power of Knowing When to Walk Away. And I hope a lot of people are going to read that book and they’re going to come to realize that it is a compliment to be called a quitter, precisely because we live in a world where so many forces push us to persist far too long at failing endeavors.

DUKE: Yes, exactly. What are the aphorisms? Quitters never win. Winners never quit. If at first you don’t succeed, try, try again. You know, everything is just quitters or losers. I don’t think that people quite appreciate how valuable the gift is that we have that is this ability to quit. Because when we start anything, we’re making that decision to start it under conditions of uncertainty. And there’s two sources, right? One is just plain old luck. So you could choose something that’s going to work out for you 80 percent of the time, but that means 20 percent of the time it’s not going to work out for you. So that’s one problem when we start things, how it turns out is probabilistic. And the other thing is that for most decisions we make, we know very little. And I think we all have that feeling, probably every day, of I wish I knew then what I’d known now. So here we are as decision makers in this awful position of having to make a decision where we’re making these educated guesses, but guesses, nonetheless, about how things are going to turn out. So the good news for us is that when we do find out this new information that, hey, things aren’t really going our way. Or I found out this new information that would’ve made me make a different choice in the first place, we can quit what we’re doing and go do something else. So it’s the thing that makes, really, decision making under uncertainty bearable. But I don’t think we appreciate that gift.

LEVITT: Yeah, so I agree completely. To summarize the basic argument in your book in one sentence, it would be that people stick with bad things, almost always, for too long. And we’d be better off if we quit things sooner. And if I understand it, the goal of your book is to get people to actually change their behavior. But one thing we know is that behavior change is hard, right?

DUKE: It’s very hard. There are so many cognitive and motivational forces that are working against us that generally we’re quitting things too late and that’s causing us to stop our progress. When the world tells you that it’s time to switch to something else, if you don’t do that, you’re going to be sticking to something that isn’t allowing you to gain ground towards your goals. But if you quit when the time is right, that’s going to speed you up. Because that’s going to allow you to switch to something, another opportunity, and really defray those opportunity costs. One of the things that I’m trying to do in the book is create motivation for people to actually try to get better at this. But just knowing about it doesn’t really help. You have to actually put tools in place that will allow you to be a better quitter. And the overall theme of the tools is, think in advance; don’t trust yourself to be good at making decisions in the moment. We have to start to be good advanced thinkers about when we should and shouldn’t quit.

LEVITT: So I think, actually, you and I believe the same thing, but we have different ways of interacting with the world. I think it’s completely obvious you should quit. And I stop there and I figure, well, everyone else should just know that too. And what you’re actually encountering is the fact that you and I, and maybe like seven other people on the planet, think that quitting is a good idea and you’re trying to meet them halfway and pull them along gently to hold their hand and take them with you to make them true believers. I don’t have the patience for that. But I’m glad that you do.

DUKE: I think you’re actually right. What I’m really trying to achieve in the book is to translate the science for people. And, it’s part, “Here I’m going to give you these tools that are going to help you be better at it.” But also: “I’m going to let you know that the problem exists.” Because I’m not sure that people know that they have this problem. I don’t think that most people do think they have a problem with quitting too late. 

LEVITT: I think most people completely mistakenly think they have the opposite problem. They think they quit too much just because they’ve been socialized so brutally and completely to believe all those things you’ve said, like how quitters never win and winners never quit. So I think there’s a lot of therapy and counseling you have to do if you really want to get people to believe in quitting.

DUKE: Somebody wrote me who knows about this work that I’ve done. And she said, “Thank you so much for giving me permission, because I just quit my job where my boss was toxic, and this had been going on for a long time. And I had been thinking about quitting, but I was unwilling to do it because I thought it would be a failure on my part. I thought that in order to show character, that I needed to stay in this situation. But now that I’ve seen what you’re saying, I quit today.” And first of all, I’m so happy that that’s the result for this woman. But also, doesn’t that say how negatively we view it? How much we think that quitting is synonymous with a lack of character and sticking to things is synonymous with character itself?

LEVITT: so I want to talk about grit versus quitting. Grit, of course, is a very popular book by psychologist Angela Duckworth, who also co-hosts the podcast No Stupid Questions, and her books about the value of perseverance. And, naively, it seems like grit and quitting might be polar opposites. But I actually think when people say that, they’re totally confused. So let me give you my take on it and then I want to get your reaction. Let’s say you’ve got a long term goal: you want to write a new book or win a gold medal in the Olympics, or quit smoking, get a Ph.D., finish a triathlon, whatever dream excites you. In pursuit of that dream, there will likely be many tasks along the way that are painful or boring or exceedingly difficult. And grit, to me, is the ability to persevere through the tough patches in pursuit of a worthwhile goal. Okay, but quitting on the other hand, it’s something totally different. It’s about assessing whether our goal is actually worthwhile. It’s about recognizing early that things are unlikely to turn out well, even if you are gritty in pursuit of that goal. It’s the awareness that time spent pursuing the gold medal you’ll never win, or the Ph.D. that will lead you to a job you don’t really want, that that time could be better spent pursuing another dream. Applying your grit to something that actually has a higher expected payoff. So, at least to me, you can be the grittiest person on the planet and be a serial quitter. What do you think of that argument?

