How To Make Better Decisions | Annie Duke | Modern Wisdom Podcast 233

1. 0:00 – 0:32

   Intro

   1. ADAnnie Duke0:00

      In order to be a great decision maker, you have to do two things, and you have to do them pretty intentionally. You have to understand and see the luck. You can't control the luck, 'cause luck by definition is outside of your control, but you wanna see it as accurately as possible. Here are, here's the spread of outcomes that can occur and here's how likely those things are to happen. But then the other thing you need to do is start really examining what are the beliefs that you have that inform the decisions that you make. (air whooshing)

2. 0:32 – 3:35

   Annies Background

   1. ADAnnie Duke0:32

   2. CWChris Williamson0:32

      Many of our listeners will be familiar with you, but for those who aren't, can you give us a bit of background what makes you an authority on decision-making?

   3. ADAnnie Duke0:40

      Oh my gosh, (laughs) that sounds so very serious. Um, well, I guess it's, I, I guess it would be kind of the, the very strange path that I took, uh, as an adult. Um, so I started off after college, uh, in a PhD program at the University of Pennsylvania. I was studying cognitive psychology, um, cognitive science. Um, in particular, uh, I was really thinking about how people learn. Um, so I was, I was gonna go off and become a professor, as one would after getting that degree. Um, and as I was, uh, ABD, as we call it, b- before I, um, defended my thesis, I actually got sick and ended up, uh, needing to take about a year off before I went out to become a professor. And in the year that I took off, I started playing poker. And I did that kind of in the meantime in order to, uh, support myself until I could go and start my career. And, uh, so the meantime turned into like 18 years-

   4. CWChris Williamson1:50

      (laughs)

   5. ADAnnie Duke1:50

      ... as it does. Um, I kind of kept playing poker. Um, and you know, a- and then about eight years into my poker career, I actually started thinking about this very interesting conversation that cognitive science, my background in cognitive science w- could have with the real-world decision-making problem that poker presents, which is obviously very high stakes, lots of skin in the game. Uh, if you're bad at deciding, you lose all your money, um, over the long run. Um, and so you really have to be kind of thinking about, in particular, how, how do people learn in these very n- noisy environments? Meaning, um, uh, where there's lots and lots of luck, uh, in terms of influencing the outcome, and also lots of hidden information, imperfect information, which is, by the way, almost any decision you've ever made in your life, including just like whether to proceed through an intersection, um, (laughs) i- is actually very poker-like in its decision. So, um, so I did, so I played poker till 2012. In 2002, I started thinking about how these two things might converge, started speaking on that topic quite a bit. And then in 2012, I retired from poker, started really focusing this, on this full-time, spent a lot of my time consulting with companies on decision strategy, um, giving talks, writing books on the topic. Um, and then I've also gone back to academics and I'm back to doing some academic research in cognitive science. So I, I took a full loop. I, I en- I ended up getting back to where I started, but it just took me a while. It took me a while to circle back. So I mean, I guess, I guess that's kind of, you know, I guess I have some practical, some practical bonafides and some academic bonafides maybe? I don't know, trying

3. 3:35 – 5:57

   What is a decision

   1. ADAnnie Duke3:35

      to answer your question.

   2. CWChris Williamson3:36

      An 18-year hiatus or a sabbatical to just go and be a professional gambler for a bit. Is that how it went, ended up?

   3. ADAnnie Duke3:41

      Yes. Yeah, (laughs) exactly.

   4. CWChris Williamson3:43

      (laughs) Um, okay. So I think that, that's as tr- credentials go, that's not bad. You've got the, the academic underpinnings and you've got a, a f- World Series of Poker bracelet and a bunch of-

   5. ADAnnie Duke3:55

      I do, yeah.

   6. CWChris Williamson3:56

      ... other accolades. So, um-

   7. ADAnnie Duke3:59

      Yeah.

   8. CWChris Williamson3:59

      ... talking about decisions, how do we define a decision? What is a decision?

   9. ADAnnie Duke4:04

      Oh, that's really interesting question. Um, uh, yeah. So, so a decision is just a choice about what you're, uh, what path you're gonna take. So at any moment in time, uh, there are actually infinite paths, although we're not considering all of them. Um, but, uh, uh, at any given time, you can go down run- one road or another. Um, and I'm talking obviously a figurative road. Sometimes it's an actual road, um, but it's a figurative road if we think about, uh, at any moment in time, you can think about the future as branching, right? So there's all sorts of futures that could unfold. Um, and when you make a decision, you're choosing which of those possible futures will be available to you basically. So, um, you know, you can think about it like, uh, um, if I'm thinking about where I wanna end up in 10 minutes, um, if I go east, there's a certain number of possibilities for where I could end up. And if I go west, there's, uh, other possibilities for where I could end up. Um, the other thing I wanna just say is like a decision is not just an action that you co- commit, comit. So let me go back to the route, um, uh, analogy. If you're on a road, I think that we sort of think that a decision would be to exit the road, to go off the road. But, um, again, at any moment we could think about y- y- there's, there's the exit choice, but there's also the choice to stay on the road. So, uh, staying on the path that you're already in is also a decision. And, and I just think that that's really important because I think that people think about decisions as actively changing something, but when you don't change anything, you're deciding also not to do that. I hope that that was a clear enough explanation. I don't know.

4. 5:57 – 9:35

   Commission vs omission

   1. ADAnnie Duke5:57

   2. CWChris Williamson5:57

      I like it. Yeah, the difference between commission and omission in-

   3. ADAnnie Duke6:00

      Yeah.

   4. CWChris Williamson6:00

      ... terms of whether you decide to enact change. I suppose as well-... a decision to do one thing is as much choosing to take that route as it is not choosing to take every other route that's available.

   5. ADAnnie Duke6:12

      Yeah, actually, uh, so that, uh, that's a really important point. Um, so it, uh, that's the point of if I go left, there are certain futures that are now available to me, but the other futures, the ones where I went right, are, are no longer, would not be available to me, at least not in the near term. So that's why when we're thinking about a decision, we're choosing the futures that are available to us, and we're sort of eliminating other futures that might be available to us. And interestingly enough, like, once, once you're sort of thinking about decisions that way, it un- unlocks a really important decision principle, which has to do with this idea of optionality, because that's what you're talking about, right? Wh- if, if I choose options, um, I may be, uh, making it so that I, I am rejecting other options that are available to me, and there are costs associated with that. And so once you sort of figure that thing out, then what you realize is that you can, uh, put, uh, sort of prioritize what the value of optionality is. And you can get optionality because of one of two things. One is that when you choose the path that you're on, that it's quite easy to get off that path and go back and pick up other futures that you might have rejected. So, uh, if I'm on a road and I exit, it's easy for me to get back onto other roads, right? So that, that would be one thing. So, um, the difference would be like, uh, renting versus buying a house, right? So, uh, if you rent a house, you have more optionality 'cause it's easier to quit. It's easier to get out of the house. It's easier to move. There, there's lower costs associated with the decision. Um, but if I buy a house, I'm making kind of a longer term commitment where I've reduced my optionality. And if you were thinking about an investment, you have liquid investments and illiquid investments, right? So, uh, there's certain, like I could buy a stock and then I could literally sell it the next day. But if I invest in a company as an angel investor, it's gonna be very hard for me to divest of that position, right? So that would be sort of thinking about that piece of it. Um, and we wanna, uh, you know, it's, it's easier and we can make decisions faster when we have, when we preserve that optionality. The other way to preserve optionality is if we can do more than one thing at once. So sometimes we don't have to choose to go left or right or east or west. Sometimes we could do both, we could go east and west at the exact same time. Um, and that also helps us as well, 'cause we don't have to reject as many options. So like the simplest example of that is that, uh, most people who own stocks don't own just one. So I don't have to choose like, "Oh, there's so many stocks to choose among, um, I'm gonna choose Google and I can only put my eggs into that basket." Um, instead I can, I can have lots and lots of stocks at the same time. Um, we can also think about like dating versus marrying again, right? So dating is like I can date lots and lots of people at once. Um, and for most people, not all, marrying means one person, right? So, um, I have less optionality if I marry than if I date 'cause I can date in parallel. So, uh, so I think it's really important a- as you start to sort of construct what a good decision process is to keep in mind what exactly what you said, that you're not just choosing the thing that you're choosing, you're also rejecting the other choices that are available to you. So if you can construct a decision where you don't have to reject as many of those things, um, then you're gonna, you'd generally be better off.

