Author and Ex-Professional Poker Player Annie Duke Talks About Knowing When to Walk Away

Annie Duke was a professional poker player before retiring in 2012. Now, she is a decision-making consultant for a venture capital firm and author of the upcoming book Quit: The Power of Knowing When to Walk Away.

Author Morgan Housel recently interviewed Duke in front of a live audience about:

• Key differences between patience and stubbornness.
• How a quitter created Slack.
• Avoiding the sunk-cost fallacy.
• How pre-mortems can make you a better investor.

To catch full episodes of all The Motley Fool's free podcasts, check out our podcast center. To get started investing, check out our quick-start guide to investing in stocks. A full transcript follows the video.

This video was recorded on September 4, 2022.

Annie Duke: We just quit too late. We want to be much more sure that we need to take that loss because that moment we quit, is when we go from failing to having failed, when we go from a loss on paper to a realized loss. As human beings, we hate that.

Chris Hill: I'm Chris Hill and that's Annie Duke. Once upon a time, she was one of the best professional poker players in the world. She gave it up a decade ago and is now a decision strategists with a venture capital firm and an author. Her latest book is Quit: The Power of Knowing When to Walk Away. At our FoolFest investing conference last week, Morgan Housel talked with Annie Duke in front of a live audience about the differences between being stubborn and being patient, why we often quit too late and how pre-mortems can improve your investing. Morgan kicked things off by asking Annie, how she got the idea for her new book.

Annie Duke: The origin story is basically, I wrote my last book which is How to Decide and released it into a presidential election. Not the best decision. But there was an about, maybe a page to half a page on quitting and why having optionality, the option to be able to walk away from something, is so valuable when you're making decisions under uncertainty. When we make decisions for most of them, we know very little in comparison to all there is to be known and obviously there's just the influence of luck on the outcome. Post decision, you start to find some things out. How is the world turning out? How is my decision turning out? Having the option to be able to react to that, being able to get out of the decision and get back to an option you may have rejected before to a new option is really important for speeding your initial decision to start up. That's what that little page and there was about, and when I was promoting the book and doing podcasts, they were asking me about all other things in the book.

I kept directing back to this one topic which I really had started to think about. That sort of got me thinking like, why I'm I so interested in this topic? Then I just realized that I think that somebody needs to have the dialogue with grit. Because grit obviously is amazing for getting you to stick to hard things that are worthwhile. But it also happens to be really amazing at getting you to stick at hard things that are not worthwhile. As a poker player, I particularly value the option to quit, which is that ability to fold, which is really one of the main skill elements, probably the most important skill element in poker. What happened after I got that bug, is I called my agent. At first, he didn't quite get it. He was like quitting what, like drugs? It's like well, I mean, that would be good to quit but no. Depending on the drug.

But no, as grit isn't always good, you need to understand the context about whether you're supposed to stick to something or not. He then got it. I then just started calling people and I talked to you and this was in the fall of 2020. I started calling people and I called up Danny Kahneman on Zoom and I had a few conversations with him and [inaudible 00:03:15] and Richard Thaler, and it's really lucky to get to Michael Mauboussin, actually was my very first call, who was just on here. I always run all ideas by him first and I would suggest you do too because that's a good strategy in life. Everybody seemed to be as energized by the topic as I was and so that was in November of 2020. By February of 2020, I had already written a 60 page proposal and then got it sold. That's the fastest I've ever gone through that process because I just felt like I couldn't not write it.

Morgan Housel: One of the things that I struggle with in investing is trying to figure out the difference between patience and stubbornness.

Annie Duke: Yeah.

Morgan Housel: Patience is so critical and you cannot have investing success without patience.

Annie Duke: True.

Morgan Housel: But patience can really look like or you can have stubbornness that you pass off as patience.

Annie Duke: That's correct.

Morgan Housel: You're being so stubborn in your worldviews but it feels fine to keep doing it because you're saying, no, I'm long term. How do I figure out the difference between patience and stubbornness?

Annie Duke: Yes. This is really the big problem. I mean, it goes back to what I said about grit. It's great for getting you to stick to things that are worthwhile but it's also great at getting you to stick to things that aren't worthwhile. When we have just very simple heuristics like perseverance is good, which is what I would say that we think about in investing, like just hold. That's all fine, except when it's not. The question is how do you figure out when it's not? The first thing that I can tell you is that almost all the cognitive forces that are working inside of your head are going to get you to quit too late.