DUKE: People have asked me this question like, “Are you more quitty or gritty?” And I said, “I’m both.” Obviously I’m gritty, I’ve done a whole bunch of stuff. For one thing, I wrote a whole book; that takes grit. But I’m a total quitter, for just the reasons that you said. I think I feel a little defensive for Angela Duckworth. Everybody interprets her as saying, just stick to things. That that is how you build character. And she would agree that that’s totally absurd. Sometimes you do decide that the goal that you’re pursuing is worth it, but that the particular road you’re on in trying to achieve that goal is not the right road. You’re not using the right strategy; you’re not using the right tactics. And I would say that book writing is actually a really good example of this. I have had book ideas where I kind of started working on it and I was trying to write the proposal and then I decided, you know what, this isn’t the right book to write. And so I’ve tossed that. But also when I’m thinking about even the book Quit or Thinking in Bets, which did get published, I did tons and tons of quitting along the way. There were whole sections that I wrote that I ended up throwing out. There are whole narratives that I wrote that never appeared in the book, ’cause I quit them. So I think about that not so much as, like, walking off the court, but as changing direction. As deciding that the particular defense that we’re using isn’t working, so we’re not going to leave the court; we’re not going to quit the game, but we’re going to change what we’re doing on the court. And I think that that’s a really important form of quitting as well, that we overlook.

LEVITT: I like what you said about classifying yourself as both gritty and a quitter, because that’s what people should strive for. People who are both gritty and quitters, they shine because they work hard, but they work on the right problems. And the very saddest cases I see are extremely gritty people who haven’t learned how to quit. And it’s so common in higher level academics because most people who enter Ph.D. programs, they have no idea what they’re getting into. I mean, I know I didn’t. And it turns out that liking a subject in college and being good at it, that just isn’t a very good predictor of being good at doing academic research, or, more importantly, enjoying doing academic research, doing what academics do. And I see so many students sticking with Ph.D. programs when they should have quit years earlier, and so many young people who get their Ph.D.’s, then they pursue academic careers, even though they know at some level it’s the wrong choice. They just feel locked in. And that’s actually a fate you narrowly escaped. Right?

DUKE: Well, I did five years worth of Ph.D. work at the University of Pennsylvania, and at the end I got sick. And I ended up in the hospital for a couple weeks with a stomach illness that was pretty chronic at that point. And I really actually needed to just take time off in order to take care of myself. And it happened to be right when I was going out for all my job talks, so I had to cancel all my job talks. And it was during that time that I found poker out of necessity, because I needed to make money. So what ended up being something that I wanted to do in the meantime to make money turned into an 18-year-long career that was actually quite successful. Whenever you’re pursuing something, that means that there’s a whole bunch of things that you can’t pursue, and all of those other opportunities that you’re foregoing have gains and losses associated with them. The problem is that when we’re pursuing something like a Ph.D., we are so focused on the path that we’re on, that we actually completely neglect the opportunity costs — what are the things that we could be doing instead that would make us really happy or allow us to gain ground toward what those broader goals are that we have in our life? Where, I assume that if you’re trying to pursue a Ph.D., it’s not so much about the Ph.D., but something about fulfillment and happiness that the Ph.D. has become a proxy for.

So when we’re forced to quit, we now have to go explore all those other opportunities that we’ve been neglecting in service of pursuing the goal that we’ve been trying to achieve. And that’s certainly what happened to me. And then I found something amazing because of it. Should it take an act of forced quitting to get us to explore the other options that we have, the other opportunities that we might have available in our lives, or should we be trying to incorporate this type of exploration without the world foisting that necessity upon us? During Covid, Phil Tetlock, who wrote Superforecasting reached out to me and asked me to do some research on counterfactual forecasting with him. And so we did four pretty large scale studies and then at the end of it, he said, “You know, you could just write this up and turn it into a Ph.D.” So I am now currently enrolled as a graduate the University of Pennsylvania, and I should be defending my dissertation in the spring. But I think that’s also a really important lesson about quitting, is that we think that quitting means closing a door permanently. But usually the door is still unlocked. You know, and if you decide that you want to go back and do it, you can usually open that door back up and head right back in.

LEVITT: We had a student at the University of Chicago who had taken his qualifying exams in 1960 and he had gone and had a wonderful career and about 50-years later he decided he wanted to come back and actually complete his Ph.D. And there was a lot of confusion around what the requirements were and whether he was still grandfathered in and could do it. And ultimately they decided that if there were three faculty members who would be willing to be on his committee, then he would be allowed to pursue his Ph.D. It turned out there was one faculty member D. Gale Johnson, who had actually been on the faculty 50-years earlier and remembered the guy who wanted to do it. And then Gary Becker and I also said that we’d be on this. And it was really fun to watch this 70-something-year-old man, who had become a big player in the strawberry industry, write a really great thesis 50-years later on strawberries, and come back and defend it. Compared to our usual students who grumble all the time, this was a guy who, there’s nowhere else on the planet he would rather be than writing up this dissertation. 