5. 9:35 – 12:16

   Gills razor

   1. CWChris Williamson9:35

      And not just the immediate rejection of whatever the options are, but the second and third and fourth order effects of going down that particular path.

   2. ADAnnie Duke9:43

      Right.

   3. CWChris Williamson9:43

      There's a friend of the show, uh, George McGill, who's been on a number of times talking about mental models and he created his own razor. Uh, and McGill's razor is when faced with two equal choices, choose the one which permits the most luck. And I think that you could probably trade the word luck there for optionality, so that as you go-

   4. ADAnnie Duke10:00

      Yeah.

   5. CWChris Williamson10:00

      ... down that particular path, there are fewer options closed to you than would've been had you have taken the other one.

   6. ADAnnie Duke10:06

      Um, I think that's a really good razor. Um, uh-

   7. CWChris Williamson10:09

      Annie Duke said it first, George. There you go, mate. Shout-out (laughs) .

   8. ADAnnie Duke10:13

      Well, it's because it's really, it's really true. And the reason, the reason that you wanna do that is that when you're deciding under uncertainty, which is pretty much all the time, um, even, even if you have perfect information, so I'm assuming perfect information, which by the way, you never have, um, uh, there's still a variety of ways that the world can unfold and tho- those worlds are gonna unfold at a particular probability, and some of those things are gonna be, uh, comprise the downside. They're gonna be things that you're not really happy observing. And the thing is that, um, you can make amazing choices that, you know, are gonna work out for you, say, 70% of the time. That would be a pretty incredible choice. We'd wanna make that choice over and over again. But what that means is 30% of the time there's going to be, uh, something that you don't like, something's gonna happen that you don't like. So, uh, given that you're always exposed to that risk, one of the ways to de-risk is to preserve optionality, to choose when you have two paths that you're choosing, say, say they're both, uh, 30% to not work out, that I would prioritize the one that's 30% to not work out but I can do, I can do something about it, I can get off the path, I can exit. It's preserving my optionality in some way. And that's sort of separate and apart from the fact that I can make, I can know sort of everything I need to know right now, and I can have an understanding of given this moment sort of what those expected worlds are, but new information's gonna reveal itself to me later that might matter to me, right, where I may wanna do something different. So that's another reason that we really want to preserve optionality, because we're not omniscient. So even if I kind of know everything about the state of the world today, I don't have a time machine, so I ca- um, I can't see what the world will be tomorrow, and sometimes the things that I find out about how the world is tomorrow are really gonna matter to me in terms of what I'd like to choose. So I think that razor is amazing.... um, and I'm going to steal it-

6. 12:16 – 15:16

   Morgan Housel

   1. CWChris Williamson12:16

      Cool. George-

   2. ADAnnie Duke12:16

      ... with credit.

   3. CWChris Williamson12:17

      W- w- we, uh, we are, uh, pushing your razor out into the world. Also, a mutual friend of ours, Morgan Housel, was on, uh, last week and, um-

   4. ADAnnie Duke12:25

      Love him. Is, is his book not, like, ridiculous?

   5. CWChris Williamson12:28

      Look, I told everyone that was listening they had to go and get it. I, I referred to it, what did I say? It was, um, Robert Kiyosaki with a better, a better-looking author. Um, it was like-

   6. ADAnnie Duke12:37

      (laughs)

   7. CWChris Williamson12:37

      ... the handsome, (laughs) the handsome Rich Dad, Poor Dad.

   8. ADAnnie Duke12:40

      I'll tell you something. Uh, what I love is, first of all, the book is so incredible, but I've seen, um, it's particularly good to get Morgan on video.

   9. CWChris Williamson12:48

      He's hilarious, man.

   10. ADAnnie Duke12:49

       I mean, you get, you get-

   11. CWChris Williamson12:50

       (laughs)

   12. ADAnnie Duke12:50

       On audio, it's, it, you also get this across, but on video, in particular, you really get this, that when he starts, like, you can just see, like, he's so passionate and he's so excited about what he's talking about. And if you get to see him talk, you know, when you get to see him, it's like you just see, it's like a kid in a candy store. Like, he's so excited to be exploring these ideas and be expressing them, um, and it's so much passion. And so I just think it's, like, really fun to watch him talk about his ideas, um, separate and apart from the fact that his ideas are amazing. Like, he's incredible at communicating these things, um, and really, like, getting to the heart of what matters without a lot of extra words around it.

   13. CWChris Williamson13:29

       Mm-hmm.

   14. ADAnnie Duke13:30

       Um, I, I would really love to aspire to express the concepts that he does so eloquently in, in as few words as he does. I tend to be wordier than he is. But anyway, everybody should get his book, The Psychology of the Money, of Money. Go get it.

   15. CWChris Williamson13:44

       I, I, uh, I said that. That was one of the first comments I made about the book. Uh, look, it's linked in the show notes below. You should have already gone back and listened to Morgan's episode, which was on last week. Go and check it out. It'll be available there. Um, I said I really enjoy the fact that 'cause he's coming from a blogging background where attention's very scarce. He's, uh, uh, had to hold people's attention so, so sharply, and that's ported across really well into a book that's only, like, 195 pages, I think, of, like, actual writing.

   16. ADAnnie Duke14:10

       Yeah.

   17. CWChris Williamson14:10

       Short, short, very short, punchy sentences. One of the chapters is, like, three-quarters of a page long and, uh, apparently he sent it off to his publisher and his publisher was like, "Do you not want to, should we, should we make this a little bit longer?" And he was like, "No, no, no, no, it's fine." Uh, he also said that he would, uh, mud wrestle me on the internet for pay-per-view money, um, if that's, uh, and I'm still proposing that to people. You know, everyone has a price and it's been a tough year for investments this year, so me and Morgan are, if the particular threshold is met for interest on, on the internet, we might mud, we might end up mud wrestling. So that's how it'll go.

   18. ADAnnie Duke14:44

       Oh my gosh.

   19. CWChris Williamson14:45

       (laughs)

   20. ADAnnie Duke14:45

       Okay. (laughs) I don't know. I don't know if I want to see that or not.

   21. CWChris Williamson14:48

       You do.

   22. ADAnnie Duke14:48

       I hope-

   23. CWChris Williamson14:48

       I mean, look, everyone wants to see that and it will-

   24. ADAnnie Duke14:50

       Okay.

   25. CWChris Williamson14:50

       ... be available for $9.97, uh, for a one-time-

   26. ADAnnie Duke14:53

       There you go.

   27. CWChris Williamson14:54

       ... purchase. Um-

   28. ADAnnie Duke14:55

       There we go. I'm so happy you had him on though. He, everybody needs to hear what he has to say.

   29. CWChris Williamson14:58

       The third time this year he's been on the show. Oh, yeah. We're, it's, it's getting pretty serious between me and him now.

   30. ADAnnie Duke15:04

       Yeah, I guess so.
7. 15:16 – 21:31

   Why cant I just use my gut

   1. CWChris Williamson15:16

      dream. That's what I really want to do.

   2. ADAnnie Duke15:17

      Yeah.

   3. CWChris Williamson15:17

      Um, so we've talked about sort of just the matrix of decisions. Why should I be concerned about having a process for decision-making at all? Why can't I just use my gut? Most people probably do.