In general, we want to be much more certain that quitting is the right choice than we actually should be. That's for lots and lots of reasons that have to do with how our other people going to judge you? How do you judge yourself? The sunk cost fallacy, which is just we take into account, like how do I get my money back in the stock if I sell it? That would be basically the sunk cost fallacy, which of course is a fallacy because it should be, where should I put my next dollar? Not did I lose like $20 in this position? You don't need to get it back in that stock.

But we do that with projects, relationships, jobs, anything like that, being endowed to something like owning the position. But this idea of how do we square our past actions and the resources that we've put into something with what our future actions might be and we want those to match up. And so we rationalize and we say, well, I'm not being reactive and I'm being patient as opposed to say, stubborn. Just to understand that these things are really working against you, that's problem number 1. Problem number 2 is, as Danny Kahneman says, the worst time to make a decision is when you're in it. What does that mean? Like I decided to eat healthy and there's a cupcake in front of me. I'm in it, like, there's a big yummy pizza. That's when it's hardest to do it.

So after you've already gotten onto the losing course, after you're already in this position and you've lost money on it, that's when it's going to be the hardest to make the decision about whether you should exit or whether you should, it's a time for patience. What we really need to do is step back and say, OK if that's the problem, if the forces are working against us and we're going to be worse at this after we're already on a losing path because that's when these things really get activated, what we'd need to realize is that we need to get out of it.

There's only two ways to do it. One is to have done the work in advance. What I do when I worked with, for example, PMs, as I say, I understand that you've written down your thesis. What are the fundamentals? What are the expectations? So on, and so forth. But in companion to that, you have to also, when you get into the position write down if the fundamental moves out of a certain band, what am I going to do and do that in advance. They look at me a little bit confused but isn't that already implied? The answer is, of course it's already implied and you have the intuition that you'll react to those implications when you see them in the world but you won't, so write it down in advance. Say what you're going to do. This is something I call kill criteria. These would be like if it hits a certain benchmark, I'm going to sell. If these fundamentals move out of this band, I'm going to sell or actually on the other side, if they move out of this band, maybe I'm going to press the position. We want to be doing that thinking in advance. So that's one way to get to it. The other way is to have the people making the decision to start something be different than the people making the decision to stop something.

This is essentially getting someone who isn't endowed to the decision as much, who isn't carrying all the sunk costs or the identity or those things that go along with it and help them to make the decision. As an example, if you have an IC that is big enough, one of the things that I've recommended with clients is the people on the IC who recommend who are deciding when you get into a position should be different than the people who recommend getting out of it. Or if a PM has a big enough team, you can divide the labor in that way and have people who enter and people exit, and those people be very different. But I could also just be a quitting coach for you as well. Just to help you see things more clearly because I'm sure you've all had that situation where you see someone from the outside looking in you're like, they should really get out of this relationship or they should quit their job, or they're sticking to this investment that they shouldn't be and you can see it so clearly from the outside, but you're the person who's on the inside when it comes to your own decision. Getting that person on the outside to help you with that is really important. Just so you know, Danny Kahneman has a quitting coach. His name happens to be Richard Thaler, but he feels like he needs one.

Morgan Housel: Both those people have a Nobel Prize.

Annie Duke: Yes, exactly. They're both Nobel Laureates. If Danny Kahneman, who wrote Thinking Fast and Slow, and Noise feels like he needs a quitting coach. I'm all for it. I want one too.

Morgan Housel: Speaking of Kahneman, there's this great story that I heard one time where when Danny Kahneman's writing, Thinking Fast and Slow, he was working very closely with Jason Zweig and The Wall Street Journal. It's like a writing coach. Is not a ghostwriter, but Jason was helping to write. Jason said that he and Danny would work for a month or two on a chapter and one morning, Jason would wake up and Danny had deleted all of it and was just starting over and Jason would be devastated, like we've been working on this for two months. Danny said to Jason one day he said, you have to understand I have no sunk costs.

Annie Duke: Right.

Morgan Housel: That is so rare. He has this quote, I'm going to butcher it, but it's like the only way that you can become a good writer is to be completely devoted to killing everything that's not great. The quote is more elegant than that, but it's an interesting story because it's so rare. The idea of having no sunk costs.

Annie Duke: Yeah. What I would guess is that he has fewer sunk costs than everybody else, be he probably, if he deleted it two months later, he's probably should have deleted it a month earlier. No, for real. For real because this is what the data shows you, is that we just quit too late. We want to be much more sure that we need to take that loss because that moment we quit is when we go from failing to having failed, when we go from a loss on paper to a realized loss. As human beings, we hate that. As Richard Thaler says, we don't like to close mental accounts and the losses. That's that moment you have to close that account and just give it up. It's why we barrel toward the summit of Everest in the middle of a snowstorm, it's wide. Blockbuster won't give up its physical stores for streaming, and then Netflix eats it. Then now Netflix.