DUKE: I have to requalify, but that’s okay.

LEVITT: It’s interesting that you end up writing a book about quitting when it seems to me quitting didn’t really come very naturally to you. Having spent some time around Danny Kahneman, the Nobel laureate behavioral economist, he’s terrible at making decisions. And I once asked him, I said, “Danny, how can you spend your life studying decision making and still be so terrible at it?” And he said, “Well, I’m good at studying decision making because I’m so bad at making decisions. It gives me intuition.” Do you think your own limitations at quitting — you’re forced quitting — do you think that helps you communicate about quitting better than somebody who naturally was a quitter?

DUKE: I think maybe I have more empathy for it because, you know, it did take that act of force quitting to get me to quit graduate school. I grew up with a father who prided himself on never missing a day of work, no matter how sick he was. And this is what was instilled in me. He would play tennis matches where his whole body would start to cramp up. And he would finish the matches. He would go to the hospital, get electrolytes, and come back for the finals the next day. I, myself, am a tennis player. And I have to confess, I have played district championships with a broken toe, through a lot of pain. It’s a little bit how I’m built, and it’s a little bit how I was raised. And when I left graduate school and became a poker player, even though I was so successful, I always felt some sense of shame for having quit. Like a pretty strong sense of shame.

LEVITT: So you suffered with two decades of guilt about making a decision to quit that was obviously a really good decision. And I think that is what is so awful about what our society has done with quitting even someone who’s thoughtful like you are, and who thinks about quitting, is forced to carry this baggage of shame. That’s the real problem. Problem isn’t just that people quit too late; it’s that people suffer when they quit. That’s what I loved about your book is it is exactly permission not to beat yourself up for the next 20 years if you decide to quit.

DUKE: That’s part of why I wrote it, because I wanted people to stop getting that message that it’s not okay to quit, that I think was very deeply instilled in me.

We’ll be right back with more of my conversation with Annie Duke.

*      *      *

LEVITT: So my son was playing soccer and he wasn’t very good at it. In fact, he was terrible. He was about nine. And on the way to one of his soccer matches, we got into a conversation and I said, “How good would you say your team is in the league? Like, how does your team rank among all the teams in the league?” And he said, accurately, “Well, I’m on the worst team in the league.” And I said, “How would you rank yourself in terms of skill among all the players on your team?” And he said, “Well, I’m the worst.” So I said to him, “If you’re the worst player on the worst team in the league, what do you think that says about you?” And he said, “I think it says I’m really bad.” And I said, “Don’t you think it says you should quit?” And he said, “Yeah, I probably should quit.” And that was the last soccer game we ever went to. Most parents would probably not use that strategy, but I thought it was right. And interestingly, he later, the same kid who was so bad at soccer, went on to be one of the top 10 fencers in his age group in the U.S.A., and quit. Quit at the top. Said, “I don’t like doing it.” And it was hard to watch ’cause it was actually really fun for me to go to his fencing tournaments, and fun to have a world-class athlete as a son. But when he said he wanted to quit, what could I say except for, “Fantastic.” He must have wanted to quit so badly to walk away from this thing that was giving him so much identity and so much positive feedback. He must have been so miserable. And I knew immediately that he must be making the right decision.

DUKE: That’s so amazing. If your son had said to you, “I know I’m the worst player on the worst team, but I love it.” Right? Then I think that that’s okay because I think that these things are a matter of values, right? Like, what are the things that you value or not? I think we need to recognize that there’s no objective way to determine as an observer, is this something they should continue to pursue or not? Except, I mean, there’s extreme cases, like you’re in the middle of a snowstorm on top of a mountain, turnaround. But in these cases it’s more nuanced than that, and it has to do with what someone’s own values are. If my child had said to me, “I’m the worst player, literally in the league, but man, I totally love it.” I’d be like, okay. 

LEVITT: But if your child said, “I know I’m the worst player in the league, but I think if I work hard, I’m eventually going to make the Olympic team at soccer, and that’s why I’m doing it.” That’s where I think as a parent, you got to figure out — maybe you let them pursue the dream, but dreams that have no chance of coming true are probably not the best dreams to pursue.