   4. ADAnnie Duke15:30

      (laughs) Yeah, that's a real... Well, I mean, the short answer is I assume you'd like to do well in life (laughs) with some reliability. I mean, look, here, here's the thing, like, it's not so much that you can't have a good outcome when you use your gut. Of course you can. I mean, in the same sense that, um, I can flip a coin and call heads and sometimes it's gonna land heads. I'm gonna win a bunch of money, right? Uh, and that's like, I could be in a situation where I'm offering you two to one, right? So I'm losing to the bat. Um, and I happen to win, I get your dollar. Okay, great. You know? Um, so the, the h- the issue has to do with, um, how are we constructing a process that's actually going to accomplish the things we want it to accomplish? Which is, number one, sort of get us to whatever the best possible judgment is, given the fact that it's subjective. So, um, we can think about there's, uh, what's objectively true of the world, what would objectively be the best judgment? And then how close are we getting to that, right? And given the limits of our knowledge, um, I would call the, you know, the, if you can get as close as possible to that thing, you're gonna do better. And generally, you're just not gonna do better with your gut. Why? Because your gut isn't actually any kind of decision tool. It's not repeatable. You can't look back at it and say, "Well, why did my gut make that choice?" Um, your gut is where all the cognitive bias lives. Your gut is where all the noise lives. Like, all the things that frustrate your decisions are living in your gut, basically, aside from all the bacteria and undigested food. So, um, so in order to be a great decision maker, you have to do two things and you have to do them pretty intentionally. You have to understand and see the luck that's go- going to be in the way that your decision turns out. You can't control the luck, 'cause luck by definition is outside of your control, but you wanna see it as accurately as possible. Here are, here's the spread of outcomes that can occur, um, and here's how likely those things are to happen. Okay? But then the other thing you need to do is start really examining what are the beliefs that you have that inform, um, the decisions that you make? And you need to actually pick those apart into, uh, a bunch of things. Number one is you have to think about what the component parts of any decision that you make are, which is kind of the antithesis of gut, right? Like, gut is like, "I just think I'm supposed to do this," and there's no kind of why that follows up with it besides, well, it's just my gut.... right? Um, so when you think about what are, what are the component parts, what are, what are the actual motivations? If I think about a statement that I make, what must be true of the world and how can I think about whether those things are actually true of the world? That's kind of the antithesis, thesis of gut-

   5. CWChris Williamson18:34

      Mm.

   6. ADAnnie Duke18:34

      ... because you're actually opening that up to a, a process that you're examining. So as an example, if, if I'm investing, um, and I choose to buy a stock, um, by definition, uh, this was something I discussed with Michael Mou- uh, with Michael Mauboussin actually just a few days ago, but when, when I make a, an investment in a stock, I'm saying that I think that I know something about this company that, or this stock that the market doesn't know. So it's not acceptable to just say, like, "Well, I, it looks like a good stock, and I have a nose for value," right? Okay, I don't really know what that means. You have to say, you have to pick it apart and say, "What is the thing that I think that I know that the market doesn't know?" So it could be something super simple, right? "I think the market is estimating the number of widgets that they're gonna sell." Okay, great. So now, now you're pulling that apart. Notice how this is very opposite to gut, right? Because I'm starting, now I actually have to forecast that thing. So you also have to express yourself precisely, um, as you're forecasting those things that you're pulling apart that must be true for you to actually believe the thing that you do, that you should buy this stock, for example. Um, so, so that's kind of like the first thing about what you're doing in terms of your beliefs. So you're pulling these things apart into their component parts. You're thinking about what are the possibilities. You're, you're trying to see the luck accurately, and then because you're doing that in all this very explicit way, you now allow yourself a way to go check back and to say, essentially, what I've done is in this very explicit way, I've created, um, what my forecast of the future is. Not in a vague sense of I think the stock is gonna go up, but I've forecasted what I think the states of the world are gonna be in the future, how likely those states of the world are going to exist, such that I can now actually close that loop and go back and check to see how well calibrated I am. And I can now do that because this is all very explicit. That's the problem with your gut, is that it's by definition implicit. It's by definition sort of intuitive. So I don't have that way, first of all, to, uh, think about the decision process in, in any kind of rigorous way as I'm making it. But worse yet, I have no way to go back and close the feedback loop in t- in order to create a really good learning loop, because I can't really go back and check. I can't say, "Well, here are the, here are my assumptions. Here are the things that I thought would be true of the world. Here was my forecast of the states of the world that were going to exist. Here's the data that I was using in order to inform that decision." And so I, I think that when you're deciding with your gut, you're, it's like the, the thing I say is it's kind of like playing pin, pin the tail on the donkey, right? It's like that old children's game. It's like you just put a, put a blindfold on, and you're sort of like, and then sometimes you actually get it right there, and you're like, "Look how smart I am." And it's like, ew, but that's not really a process.

8. 21:31 – 24:26

   The importance of repeatability

   1. CWChris Williamson21:32

      Repeatability is one of the important things that we need for a decision tool. What else do we need to make a decision tool effective?

   2. ADAnnie Duke21:40

      Um, so repeatability, meaning if I use this process in the same way over time, I should expect to get similar results. Um, I also need it to be something that can be examined, right? So I can examine it like an object and understand whether I'm actually using it in the appropriate way. Um, uh, I need to be able to, uh, give, uh, the process to somebody else such that they could actually apply that same process. Now we can start to see the problem with the gut, right? Is I can't be like, "Ew, here's my gut." (laughs)

   3. CWChris Williamson22:15

      (laughs)

   4. ADAnnie Duke22:16

      Right? Well, I guess in Star Wars, they, he, they open that animal up-

   5. CWChris Williamson22:20

      All the time. All the time.

   6. ADAnnie Duke22:20

      ... sleep inside of it. They're always sleeping inside of guts in Star Wars. Um, (laughs) I don't know if that helps them with- make decisions, but it keeps them alive. Um, ew. But anyway, so I can't, I can't just say to you, like, the, the problem is if you say to me, like, "How'd you make that decision?" And I say, "Oh, well, I had gut feel. Like, I just felt like that was the right thing." This is completely unhelpful to you, um, because that doesn't mean that you can do it, right? So I need to also be able to give it to somebody else and explain how you would use this process such that they could also use it in the same way. That is not to say that they will get the same results from it, because, um, there are issues of talent, right? I mean, two people can learn how to hit a tennis ball, and one person will be much better because people just sort of are better at some things than others. But I should be able to get you part of the way there. I can get you to hit a ball, right? Um, and hopefully I can get you to hit a pretty good ball. If it's a really good process, I should get more fidelity out of what you're doing. So r- repeatability, I have to be able to examine it, um, uh, so that I can understand what its use is, what its purpose is. I have to be able to teach it to somebody else. That's really important. Um, and then I have to be able to go back. I have to loop back to see, I have to be, it has, there has to be accountability. I have to be able to loop back to see whether it produced, it, I actually applied it well. Um, did I use the right tool for the right job? That kind of thing. So, uh, it's, it's very simple. It's like a really good decision process, which has a whole bunch of decision tools in it. It's really like, uh, just using tools, right? So I, I wanna know, uh, we can think about like a screwdriver, right? Okay, I can look at it. I can figure out what its purpose is, right? I can, for the purpose that I use it for, to, to screw screws in the wall, I can repeat that with fidelity over and over again. Um, I could hand you the screwdriver, and I could teach you how to use it such that you could put screws in the wall as well. And then I can go back and I can check your work, and I can see, oh wait...... it seems to me that you stripped the, the threads on this. Did you use a hammer?

9. 24:26 – 30:16

   The ultimate decision tool

   1. ADAnnie Duke24:27

   2. CWChris Williamson24:27

      (laughs) Well, I mean, a hammer is the ultimate tool. As everyone knows, there's no job that can't be fixed with a hammer. Which would be the-

   3. ADAnnie Duke24:33

      That is totally true.