This is this issue that we have. What I love is this one really fun thing that Stephen Levitt did that really shows this problem. He just had people like, OK you're thinking about do you want to quit your job, do you want to quit your relationship, do you want to move to a new town. These big life decisions, you are just, should I stay or should I go? He puts up a website. He says, we're going to flip a virtual coin for you. Heads you quit, tails you stay. Now you'd say nobody would ever do this, but 20,000 people did. Let's agree that if you're going to go flip a virtual coin for whether you should stay in your job or not, that you think it's 50/50. You are so sure that this is either or 50/50 because you're going to go do this. They do that. What he finds is he measures their happiness two and six months later is something surprising.

That if it were really a 50/50 decision, the quitter's post decision should be equally happy as the stickers. But that's not what he finds. What he finds is the quitters are happier. Why? Because by the time you think it's 50/50, it's like 90/10. It's not even close because the decision to quit is also made under uncertainty. We don't know what the future holds. We don't know whether could we turn it around? Could we get our money back? Could we not have to shut this start-up down and become the next [Alphabet's] Google? The only way for us to know for sure is to keep going on the path. What happens is by the time we get to OK, I'm ready to quit. It's not really a decision anymore because you're already at the bottom of the crevasse. Like obviously I shouldn't have continued. You're already at the point where you can't make the chapter work after two months and so now you're throwing it out. That's something I think we need to remember, is that we get to quitting just way too late in general.

Morgan Housel: Whether it's a person or a company who has quit really well so much that their quitting was key to their eventual success?

Annie Duke: Stewart Butterfield.

Morgan Housel: Of Slack. What became Slack?

Annie Duke: So, Stewart Butterfield I think is the best quitter in the world. He started a company way back during the dot-com crash era that was trying to create an online multiplayer world-building game. He was creating that, and it was just a bad funding environment at the time. This was late '90s, early 2000s, he couldn't get any money for it. He shut that company down because he had to. But there was a piece in it which had to do with photo-sharing, which basically allowed you to see pictures in my inventory and that became Flickr, which was a modest success, sold it to Yahoo. Then he quit Yahoo because he still had this dream of creating an online multiplayer cooperative world-building game. He rounded a bunch of people up. This was in the late oughts. His founders, you know, the co-founders, and they started this new company called Glitch. They created a game called Game Neverending, which was a huge critical success. It was Monty Python Meets Dr. Seuss is the way it was described. The gaming community, like the core gaming community, seemed to really love it. At least the critics did.

Through word of mouth and PR they started to build an audience and they, a ways in, they had 5,000 really hardcore users who were using the game 20 hours or more a week. They had the backing of Andreessen Horowitz, and [inaudible 00:14:23]. So they had a lot of money. They had six million in the bank. But the problem was that in order to get those 5,000 users, they needed to get basically for every 95-99 people who came to the game, you'd only get one hardcore user and the rest of the people would be there for only about seven minutes. This was a problem. Along with the investors, they did a huge marketing push over six weeks where they started going from PR and word of mouth to traditional marketing buying ads. They did that, and they actually experienced user growth of 67 percent week-over-week for six weeks up until the weekend of November 11th, and 12th, 2012. That was their biggest week ever.

That Monday morning, Stewart Butterfield wrote a note to his co-founders and investors saying I woke up with the dead certainty this morning that Glitch was over. That's weird. His investors a little bit were like, what? They kind of lost their minds. He basically explained that he had figured out that they would have to maintain that exact same growth for 31 weeks just to get to breakeven. That it was absurd to assume that you could, because they probably already reached the gaming community that was going to use the game and that ads were going to get more expensive, they were going to go into the same eyeballs that had already seen them, and they were going to be getting outside of the same core gaming community. What he realized was that this wasn't a venture scale business. He shut it down, not clear whether his co-founders or investors were on board with it.

But as he said, if he wasn't into it, that was it because he was really the driver of the operation. He even said the thing that I regret the most is that I feel like I was six weeks too late because I thought people would think I was insane. Even he said he quit too late, and he's like the best quitter I've ever heard of in my life. This is a success story in itself. He realizes that this is not a venture scale business. He realizes that for his employees, the equity that they're working for is no longer worth their time. So he wants to free them. Because this is something like, oh, I'm being patient. When I'm really being stubborn, I owe it to my employees. He realized he owed it to his employees to let them go because they were mostly working for equity, very little cash comp and that wasn't worth their while. He didn't think that this was the best place for his investors' money. He sort of frees everybody up to go and be brilliant somewhere else.