DUKE: One of the best ways to become better at quitting is to think in advance. There’s some pretty deep scientific work, particularly by a wonderful scientist named Barry Staw that says, “When we get signals from the world that the path we’re on is not working out, that we ought to quit what we’re doing, we don’t actually pay attention.” As you described with the Ph.D. student, who sort of keeps going, even though they’re unhappy, we know that intuition is just plain wrong. In fact, when we get those negative signals, we’ll escalate our commitment to the losing cause. In other words, we’ll double down and triple down on the path or the goal that we’re trying to pursue for a variety of reasons. One of which would be, we don’t want to feel like we wasted our time up until this moment. So if we know that we’re not going to pay attention to the signals, one of the things we can do is as much as we possibly can, determine those signals in advance. Okay, I’m going to start this thing. Let me imagine that things go quite poorly. What are the signals that I’m going to be seeing that would tell me that I ought to walk away, that things aren’t going well? And we write down a list of those signals and we call those “kill criteria”. Doing that advanced work will actually get you to be much more likely to pay attention to them and quit earlier than you otherwise would have, which is really good for you. ‘Cause if things aren’t going well, you want to quit as quickly as possible so that you can go do something that would be better. So, like, in the case of your child, if they’re playing soccer and they say, “Yes, I’m the worst in the league, but I’m going to make the Olympic team,” which would be, overly optimistic. And over optimism is one of the things that stops us from quitting. So we want to sort of help our child through that. We can say, “Okay, I’m totally with ya bud. You have a dream of becoming an Olympian. Maybe that can actually happen for you. What do you think would have to happen in the next year, or the next six months, or during this season that would tell you that you’re getting way better at soccer and that eventually making the Olympics is actually like a realistic goal that you ought to keep pursuing?” And you can work that out with them. Like, how many goals do you think you would have to score? Or, you know, if they’re goalie, how many stops would you be making? An example of this that’s really simple, would be like, if you’re thinking about letting somebody go who’s in your employ, which is something that we know that we do too late and firing somebody is quitting. When you have that moment of thinking, maybe this person isn’t working out. Instead of assuming that you’re going to be rational about making that decision to let the person go, sit down with them and say, “Alright, like, I think we can both agree that things aren’t working out.” They’ll say, “I know I can turn it around.” And you’ll say, “I have no doubt that you can turn it around. So let’s look at what does turn it around mean? Imagine it’s six weeks from now and your performance has really improved. What does that look like?” And you can write those benchmarks down and you can agree that if you don’t hit those benchmarks or if you’re still seeing some other signals that would be adverse or negative, that you’re then going to have that conversation and it’s going to be reasonable for you to terminate the employee-employer relationship. And it just turns out that we’re better at actually following through on that than if we sort of leave it to our own devices to pay attention to those signals when we come across them.

LEVITT: It’s my strong impression that a lot of what trips people up is that they don’t keep track of time very well. The key to me about these kill criteria, is that they’re thresholds with a timestamp attached to them. If I haven’t grown revenues to a million dollars in three years, then I’m not on track to be a billion-dollar company. That three years is really, really, really important because once you get into the thick of it, you always feel like, ah, I’ve got a million dollars in revenues right around the corner, I can just do this, I can just do that. But the problem is you can just do this and just do that forever and never get resolution. And so that, I think, is a really powerful idea that you have to tie your goals to a timeframe to make them more realistic.

DUKE: Yeah. It’s what I refer to as “states and dates”. So the best kill criteria contain a state and a date. If I am not in this state by this date, I must quit. If I am in this state by this date, then I can continue. It’s just essentially just a deadline because the deadline could also be if I haven’t achieved these particular benchmarks after having spent a certain amount of money. So that would then basically put a limit on the time as well. So just as you say, you’re trying to limit the amount of time that you’re in a losing endeavor for exactly the reason that we lose track of time, otherwise, it’s always right around the corner. And I think this brings up a problem with quitting, which is this interesting asymmetry. Like, once we start something, we kind of want to know how it turns out. And this is where we get into this issue, which is that if you want to know how the thing you’re doing turns out, the only way to do that is to keep doing it. Richard Thaler a Nobel laureate in economics, what he said to me is,
“Generally we won’t quit until it’s no longer a decision.” In other words, there’s no hope. You’ve butted against the certainty; your startup is out of money, and you can’t raise another round. You’re in a job with a boss that is so toxic that you have used up all your vacation and sick days and you’re still having trouble getting yourself into work. And I thought that was so insightful to the point of what you just said. The reason why kill criteria need to have a date, you need a timestamp, a deadline, is because as long as there’s hope, there’s always some chance we can turn it around. And that moment where we go from failing to having failed is so horrific for us as humans that we’ll just keep going until we’re certain that we had no other choice. 

LEVITT: So let’s talk about the academic evidence on quitting. Quitting is a hard question to study. What’s the academic research on the subject that you find the most compelling?

DUKE: There are so many, but I would love to talk about one by a certain person named Steve Levitt, if that’s okay? Because this is one of my favorites.

LEVITT: Sure. That’s my all time favorite study I ever did.  

DUKE: Here’s the thing. If you quit on time, meaning at the objectively right moment, it will feel like you quit too early — usually. Do you want to describe the study?