   4. CWChris Williamson24:34

      ... ultimate decision tool, yeah.

   5. ADAnnie Duke24:35

      Yes. Well, yeah, absolutely. That's what I was gonna say, like, breaking asphalt, but you could do it, it would just be really tiring 'cause you could use the claw part of... That, it would just take you a very long time, but you could actually do it. But yes, you could, you could-

   6. CWChris Williamson24:46

      Hammer, underrated tool as far as I'm concerned.

   7. ADAnnie Duke24:48

      You could use a hammer, right? So, so you can think about gut as, like, what if you came in and you saw a screw was in the wall, and you said to me, "How'd you get it in the wall?" And I was like, "It just happened because I'm really good at putting screws in the wall." Like, I mean, think about how bad that is. First of all, you can't do it. Second of all, I can't actually figure out, like, was that the best way to do it? Was there a better way to do it? Because I just left it in a black box, right? So, and one of the things I wanna be really clear about is that, uh, there are lots of decisions you're gonna make in your life where you are actually using your gut. Just because we... You know, you're not gonna take a really long time on every single decision that you make. The issue is, first of all, you have to know what a good decision process looks like so you can understand when it's okay to kinda go with your gut, um, because there is just less risk to sort of not having a high-fidelity process. Um, and then the second thing that's, that's just really kind of important about that is that you have to have a way, even with your gut decision, to go back and say, "Well, there were things that were implicit in my gut decision, that for me to make that decision, I had to believe were true." So let me go back and actually think about that now, right? Let me think about what were the things that ha- that I had to believe were true. Were those things actually true? Was I missing anything? What wasn't I incorporating into my gut feel about it? Um, so that the next time I make a gut decision, having gone through that examination, one would assume I would have honed it better. So, uh, look, uh, there's lots of things that you're just gonna kinda go fast on. And that's okay, as long as you understand what a really good process looks like and you understand how incredibly important it is to make the implicit explicit in your decision process so that when you go and take a look back on really what is the quality of the decision that my gut produced, that you have a way to actually answer that question, such that you could tweak and hone and reali- you know, and then sometimes you'll realize, um, for, say, the market that I was in, uh, my gut feel was pretty good, but there's something that's changed about the market that I didn't notice, and now I actually have to go back to being pretty deliberate. You know, I, I actually think about it, like, like, this so that you can see these videos. So, um, when you're riding a bike, we know that when you turn the handles to the right, the bike goes to the right, and when you turn the handles to the left, the bike goes to the left. So, when we're starting to learn how to ride a bike, it's a very deliberative process, right? We're trying to figure out, like, how do we balance the bike, how do we keep it upright if we turn this way, are we gonna turn... And you've seen people, they're sort of s- you know, wobbling and whatever. And then eventually, we become really good at riding bikes, and it's pretty much something that we're not thinking about very much, right? Like, I, I, I just sort of get on a bike and I go, and, um, not really deliberating about it. So that would be like, uh, it's turned into a gut decision, and it works really well. But, um, the world changes. So, what happens when someone hands you a bike where now, when you turn the handles, it goes to the right, it goes left. And what you can see is that people fall down a lot again, because their gut is telling them when they wanna go right to actually turn the bars right, and now they have to sort of re-figure out, "Well, hold on, this is, like, a different type of bike, and so I have to now learn this process," and it takes a while. But then once they learn it, they can go back to sort of go, you know, using their gut. Th- decisions and gut decision-making are just, like, a lot like that. That your gut can get you pretty far, but you always have to be sort of checking back in to see if the world is handing you a bike where, where the handles- bars work differently, right? Um, because that happens all the time. Environments change a- and, and markets change. So that's, like, things external to you, um, changing. But then the other thing is that, uh, sometimes it turns out that there's a better way to ride the bike, and the way that you were doing it worked pretty good, right? Like, you were winning to it. Uh, you were getting to where you wanted to go. It was all kind of fine. But someone shows you a, something that you could do that's even better than that, would actually help you to be more efficient or, or whatever it is, and you a- so that's why you actually... Another reason why you wanna be examining your gut is even if nothing has changed about the world, you may not have the best strategy. Y- the big danger is you have a strategy that's winning. That's, like, the biggest danger that you can ever be in.

   8. CWChris Williamson29:29

      Why?

   9. ADAnnie Duke29:30

      B- because it means you're very unlikely to go explore if there's a way to win more. And, you know, you... By the way, uh, you can see this in investing. There are people who are winning in investing, and you, they haven't answered the question is, okay, you're winning, that's fine, would you be winning more if you were indexing? (laughs)

   10. CWChris Williamson29:51

       Yes. Morgan, Morgan says yes.

   11. ADAnnie Duke29:53

       Yes. For most people-

   12. CWChris Williamson29:55

       Dollar, dol- dollar price average into the market, index funds, leave it for fif- 50 years, millionaire. That's what, that's what Morgan says.

   13. ADAnnie Duke30:00

       Right. So, and it, but that's not true of everyone, but what happens is that we get focused on the fact that we're winning, so we're doing better than zero, so I'm making 4% with my active, you know, I'm sitting here on Robinhood. The worst thing that happens is you're making

10. 30:16 – 31:56

    System 1 vs system 2

    1. ADAnnie Duke30:16

       4% on Robinhood. (laughs) Like-

    2. CWChris Williamson30:18

       That's, that is, that is the worst situation to be in. It's so interesting to hear this, um...... this same sort of reflection come about in multiple different areas of human nature, right? So let's think about the system one versus system two thinking, or the fast and slow from Daniel Kahneman, the explore versus exploit that James Clear talks about in Atomic Habits when you're doing habit setting. Uh-

    3. ADAnnie Duke30:39

       Yeah.

    4. CWChris Williamson30:39

       ... Taylor P- Taylor Pearson's Consciously Designed Life versus Execution Designed Life, which is sitting back, thinking very, very deliberately and having to pay the entry price that is being super, super conscious about every little area that you do. And then the hope is that over time, that becomes absorbed into your mode of being, uh, because you can... Perhaps when you- when you're a child, it's really, really important that you don't put your hand on the hot stove. As you get a bit older, that decision still has the same level of consequence if you get it wrong, but the chance that you get it wrong is significantly lower, so you don't have to be conscious about it, but you do when you're a kid. The same as the reason why they've got those little foam cups on the edge of all of the sharp tables that are at eye height-

    5. ADAnnie Duke31:22

       Yeah.

    6. CWChris Williamson31:22

       ... for children in houses, because they might run around and not see a table and then lose an eye. When you're older as a adult, that's usually not the concern that you have, despite the fact the consequences are the same. So, this is how we expand our domain of competence, by being deliberate, by learning, by probably making some mistakes, hopefully not losing an eye, and then we allow ourselves to move on to whatever comes next. What do you think-

    7. ADAnnie Duke31:44

       Yeah.

    8. CWChris Williamson31:44

       ... most people presume wrongly about good decision-making?

    9. ADAnnie Duke31:50

       Uh, that a good decision will get you a good result. And that

11. 31:56 – 39:56

    Good decisions vs bad decisions

    1. ADAnnie Duke31:56

       a bad decision will get you a bad result.

    2. CWChris Williamson31:58

       So, this is the difference between being-

    3. ADAnnie Duke31:59

       Right? I think that's literally the main thing that people get wrong.

    4. CWChris Williamson32:01

       This is the difference between being right and being accurate, right?