But two days later, he says, you know, this is internal communication tool that we develop doesn't even have a name, that people seem to really like, maybe I could develop that. He gave it a name which was searchable log of all company knowledge, which was Slack. Two days later he founded Slack. We know what happened with that. I think there's a temptation there to say, well, what makes that a really happy story is that Slack came out of it. But I think the thing that we need to remember is if he had quit that company, just shut it down and never founded Slack, that that in itself is a happy story. Because he quit maybe six weeks too late, but somewhere around the point that he really realized that this was not venture scale, and it wasn't worth everybody's time to be doing it.

Morgan Housel: That's a great story. One other point that this all reminds me of is just the concept of having enough of when to quit and when to pull back. Someone once asked John Rockefeller how much money he needed to be happy, and his answer was just a little bit more.

Annie Duke: Was it all of it?

Morgan Housel: All of it. Every dollar. But I think it's so critical to happiness that you do have some sense of enough. Enough doesn't mean that you have no ambition for more, it's just an idea of keeping your expectations and check relative to results. Maybe it's not a full-blown quit. But how do we, individual people use what you've looked into and researched and your knowledge about this to have some sense of enough money, enough success, etc.?

Annie Duke: Again, I think that the key here is that you have to do the work in advance. You have to say to yourself, at this moment in time, as I envision myself six months from now or a year from now, what is it that I'm trying to achieve? Then given what I'm trying to achieve, what are the signals that should tell me that I've done it? What are the signals that should tell me that these goals are unrealistic or that this isn't actually the right way to achieve those goals or I'm not on the right path to be able to do it? Do that in advance for yourself. I do this all the time. The thing is, what's really important to understand is that you don't need to do this. If you didn't do it when you first started something, you can do it at anytime in-between. This is something I coach employers on all the time when they're saying, oh, there's this employee and I think I should let them go, which I happen to know is the moment they should let them go. But usually they're not ready for that.

What I say to them is, how long do you think is reasonable for this employee to turn it around, and we figure out what that is. It's usually somewhere around a month to two months with feedback obviously. I say, OK, let's actually see what are the key performance indicators, what are the benchmarks they might hit? You tell me what the criteria are that would tell you that they've actually turned it around. Sit down and actually work that with the employee so they're in on it, and then you can have the conversation. What's interesting in that situation is the employee usually quits. Well, because I think they just realized that I can't actually achieve this. At some point it does get them to realize it's not the right fit for them, which is better for them because they get to go find a better fit for them. But you can do that with yourself as well.

That's what I say, don't try to figure out it in the moment. When you start to have those questions sit down. I think people should do this twice a year and say, what I'm I trying to achieve? What is the next six months look for me? What are going to be the signs that I should exit? What are going to be the signs that I should considering an exit? What are we going to be the signs that I need to go gather more information about what is right for me? Because in the end, what we don't want to do is end up continuing when we ought not to. This is one of the things that I think to myself now all the time, I don't want to be Siobhan O'Keefe who started experiencing pain in 2019 in the London marathon on Mile 3 and on Mile 8, she snapped her fibula, broke her leg. I know, right? She finished it. You think that's bizarre. There were four other people in that race who broke something and kept going.

If you look up just broken leg or something in a marathon-finished race, it's pages of Google results of people doing that. Obviously that's putting the rest of their ability to run future marathons at risk. I think that we all share the intuition that if I broke my leg during a marathon I wouldn't keep going. But what I realized from Siobhan O'Keefe is, yes I would, because it's one more mile. It's your question about just a little bit more. I'll just go a little bit more and not being able to see past being in it. Because John Rockefeller was asked that when he was in it. That's the issue. Because if I'm in it, I want a little more, you have to step back from it and say, what could happen in this marathon that would make me quit and write down breaking my leg and then tell your friends who were rooting you on, if I break my leg, tell me to stop.

Chris Hill: Annie Duke's new book Quit: The Power of Knowing When to Walk Away comes out October 4th and is now available for pre-order. We are off on Monday for the Labor Day holiday, but we'll be back on Tuesday. As always, people on the program may have interest in the stocks they talk about and The Motley Fool may have formal recommendations for or against, so don't buy or sell stocks based solely on what you hear. I'm Chris Hill. Thanks for listening. We'll see you on Tuesday.