LEVITT: Sure. It was a super simple study. I just built a web page, and on that web page, all we did was we said to people who are having trouble making decisions that we would flip a virtual coin for them. And so we would have them tell us what they were struggling with maybe they couldn’t decide whether to quit their job or to enter a relationship or to get a tattoo or which college to go to. And then we would flip a coin. And if the coin came up heads, they should make a change. And if the coin came up tails, they should stick with it. And what was great about it is that a lot of people came, something like 25,000 people came. And believe it or not, they actually followed the coin toss. Not always, but something like two-thirds of the time, people followed the coin toss. And so what we were able to do was compare people who were thinking about quitting their job and got heads relative to people who were thinking about quitting their job and got tails. And we could follow up in three months or six months later and ask them how life was — was their life better? And we also tried to get them to give us the name of a friend who might give us an even more truthful response about how life was treating their friend. And overwhelmingly what we saw is that the people who got heads, who were prompted, therefore, to make more changes, reported having higher happiness six months later, saying they would make the same decision again given the chance. I found it to be overwhelming evidence that people who thought they were indifferent between making a change or not, it turned out those people actually did much better when they made a change, when they didn’t.

DUKE: This is such great work showing that when we quit on time, it will feel like we’re quitting too early. For the reason that by the time we experience the choice between quitting and sticking as a close call, it’s not actually close at all. So by definition, that means that we’re really getting to that decision too late. I mean, who’s going to flip a coin to make a big life decision except someone who thinks it’s a 50-50 choice. They have to, by definition, believe that they’re as likely to be happy staying in their job as quitting their job. They’re as likely to be happy staying in their relationship as quitting their relationship. So they have to be experiencing this decision as a close call. But what you found was that no, it wasn’t actually close at all. On average, the people who quit were happier. Now, if it were, as they experienced it, an actual close call, then the quitters and stickers would end up equally happy six months later. So what that shows us is that we’re just getting to this decision too late, partly because of a lot of the cognitive biases that make it very hard for us to walk away from things. And that’s what I think is so beautiful about that work that you did is that it shows so clearly that our calibration is just way off on this. 

You’re listening to People I (Mostly) Admire with Steve Levitt and his conversation with Annie Duke. After this short break, they’ll return to talk about why Annie quit poker.

*      *      *

Well, I can’t bring Annie Duke on this show and not talk about poker. She was a true poker celebrity and she earned more than $4 million in live tournaments over her career. She retired from poker a decade ago, but she still remains the fourth highest earning female poker player of all time.

LEVITT: Alright, so let’s talk poker. You spent roughly 20 years playing poker professionally and you had great success in big tournaments. Now, the oddest thing about you as a poker player is that you’re a woman. And back when you were competing, only 3 or 4 percent of the competitors in these big tournaments were women.

DUKE: That’s actually still true.

LEVITT: Is that true?  

DUKE: The absolute number of women are greater but not the proportion.

LEVITT: So do you think it helped you, or hurt you, to be a woman at the poker table?

DUKE: Separate from what was happening at the table, it was easier to get an endorsement ‘cause I was unusual. But in terms of the actual poker play, I think it was quite helpful. Partly because I was unusual, and also reasons having to do with just like, stereotyping somewhat with misogyny. 

LEVITT: So the men you were playing with—

DUKE: Not all.

LEVITT: So many of the men you were playing against, what did they think that you were going to do that you could then turn on its head and do the opposite?

DUKE: There were two main ways that it expressed itself. In one case, you would be playing with somebody who seemed perfectly nice, but they didn’t give you credit for being able to think more than a level deep. What that meant was that they sort of created a one-to-one mapping between your action and your cards. So if you raised, they assumed you had to be strong, because they didn’t really give you credit for the creativity it would require to bluff in that situation. If you check, they assumed you were weak.

LEVITT: And that’s great, right? 

DUKE: Oh, it’s amazing.

LEVITT: What a gift, because it’s a zero-sum game, and if they make such blatant mistakes, you just exploit them.

DUKE: So in that particular case, if you’re playing with someone like that, you can bluff them a lot. Because they’re giving you credit for a hand that’s stronger than it is, and they don’t think that you have the masculinity, I guess, to bluff, like the courage, whatever. There was another type of person who also really disrespected you, but in a particular way that had to do with a threat to their masculinity? So the idea that a woman might outplay them or outthink them would be horrifying. This would be sort of the category of person who thinks, what are you doing at a poker table? This is my space. You don’t belong here. You should be barefoot and pregnant in a kitchen somewhere. And they wanted to assert themselves over you. So that expressed itself as like hyper-aggression. So their goal was like, I’m going to bluff you and then I’m going to show you my cards. That was kind of what they were trying to achieve so that they could dominate you in that way. So also that’s wonderful.

LEVITT: I’m actually surprised to hear you say that you like it when people are super aggressive. ‘Cause I would think that that’s a tough thing to deal with at the table.