    5. ADAnnie Duke32:04

       Yeah, exactly. So, um, mostly, if you have enough cracks at it, uh, good decisions will actually get you good results, but that's in the aggregate. So, in the aggregate, that- that is true, um, that, uh, assuming that you have managed your risk of ruin well, so that you still have something to bet, um, whether it's an eye or money, um, making repeated, you know, repeatedly making winning decisions, you will actually end up winning. That's true in the aggregate. But what people think is that that's true in the short run as well. Um, and I think that's- that's really, like, the biggest mistake that people make. So, what ends up happening is that, uh, essentially, we over-index on the outcome, this is just resulting, um, in order to sort of try to work backwards to what's that mean? Just take us through over-index on the outcome that's resulting. What's that mean? Okay, so, um, let's... I'll- I'll... So, let's say that I, uh, traveled through an intersection safely, um, did not get in an accident, nothing remarkable happened. What people will say is that that must mean that I made good decisions as I went through the intersection, and if I travel th- through the intersection and I get in an accident, they'll say I must have made poor decisions. This is true, like, um, uh, I make an investment and it loses. I made a mistake. I make an investment and it wins. I must be a genius. Um, I order the chicken instead of the fish in a restaurant and it's gross. I made a mistake. I order the chicken instead of the fish and it's amazing. I knew it, right? Like, obv- obviously great decision. Um, so that's resulting. It's basically saying if I know what the quality of the outcome is, that this tells me something about the underlying decision process. Um, the issue, of course, is that I might have gone through a red light when I got through safely, and I hardly think, now that I've told you those details, that you would consider me a good decision-maker in that case. Um, I may have gone through a green light and a car came from the other direction and, you know, what could I do? Um, and that doesn't really make me a bad decision-maker, that it happened to turn out poorly. Um, obviously same things with ve- investments. I mean, the- the- the whole thing, obviously, is that there's volatility. So, uh, you can make a perfectly sound investment, um, when you're thinking about it in terms of just expected value, sort of what's my expectation in terms of whether I'm gonna win more often than I lose, um, and if you keep making choices where you're gonna win more often than you- you lose, you're probably making pretty good decisions. But you can lose, like on, in the short run. I can make an investment and I can lose, and it just turns out really poorly. So, uh, so I think that this is the biggest thing that kind of, like, really frustrates our ability to learn, because the only thing that we can learn from is experience. That's it. That's- that's what we have, aside from, you know, sort of what the build- built-in software is that we come with. Um, we've got to learn, uh, about the world as the world kind of tells us, like, you know, here's some data, and then we sort of process the data and we figure this out and we're trying to close these feedback loops and we're trying to learn. And what ends up happening, of course, is that we- we're pretty poor at learning, because we aren't aggregators. We just naturally don't sort of wait around until we have enough data to understand something that's significant about, uh, the set of outcomes that we might be observing, and instead, we process those outcomes in sequence, um, and we don't really know what we're supposed to take from the outcome. And in general, we just kind of take really bad lessons from it, and it's a really strong cognitive illusion. So, in Thinking in Bets, you know, I talk about the Pete Carroll example in the Super Bowl. He makes a pass at the end, mathematically a really good play. Um, it- he loses the Super Bowl because he chooses this- this particular play to call, and people still today talk about it as a mistake. Um, likewise, um, you know, I think people are pretty familiar with America's 2016 election, um, and Hillary Clinton losing that election. So, uh...D- again, to show you like how widespread this is and how much it really frustrates our decision-making and our learning and the lessons that we learn, there's general consensus that Hillary Clinton made a really huge error, like really bungled her election strategy, her campaign strategy. Um, and it was the error that everybody seems to agree upon is that she did not, uh, campaign in the Rust Belt, which is really kind of combined of, uh, we're talking about three states, Michigan, Wisconsin, and Pennsylvania. Interestingly enough, Ohio is one of those states that doesn't get mentioned as one of the states that she bungled, uh, even though it's part of that Rust Belt. And I think the reason why is that she wa- she w- lost that state by a lot, so there would have been no reason for her to go and actually campaign there. But so she loses Michigan, Wisconsin, and Pennsylvania, razor-thin margin. Across the three states, it's a total of like 80,000-ish votes. Okay. So, um, so everybody starts writing pieces about her terrible Rust Belt strategy and how she really screwed it up and what a mistake she made and how she bungled that part of the US and, uh, and this was horrible what her strategy was. So here's how we can understand, um, where our ability to understand what a good decision and a bad decision is really goes wrong, 'cause what people are saying is she made very poor decisions. But here's the thing about presidential elections, it's true about any national election, it's true whether you're in Britain or Germany or France or America or wherever, is that everybody has something to say about it while it's happening. That's whether it's a seasoned political strategist, a pundit, people come crawling out of Si- Silicon Valley who are like data analysts who write a billion pieces on Medium. Um, (laughs) so everybody's got something to say. So I think that we should assume that given the data that was available at the time, that if this was a really huge mistake, there would have been lots of people writing about it. Because we have to remember, that's that idea of like what do we know at the time that we decide, right? And our knowledge is always limited in some way. We don't have perfect knowledge. We certainly can't foresee the future. And there's just a whole bunch of stuff we don't know. Um, so we have to judge decisions based on the knowledge that's available at the time that somebody makes the decision. And I would agree, it would be a huge mistake if there were a lot of people who were yelling and screaming about Michigan, Wisconsin, and Pennsylvania and the fact that, uh, it seemed like even though the poll said that she was way ahead there and running even in places like Florida and New Hampshire and even-ish in like Arizona, um, which is where she was spending her time, that maybe she should skedaddle out of Florida and get to these three states. There, if there were lots of people saying that, uh, and she did not know that herself, I would agree. She was not using the information that was available to her. But you can look it up on Google for yourself. There are crickets. Now, you do start to see tons and tons and tons of articles about this particular issue, but the first one of them appears on November 9th, which is the day after-

    6. CWChris Williamson39:49

       (laughs)

    7. ADAnnie Duke39:49

       ... the election.

    8. CWChris Williamson39:50

       How fortunate.

    9. ADAnnie Duke39:52

       Yeah. There w- were, by the way, two really big long articles about,

12. 39:56 – 41:46

    The cognitive illusion

    1. ADAnnie Duke39:56

       I think it was Pennsylvania in particular, and they were critiques of Trump.

    2. CWChris Williamson40:01

       Post-hocking like that, to me, reminds me a lot of when I watch my housemate play COD. So I've, I've got into, during the pandemic, I've really, really gotten into watching him play Warzone online, terribly. He's awful.

    3. ADAnnie Duke40:18

       (laughs)

    4. CWChris Williamson40:18

       Toby, you, I love you, man, but you suck at (laughs) at COD. You know, e- e- this is, not only does he suck in terms of how I get to see the, uh, more subjective areas of his game, like when he crouches and when he sprints, but there's objective measures as well in that he sometimes goes through a whole game with zero damage being made, et cetera, et cetera. But I'll sit there. I know, it's, uh, uh, you could just pull the trigger for the whole game and probably hit s-

    5. ADAnnie Duke40:42

       But I literally don't even know the game, and I know that has to be a bad strategy. (laughs)

    6. CWChris Williamson40:45

       It's not a good thing. It's not a good thing. Um, but I'll happily sit there and abuse the g- with no control on myself, just with like a chocolate bar or like a glass of water in my hand, and I'll happily abuse him and tell him that, "No, no, no, you should have gone left. Oh, why didn't you go left?" Like how p- I knew that left was a good thing, but I don't, I only ever say that like after he's dead and on the floor or like doing this-

    7. ADAnnie Duke41:07

       Right.

    8. CWChris Williamson41:07

       ... signal thing where he needs someone to come and revive him. So it's, it's precisely the same as this, right? And everyone's a, everyone's a critics, everyone's uncle.

    9. ADAnnie Duke41:15

       E- exa- exactly. It's like e- d- and this is why I say it's a real, it's a, it's a really robust cognitive illusion, um, in the same way that when I look at a visual illusion, you can explain to me what the illusion is, but I cannot unsee it, right? Like you can see like the, on Twitter, you'll see like all these great illusions, like it'll be a stationary cube and because the background is flashing, you'll see the cube spinning, for example.

    10. CWChris Williamson41:43

        Is that Steve Stewart's thingy that you follow?

13. 41:46 – 46:38

    Steve Stewarts Thingy

    1. CWChris Williamson41:46

    2. ADAnnie Duke41:46

       Well, so he... Yeah, so he'll post them. Phil Te- Tetlock will post them. There's a Japanese person who posts tons and tons of visual illusion.

    3. CWChris Williamson41:53

       Yes, it's Steve, Steve reposts his stuff, doesn't he?

    4. ADAnnie Duke41:56

       Yeah, exactly.

    5. CWChris Williamson41:57

       Yeah.

    6. ADAnnie Duke41:57

       I'll repost them sometimes. Phil Tetlock has been reposting a lot of them lately. So the thing is, you can explain to me what's happening, right? You can say to me that, "The cubes are not moving. It's the fact that the background is flashing that's making you see the cube spin," and you can... and I'll go, "Oh, that's really interesting," and then I look at it and the cubes are spinning.

    7. CWChris Williamson42:13

       Yeah.

    8. ADAnnie Duke42:15

       Um, and this issue of resulting is like that. So, um, I know what... I know that nobody was talking about this before November 9th. I know that. And it's very hard for me not to feel deep down inside of me that Hillary Clinton made a mistake.... it, it's hard for me to get over that, um, because it really does feel that way. Like, God, when that, that food comes to you in the restaurant and it's disgusting, it's just hard not to go-

    9. CWChris Williamson42:43

       Knew I shouldn't have ordered that.

    10. ADAnnie Duke42:45

        ... "Ah, I should have... Ugh." So, so that's kinda the, the first thing that has to do with, like, this over-indexing on outcomes, but the other thing that comes from that is also that we misremember what our state of knowledge was at the time. So, that's what you're talking about with, "I knew you should have gone the other way." Right? So, there's a little bit of a difference between, um, uh, you know, somebody saying, "It was the wrong choice for you to go right," as if that has now revealed itself to them afterwards, versus saying, "I knew beforehand that you should have gone the other way," regardless of whether you said it or not. Um, this actually happened to me with this Hillary Clinton thing, 'cause I pitched it to somebody at a, a major newspaper, and they said, "Oh, I can't run this because I did actually know this, and I had a bunch of friends who knew it, and we were all talking about it and reading about it." And I'm thinking, "You're at a major newspaper. You could have literally published that, but you did not. So, I'm not exactly sure why you kept that incredibly contrarian opinion that, that may have gone viral (laughs) to yourself."

    11. CWChris Williamson43:49

        It's so, it's like-

    12. ADAnnie Duke43:50

        But you apparently did, but, but literally, they're misremembering what their state of knowledge is. So now think about how hard it is when we talk about, like, we were talking about what makes a good decision process, and part of it is this ability to go back and examine. If you can't even remember- (laughs)

    13. CWChris Williamson44:06

        (laughs)

    14. ADAnnie Duke44:06

        ... what you thought at the time, if you can't remember what your state of knowledge was when you made the decision, how on earth are you gonna create any fidelity around that process? Like, how are you gonna go back? And then what are the lessons we're learning? So, I'm seeing it in the United States right now, where people... Everybody... Like, you can... Everybody's talking about like, "Oh, yeah, you know, Biden's running ahead in these three states, but I'm sure it's wrong," that, you know... 'Cause obviously there was a polling error in those three states. The thing about a polling error is you can't find out until after the fact. So now everybody has decided that we all knew this, and, you know, now the only strategy that you could have possibly applied to a presidential election would be to spend all your time in Pennsylvania, Wisconsin, and Michigan, apparently. What a terrible lesson to take from that. The lesson should actually be, is there a way for you to detect which polls are more accurate and which aren't? And if they're not accurate, can you predict which way they might not be accurate? 'Cause there's no reason in this election cycle that we d- we should assume if there's a polling error that it would go in Trump's favor versus Biden's favor, right? It can go... That, that can actually go both ways. And we know that this is actually quite hard because in 2016, the polls nationally were dead on, and in states like Florida and New Hampshire, they were also dead on, right? So, we had this very specific polling error. I would actually argue that in this election cycle, it's probably more likely... And I don't know, I'm just making this... I would have to talk to a pollster 'cause there may be just, um, structural problems to polling those states. But assuming there aren't structural problems, I would assume that they would be polling more accurately, because I have to imagine the pollsters are like super-duper, like, "Holy crap, I can't get these three states wrong." And so if there were a polling error, my guess is if, if I were to bet, given my knowledge, I would want to go find out more before I place money on it, but if I had to bet, I would say the polling error would be more likely to occur outside of those three states right now, just because I have to assume there, that people are really kind of looking at those. That's assuming there's not something structural about those states that would make polling them quite hard. Uh, but you can see that people are kinda taking the wrong lessons. So, you can have a candidate now come along and just sort of based on this idea that she made a horrible mistake, say, "My campaign strategy should be to really campaign hard in those three states," but that could be completely a terrible campaign strategy. You're just learning really bad lessons.

    15. CWChris Williamson46:27

        I don't envy the job of a campaign manager at all.

    16. ADAnnie Duke46:31

        No.

    17. CWChris Williamson46:31

        It's just-

    18. ADAnnie Duke46:32

        No, you're just gonna... Like, if you win, you're a genius. If you lose, you're an idiot.

    19. CWChris Williamson46:35

        I know.

    20. ADAnnie Duke46:35

        There you go.

    21. CWChris Williamson46:36

        It's, it's so, so hard.

14. 46:38 – 54:56

    How to spend your time more wisely

    1. CWChris Williamson46:38

       Um, how can people spend their decision-making time more wisely? Talk about this in the book, and I'm very, very, very interested in the opportunity cost of not making decisions, of how long it takes to make one. Obviously, it's all well and good us saying, "Here, be conscious with your decision-making process, and let's have a... This, that, and the other," but you need to spend that time as wisely as possible, right?