DUKE: Well, I think that where most people go wrong is when someone’s really aggressive, they try to be aggressive back. But that’s not actually the right strategy. You can think about it as a pricing problem. If I’m playing in a game, when I think about the hand that I’m going to play, I have some sort of expected return on the hand, because there’s only so much that you’re going to bet. So If I’m playing in a game with people who are pretty tight, where the pots are pretty small, I’m not going to get as much return on any hand that I play because they’re keeping the pot small. So that means that the price for whatever I bet is going to be smaller. But if I’m playing with somebody who’s being very aggressive toward me, that means that they’re going to bet more than somebody else would in their circumstances, which then is going to increase my R.O.I. ‘cause they’re sort of artificially increasing the size of the pot.

LEVITT: So I did play some poker and I decided that I should participate in two World Series of poker events. And the first one was a complete disaster. I literally did not win a single pot. 

DUKE: Oh my gosh. That’s horrible.

LEVITT: I sat there at the table for two hours and I either folded or lost every hand. Undeterred though, I decided to enter another event the next day. My luck is a lot better than it was in that first tournament. I keep on getting into big hands with this guy who’s sitting just to the right of me and I’m beating him over and over. And people at the table are shaking their heads and they’re muttering because I obviously have no idea what I’m doing. I don’t know how to stack the chips and I forget to ante and everything. And this guy is swearing at me repeatedly. He’s insulting me. And finally he barks at me, “Don’t you know who I am?” And I look at him and I say, “Honestly, I have no idea who you are.” And he says, “I’m David the Dragon Pham.” And that was supposed to impress me, but I still didn’t know who he was. Now it turns out he was a two time Card Player magazine Poker Player of the Year. And he’s won over 10 million in live tournaments. But like, I didn’t know who the guy was. And here’s how my story relates to what you were saying, Annie.

DUKE: Oh, it totally relates.

LEVITT: It was obvious from his comment and the reactions of the people around the table that I was somehow supposed to play differently because this was David Pham. I was supposed to defer to him maybe, or believe he had better cards than everyone else. Anyway, it made me wonder whether just being a big name might give you an advantage in the tournaments, making people fear you. But I also think that that works the opposite of the biases that I think many poker players had about women. And so I wonder whether overall maybe it was harder to be a professional woman poker player because you weren’t getting all the benefits that other male professional poker players were getting in terms of fear or respect.

DUKE: I mean, I think that’s true. Look, when someone’s playing really aggressively at you, you sort of become what’s called “a nuts peddler”. The nuts it would be the best hand. But when you’re talking about a nuts peddler, you’re talking about someone who’s playing very, very good hands. Which it sounds like you were, like, you were just getting really good cards. And then David was trying to assert himself over you. You were sitting there with the best hand waiting for him to do that. So you actually are displaying exactly the strategy that I was talking about, which is when someone’s being very aggressive towards you, you actually tighten your play up. You can afford to wait longer because the size of the pot you’re going to win is going to be so much greater for your waiting. The other reason that you’re supposed to wait is that when you’re playing someone who’s trying to assert themselves over you, it’s also much harder for you to bluff them. We can think about this relationship between the strength of your hand and your ability to bluff. The weaker the hand that you play, the more likely that the only way that you will have eventually out of that hand is to bluff somebody — to win with the worst hand. The stronger the hand you have, the less likely you’re ever going to have to bluff. In a game where bluffing is very easy because it’s very easy to push the people around, and make them fold good hands, you can play many more hands. You can play hands that are much weaker because you have really two ways to win the hand. One is you could have the best hand. The other is you can make the other people fold.

LEVITT: And do you think that in general that was an advantage to the well-known players that people somehow attribute to them better hands than they really have because they’re famous?

DUKE: I think in general that is true. You will sometimes come across somebody who just really wants to have a story. You know, like people wanted to have stories about knocking Phil Holmes out of a tournament. And I think that that’s a lot of the advantage that Phil Holmes has because people want the story. They want to knock him out, and then he’ll have the best hand, and they knock themselves out in going for the story. And I think that from the day that I sat down at a poker table before I was ever well known, I came across both of these types of people. The ones that didn’t give me credit, I could bluff a lot against because they didn’t think that I was capable of bluffing. The ones that just were kind of mad that I was there and didn’t want to lose to a woman, well, those people, I couldn’t really bluff because they were going to call me because they wouldn’t want to be outplayed by me. So therefore I had to be much more careful about the types of hands that I was playing against those people. Now I think that as I got more famous, people started giving me credit. So everybody started shifting into that category of, I just want to have a story, I want to knock her out, and so I had to actually tighten my play up because it just became harder and harder for me to bluff people.

LEVITT: Now, you were obviously extremely successful at poker, but I’m curious whether it was hard for you to just be a poker player? I mean, there’s nothing at all wrong with being a poker player, but it’s not necessarily viewed by society as a high-status job. And you were on this high-status track before you stepped off. Were you haunted by that?

DUKE: When I first started playing, I was just so deeply interested in the game. And trying to figure out how do you solve for playing in this environment where you can’t see other people’s cards? There’s a lot of luck involved. It’s very, very hard to close feedback loops. When we think about biases like the sunk-cost bias, for example, you would hear players say, “Well, I couldn’t fold ’cause I had too much money in the pot.” Like, because of the sunk-cost fallacy. And I could feel those forces exerting themselves on me. I could feel like self-serving bias exerting itself on me, which is just like, if you lose, it’s ’cause of bad luck. And if you win it’s ’cause you’re great. I could certainly see that in other players and trying to solve that problem really captivated me for a few years where everything else went out of my head. But, you know, when I think about that regret, that shame that I felt about quitting academics, it’s a lot of because of what you said. You know, which is, a lot of people had put time into me. They had trained me to do something that was supposed to be better for the world than what I was doing in poker.  

LEVITT: Which is interesting because you were providing so much utility and joy to so many people. People loved to watch you play poker on TV. It really was socially, probably, far more rewarding than what you might’ve done as an unhappy academic, the chances of doing anything useful are probably really, really low. So it’s interesting that society has such a pull in that regard.

DUKE: It’s a little bit different now. You know, I think that poker players are cool, in a lot of ways now. But when I started playing poker, it wasn’t on television. It wasn’t on the internet. People would talk to me about it and ask questions about it — it would generally end with something like, “Have you considered gamblers anonymous?” Interestingly though, I don’t think that that had anything to do with my starting to really come back to cognitive science, which happened in 2002. I was very pregnant with my fourth child, and I got asked to come to speak to a group of options traders about how poker might inform their decision making in that environment. And it was this real ah-ha moment for me where I started thinking explicitly about the way that those two disciplines informed each other. And I had this realization that I never had stopped doing cognitive science because the thing that I was obsessed with in graduate school, which was learning under conditions of uncertainty and uncertain systems, was really what I was doing in poker. It was just a different way to explore the topic that was really fun and really interesting and really hard. And when I got asked to give that first talk in 2002, I could feel something light up in me. And I knew that was something that I couldn’t let go of, that conversation between those disciplines. And I still haven’t.

LEVITT: Looking back, knowing what you know now about quitting, should you have quit poker sooner?

DUKE: Oh, for sure. No question about it. There were a lot of things that were going on in poker that were kind of miserable. And I think the reason why it took me so long to walk away is the reason why a lot of people have trouble walking away from things, it was part of my identity. I was a poker player and in the most public way, I was a poker player. Even though I was spending most of my time doing speaking and consulting, my public face, the way that people knew me and the way that I identified myself was as a poker player. And if there’s one thing that I learned from writing this book, the hardest thing to walk away from is who you are. That’s the most difficult thing we ever face.

Quitting is hard. It’s hard for me. It’s hard for Annie Duke. It’s probably hard for you too. And given the data I’ve seen, about half of the people who are listening right now should be quitting something. And by that I mean there’s something: a job, a relationship, a habit, a limiting belief that you’re on your way to ending, that in your heart you know you need to quit; that you will eventually quit, but maybe not until years from now. The time you waste putting off quitting, you’ll never get that back. We’ve all been socialized to think quitting is wrong. The data say the exact opposite. So maybe today will be the day you decide to quit. And if so, I’d love to hear from you, because there’s nothing I admire more than a quitter.

And now a question from a listener and now a question from a listener.

LEVEY: Hi, Steve. So our listener Will wrote to us, he’s an academic in Australia where there’s an increasing interest in using academic research to launch spinoff companies. Will is in favor of this, and he wants to know what you think of this trend?

LEVITT: Well, you know, Morgan, I’m often quite definitive in my answers to questions. But this is one where even I am going to be wishy-washy. ‘Cause I think it’s just a hard question. Now let’s just be clear what behavior we’re talking about. Academics do their research and they come up with some kind of insight or invention, and then they’re able to go and launch companies in which they can benefit, potentially dramatically, financially, from the insights that they’ve gained by doing research in an academic sphere. And that’s the issue under debate: Should an academic who’s doing research while employed for a university should be allowed to have direct and potentially enormous financial gain based on the insights that they come up with.

LEVEY: So just as an example from academics we’ve had on this show, chemist Carolyn Bertozzi has actually launched, I think, 10 spinoff companies based on research she or her students have done. And Jennifer Doudna, the creator of CRISPR, has also launched a company based on that research. Those two both have won Nobel prizes for their research, and I think they’ve been pretty successful with their companies. So what are the downsides of academics doing this?

LEVITT: The first one is that if you’re spending your time trying to commercialize an idea, you’re distracted from the basic tasks that the university pays you to do, teach your classes or come up with other inventions. That would be good in the public domain. The second con is that these incentives could distort people away from the projects that have the greatest social benefit, and towards the projects that have the greatest personal benefit. In other words, those in which the opportunity for profit are the highest. And the third con, which is quite extreme, and hopefully not very common, is that we know that people respond very strongly to strong incentives. And so you could imagine real academic malfeasance — fabricating data, intentionally misinterpreting results — that would be more likely to happen if you have a billion-dollar company on the line then if you’re just trying to get your next paper published in an academic journal.