    2. ADAnnie Duke47:00

       Yeah, so, so we've hinted at it a little bit 'cause we've been talking about, like, what road are you on, and are you getting off the road, and that kind of thing, which is a little bit of unlocking it. So, the idea is, look, you wanna, you wanna obviously understand what a really robust decision process looks like, um, but then I said sometimes you're gonna go really fast. I would not want someone to think that you should never go fast. The question is, what are you going fast on and what are you going slow on? How are you valuing the time that you're taking in order to make the decision? Because every minute that you spend gathering information and considering options and try to kind of map these things out is a minute that you can't get back. So, what we want to generally be thinking about is, how can we be sort of maximizing the amount of time that we're spending on things that are going to, um, bear fruit and minimize the amount of time that we're spending on things that aren't gonna be particularly fruitful? And that includes your decision-making time. So, you want to spend your decision-making time on places where there's really gonna be a lot of value to doing so, and not spend so much time on places where there isn't gonna be a lot of value to doing so. And actually, sometimes, um, you're costing yourself a lot by, um, not going super fast, and there's two ways that you can cost yourself a lot by not going super fast. One is that options expire. So, sometimes by going slow, the opportunity actually goes away. Um, uh, so actually, that was a really big problem with coronavirus response, was that options kept expiring. Um, so a simple option, for example, uh, would be, um...... uh, make sure when you get the very first intelligence that there's something bad a-brewing, um, that you ramp up production of PPE. Right. So, uh, the longer you wait to do that, uh, the option to actually have that be a reasonable solution to mitigating the risk of the virus kind of s- goes away, 'cause obviously as soon as you have, like, widespread community spread, it's helpful, but then you're trying to sort of roll back a disaster as opposed to prevent a disaster. So that, that's an example of, like, the opportunity going away. Obviously, if you're thinking about investing in a company, at some point they close the round. (laughs) So, so there's that cost to it, but the other, the other cost can be that sometimes the thing that you really need to know or that you wanna know is, um, how the world responds to the decision that you made, or how you respond to the decision that you made as you're trying to sort of gather information. Um, and the best way to do that is to, like, to do a lot of stuff really quickly in order to, to get lots and lots of information so that when you do have to make really significant decisions, you have much better models of the world. And this is true, like, of our own preferences. You can think about this as, like, um, a priori, like ex ante, there are certain things that I won't know whether I like or not. Right? It, it could be a food that I've never encountered before, and I, I kind of don't know ex ante. The only way that I can know whether I like that food is to actually taste it. So there's no reason for me to research it and go ask a lot of people their opinions about the food and so on and so forth. I might ... Uh, the best thing I can do is just try it, um, and that's true of a lot of stuff that we can ... that, that we wanna be doing. So w- the question then is, okay, so how do you sort of sort through that stuff to figure out what are the things that you can go pretty fast on and what are the things that you can go pretty slow on, or you should go pretty slow on? And the answer basically all comes down to a core principle, which is in cases where when you experience the downside it's not that bad, you can go really fast. Because what we're really trying to do when we make decisions and we take our time is to try to reduce the probability of a bad outcome occurring and increase the probability of a good outcome occurring. So we wanna make sure that on balance, we're winning the d- the decision, that the upside potential is greater than the downside potential. And the more that we can create a spread between those two things, the better off we are. Occasionally it's sort of like you're choosing among bad options, 'cause that does come up. Sometimes you're kind of in a no-win situation, and even then you're trying to increase the spread between those. You're trying to figure out which one has the least ... is sort of the least bad. Um, sometimes you have all good options and you're trying to figure out which one is the most good. But all of these are just sort of weighing upside and downside potential. So presumably what we're getting out of sort of s- most- mostly what we can get out of spending more time is that our, um, forecast of what the upside and downside potential is gets more accurate. So if we spend less time, we're sacrificing some accuracy. So we may end up with, uh, a little bit more downside than we would have if we had spent a little bit more time. That's why it all comes down to, in terms of the time that you spent, you spend how bad is the bad? Because if the bad isn't that bad, it doesn't really matter. You should probably just, like, move fast and break things, as they say. You should just go around, because what you're getting in return is you're not missing out on opportunities and, um, you're learning a lot. I mean, this is a way to do this fast cycling learning. So let's sort of figure out, like, okay, so how do we figure out when bad isn't bad? So we've already talked about a couple of those things, right? So one is, uh, when something's really quittable. So if I'm, uh, thinking about a decision that I'm making and I'm trying to decide do I wanna exit, um, uh, when- whenever I choose a route, um, I might experience downside on that route. There could be a road closure. But also, I'm giving up the gains that are associated with other options. Okay? So this is one of the things that I'm risking. When I know that I can easily get off the road that I'm on and go get on another road, what that means is that I'm naturally mitigating the downside. Because I can just quit when the world starts to reveal itself that a particular future is unfolding that I don't pre- find particularly favorable. Or when I think that I'm gonna like durian and I taste it and it's yucky-

    3. NANarrator53:20

       (laughs)

    4. ADAnnie Duke53:21

       ... I don't have to finish the whole fruit. I, I can just throw it in the trash and go find something to wash the taste out of my mouth. Right? So, um, so the, the ... Basically we can sort of think about this as the more liquid the thing that we're doing is, the faster we can go, because we can gather information. And as we gather that information, if we don't like what we're seeing, we can go ... we can get out and go to other options. So that's quittability. Then we also already talked about doing things in parallel. So we know that we're always exposed to a certain amount of risk. Um, but what we can do when we do things in parallel is we can just basically sort of have a bar, right? We can say, "This is good enough. I'm gonna do it, because I'm gonna do enough of it that I don't really have to worry about whether I'm gonna realize the downside on this particular choice, because across all the choices I make, I know that I'm gonna be positive expectancy, and so therefore I should make a bunch of money." Um, so obviously that's just portfolio theory, right? So I can own a whole bunch of stocks and then, um, that's great. Um, another way to think about doing things in parallel is that, uh, if I can get on both sides of a position at the same time, um, uh, then I can obviously go faster because I'm sort of betting on both things at once. I'm betting that I'm right and I'm wrong, um, at the exact same time. So that would be like, uh...... uh, in, not necessarily in every single market, but in a lot of markets, stocks and bonds would work that way, against each other-

    5. CWChris Williamson54:55

       Would-

    6. ADAnnie Duke54:55

       ... um, not

15. 54:56 – 57:41

    Exercising options in parallel

    1. ADAnnie Duke54:56

       so recently.

    2. CWChris Williamson54:56

       ... perhaps, uh, doing the night class of the career you're considering doing whilst staying in your current main, mainstream job to see if you like it or-

    3. ADAnnie Duke55:04

       Yes.

    4. CWChris Williamson55:04

       ... moon, moonlighting on a weekend as a veterinary nurse whilst you continue to do your art midweek or whatever it might be?

    5. ADAnnie Duke55:10

       Exactly. Yeah. Um, you can't decide between the chicken and the fish on the menu, so you ask me if I'll order the chicken and you order the fish. Do that. Um, so hedging would go into this category, in general. So I, I wanna have an outdoor wedding. It's my dream, but I'm concerned that it might rain. Uh, I could choose to have my wedding indoors, like in a banquet hall or something like that. Um, but notice that I lose my optionality then 'cause I can't actually have it outdoors, uh, then. Um, but what I could do instead is I could choose a location where I could set up a tent, and I may never use the tent. That's okay. Um, because I'm thinking about that in advance. So I'm, I'm basically paying something for the optionality' cause the optionality is worth something to me. Um, so I don't want my day ruined, and if the price of the tent is tolerable to me for insuring against having my day be ruined, I'd rather have an indoor and outdoor wedding essentially at the same time in a theoretical sense.

    6. CWChris Williamson56:15

       Yeah.

    7. ADAnnie Duke56:16

       Right? So I can, I can do them both at the, the, theoretically at the same time even though I'm o- only one of those things is actually gonna occur. But I've allowed myself the, a future where I could do either. Right? So that's, that's exercising options in parallel. So that al- also really helps mitigate, um, the risk, and you can go much faster on decisions that are liquid, um, and hedgeable basically. So you can think about those two things, um, or, or where you can just straight up do two things that you think are winning at the exact same time, so you're not... Then that's not really a hedge. The other thing you can think about is just generally there's a whole bunch of decisions that we take a lot of time with that actually aren't particularly high impact. And if we could actually sort of step back and say, "On its own, separate and apart from whether there's any optionality, um, is this something that's just generally low impact and I'm just getting hung up on it for reasons other than, um, what the impact of the bad outcome might be, or at least the long-term impact of the bad outcome? Um, maybe I could actually speed up a bunch of those decisions." So, uh, so an example of that is, so I, I know a lot of people who do this. I may have sometimes done this myself. Um, people who really spend a lot of time like, you know, back in the old days when we went to restaurants and dined in, um, where people will spend like forever trying to decide between two things on a menu. Do you know anybody like

16. 57:41 – 1:14:57

    The Happiness Test

    1. ADAnnie Duke57:41

       that?

    2. CWChris Williamson57:41

       Yes.

    3. ADAnnie Duke57:41

       (laughs)

    4. CWChris Williamson57:41

       Including, including some of my good friends.