LEVEY: How often do you think it’s true though a company that was started from academic research actually goes on to be a very profitable thing? I think Carolyn Bertozzi even said that while the hope is to make money off of these companies, her drive was really more to bridge the gap that often occurs between academia and the real world. Where unless she starts a company, her research never really sees the light of day and never lands with the people who could actually benefit from it.

LEVITT: So in some ways, I think if all of these academic spinoffs, all these commercialized ideas never actually make anybody any money, then that’s worse than them making money, right? ‘Cause then it sounds like the academics are spending a lot of time trying to promote their ideas, but their ideas aren’t actually very good. That being said, it often takes a lot of imagination and vision to see how a basic research finding after tweaks and twists and changes and improvements is something that people are willing to pay a lot of money for because it does a lot of good. And I think it’s often the creator, the academic himself or herself, who is likely to have that vision to be able to see what’s really important about what they’re doing. So we’re talking about R & D, there’s basic research, and then there’s more applied development. That gap is big. And it’s often the case that if the researcher doesn’t have the enthusiasm to bridge that gap, that it will never be commercialized.

LEVEY: In terms of you being a little undecided in how you feel about this, what are the benefits? Just so we are spelling out both sides of the argument.

LEVITT: Sure, so the obvious benefit is that it’s motivating. If you can financially benefit, then it gives you stronger incentives to do good research. At a more subtle level, you could imagine that knowing that you can get rich as an academic might lead really good people who otherwise wouldn’t be drawn to academics into academics. A second potential benefit is that it distorts the incentives of academics to work on projects that are more commercializable.

LEVEY: Wait a minute, that sounds awfully like one of the cons you listed just a minute ago.

LEVITT: Ah, it is exactly what I said for one of the cons. Because it’s just not clear whether distorting the incentives of the researchers is a good thing or a bad thing. Many pet academic projects are of interest only to the academic. There’s no way in the world that it could ever lead to any good in society. Whether this kind of distortion is a pro or a con, depends on whether you think academics left or their own devices are going to work on the most important social problems, or whether they’re gonna piddle away their time working on pet projects that nobody cares about. And the third benefit that comes to mind, which is a little more subtle, which is if you are a university employee and you develop ideas, the university owns that intellectual property. So it’s not that these researchers are getting all the benefits themselves, they’re sharing that benefit with the university. So most universities are of the mind that this is good research. Research that leads to a lot of money coming to university coffers, that’s the kind of research they’d like, they support.

LEVEY: So am I right in thinking that it’s pretty rare for an economist to be able to create a spinoff company from their research?

LEVITT: My 25 year career, I’ve never had an idea that anybody would ever want to pay for, so there hasn’t been anything I’ve ever had to try to commercialize.

LEVEY: Wait a minute. A lot of people paid money to read Freakonomics. Now I get that that’s not a company, but does the university benefit from having a professor publish a popular book?

LEVITT: Well, they do not benefit financially. If you go out and write a popular book like Freakonomics that is heavily based on the research I did while I was employed at the University of Chicago, they don’t try to take any of the royalties. And believe it or not, I was actually on the intellectual property committee at the University of Chicago for a year, and I’d never thought about that. And the more I sat in and listened on these disputes, which were happening, in the hard sciences where professors were complaining because they felt like they owned the intellectual property, but the university claimed they owned it. The clearer it became to me that it was quite odd that the university had never come knocking on my door and asked for any of my royalties. In that year I was on the committee, I was very tight lipped about that particular issue and didn’t want to even get people thinking about what I think really would’ve been quite fair. But our royalties have gone way down now so I’m ready to have that topic be discussed.

LEVEY: Will, thank you so much for writing. If you have a question for us, our email is That’s It’s an acronym for a show. We read every email that’s sent, and we look forward to reading yours.

In two weeks, we’ll be back with a brand new episode featuring my good friend, economist John List. Who, I suspect, will come to be viewed as the most important economist of my generation.

LIST: I applied to 150 schools when I went on the original job market in 1996. One of the 150 gave me an interview.

*      *      *

People I (Mostly) Admire is part of the Freakonomics Radio Network, which also includes Freakonomics Radio, No Stupid Questions, and Freakonomics M.D. All our shows are produced by Stitcher and Renbud Radio. Morgan Levey is our producer and Jasmin Klinger is our engineer. Our staff also includes Neal Carruth, Gabriel Roth, Greg Rippin, Rebecca Lee Douglas, Julie Kanfer, Zack Lapinski, Ryan Kelley, Katherine Moncure, Eleanor Osborne, Jeremy Johnston, Daria Klenert, Emma Tyrrell, Lyric Bowditch, Alina Kulman, Elsa Hernandez, and Stephen Dubner. Our theme music was composed by Luis Guerra. To listen ad-free, subscribe to Stitcher Premium. We can be reached at, that’s Thanks for listening.

DUKE: Steve, I’m tone deaf, like truly. If you heard me sing, it would be painful to you.

Read full Transcript


  • Annie Duke, author and former professional poker player.



Episode Video