    5. ADAnnie Duke57:44

       Yes. And they're like liter- they're asking everybody like, "What are you ordering? What are you ordering? What are you ordering? What are you ordering?" You know? Like, okay, well, they're different people. They have different tastes than you do. Um, and then, you know, they'll narrow it down, and then they're calling over the wait staff. They're being like, "Which one do you think is better?" Um, uh, and it's like literally they're like taking forever on this decision. But the pro- the, the reason why that's such a bad use of your time is that, first of all, you could be using that time for things that might make you, bring you more enjo- enjoyment by like actually talking to the people that you're with. (laughs) That might be fun. Um, uh, but it's really more that the, the impact of sort of like not g- being quite as accurate about which dish you're gonna like and ending up with a dish that maybe isn't so good or is a little bit more likely to maybe not be so good is so small that there's no reason to be spending really any time on this decision whatsoever. And the way that you can get there is you can, uh, I talk about this thing called the happiness test, um, which is, uh, to do a little time traveling to try to figure out what the impact of the decision is. So, um, if I see you, uh, if you order something and you eat it and it's terrible, I guess we do takeout right now, and you get, it's like gross, and it's like yucky, and you end up throwing it out. Um, and I see you a year after this happens. And I just say to you, "Hey, do you remember a year ago when we ordered takeout and the dish you got was like gross?" Um, and I, I just say, "How much of an effect did that have on your happiness today?"

    6. CWChris Williamson59:16

       Almost none.

    7. ADAnnie Duke59:16

       Any?

    8. CWChris Williamson59:17

       Almost none.

    9. ADAnnie Duke59:18

       (laughs)

    10. CWChris Williamson59:18

        Unless it un- unless it was that dish I had that had salmonella in just after I came back from Africa last year, which that, that did have a fairly long-lasting impact for about-

    11. ADAnnie Duke59:26

        Yes, so-

    12. CWChris Williamson59:27

        ... too much. (laughs)

    13. ADAnnie Duke59:28

        Apart from food poisoning, but food poisoning is, um, all things being equal, because e- if you're at a restaurant where you might get food poisoning, that's gonna be true of every single dish that you get. Um, so, (laughs) so I agree, separate and apart from food poisoning. Um, so what if I catch you in a month? You didn't get food poisoning, it was just yucky.

    14. CWChris Williamson59:47

        Don't c- I can't even remember it.

    15. ADAnnie Duke59:48

        Right. What about a week?

    16. CWChris Williamson59:50

        Mm, maybe.

    17. ADAnnie Duke59:50

        Are you still like a week later, or are you like, "I'm so sad my week was ruined 'cause I had a bad takeout dish"?

    18. CWChris Williamson59:56

        Probably not. Probably not.

    19. ADAnnie Duke59:57

        Probably not. So basically what you can get at with the happiness test, and I use happiness as like a proxy for, um, in general, like are you reaching your goals? Right? Um, is that what you're, what you can see is what is the long-term impact of, of sort of observing the bad result? Right? Like, if I actually observe the future in which the dish is gross, or I don't like what I watched on television, or the book was kinda bad, or the movie that I went to go see kind of sucked, or I had a bad date, or whatever it is, um, the l- the shorter the time period in which I can catch you and you tell me that didn't have any effect on your happiness, the lower impact that decision is gonna be. So we can see that. Like, if you're making a choice about getting married, where for most people, but not all, but for most people, that would be, um, a choice to be committed to one person...... and not have other people involved in the equation, um, that is gonna have a pretty long-term effect on your happiness. If your marriage isn't gonna go w- isn't going well, it's gonna matter to you, right? But if you kind of have a bad date, it's not gonna have a long-term effect on your happiness. So here we can separate out the things that really are gonna have a long-term effect on your happiness and the things aren't, and the things that aren't gonna have a long-term on your happiness, honestly, you spent- shouldn't spend too much time with. And in general, as we think about where are we gonna allocate our decision-making time, what we wanna be doing is thinking about it like a menu. The, like, a really good approach to a menu is basically to sort the menu into options that you like and options that you don't like, and once you get to options that you like that are kind of in that bucket, you can generally pretty much sort of be flipping coins among them. Um, and one of the good ways to kind of clarify that for yourself, like, what belongs in the bucket of things that I like, is to apply the only option test, uh, which is essentially to- to take anything out of that bucket. So we can think about it like a menu, right? If this were the only dish I could order, would I be happy with it? And if the answer is yes, you can put it in the like bucket. If the answer is no, I'd be really sad if that was the only thing I could order, you- you sort it out. So your- your decision-making time should be spent on the sorting, but not on the picking. You can think about once things are sort of in this bucket of stuff that I like, then anything in there is just picking, and you may be under the illusion that you can separate out whatever small differences there are in expectation for any of those individual options, but we're deciding under uncertainty, right? Like, we don't have a whole bunch of information, least of all how it's gonna actually turn out, which is the thing that you would wanna know. So the- you're not actually probably gonna be able to sort out with any fidelity, with any accuracy, what the slight differences are between the items that are in your "okay, this is a thing that I think I'm gonna win" too bucket, um, so stop trying. Like, just go ahead and flip a coin once you've got them in a bucket. Now obviously, um, the more information that you have available to you, the more, the more accurately your modeling that- that information and that data, your bucket is gonna be smaller. But regardless, that bucket is always gonna hold more than one option. Right? So the less information you have, the kind of, you know, if your model isn't very exact, your bucket may be much bigger. The more information, the more data, all that stuff that you have, that bucket mu- might be much narrower. But once things are in the bucket, stop trying to pretend that you can actually split them apart. Really bad use of your time. That's where you really lose opportunities to go find other things to get into the bucket. That's where I wanna be spending my time. I wanna find other things that I can put in the bucket. I don't wanna spend my time sorting among things in the bucket. I wanna go get other things into the bucket. That's where all of my time should be spent. And then you're gonna have fewer opportunities expire when you do that.

    20. CWChris Williamson1:04:06

        I like it.

    21. ADAnnie Duke1:04:06

        And the only thing that I would say is that once you do have things in the bucket, to your point, um, one way to decide, not, instead of trying to figure out, like, what's the difference in expected value, just say they're both in the bucket. Is there one that's more equitable? Can I do these two things at the same time? Right? You can ask those types of questions about optionality, and that, that will help you do some sorting of the things in the bucket that may help you actually get a higher fidelity choice. But all things being equal, once they're in the bucket, just flip a coin.

    22. CWChris Williamson1:04:38

        I wrote a newsletter last week, which is actually about my golden rule for decision-making, which is far, far less sophisticated than the stuff that you go through in your book. But, um, I found this quote, I can't even remember where it was from, that said, "The quality of your life depends upon the quality of your decisions." Therefore, making precise decisions is, is quite important. Uh, and the rule that I came up with wa- or the rule that I tend to use, which takes the third-party perspective from Shane Parrish from Farnam Street, uh, and a little bit of stuff from James Clear, was to ask myself the question, "What would me tomorrow have wanted me today to do?" Like, you've got these two things. I want a cookie, I want to not go to the gym, I wanna do whatever. Um, the third-party perspective, being able to take that, that third-party perspective is super useful, um, because it helps to give you just a little bit more distance between the present self, which we all know is just a, a total bastard, uh, and the remembering self, which is actually really, really useful. So if you were to write-

    23. ADAnnie Duke1:05:36

        Yep.

    24. CWChris Williamson1:05:36

        ... a couple of golden rules that people can take away and can help them to make better decisions, is there stuff that people can remember to sort of keep in their mind? A couple of little maxims, Annie Duke's maxims that you've got that they can take away?

    25. ADAnnie Duke1:05:51

        Yeah, so, well, first of all, let me just say, uh, your point about the time travel is really, uh, that's really key. That's what the happiness test is doing, right? In the moment, you're like, "Ugh, if I get the chicken and it's bad, I'm gonna beat myself up thinking I made a mistake." Whereas what this does, it says, when you're looking back in a year, are you even gonna care? And that allows you to get into a more rational place about, about, um, sort of how much time you're spending and how much worry and how much anxiety you have around that decision. Do I have any maxims, um, that will help people with their decisions? Mm. Yes. Okay, here's one. If you wanna know what somebody thinks, don't tell them what you think first.

    26. CWChris Williamson1:06:32

        Why?

    27. ADAnnie Duke1:06:34

        So, one of the things that you're talking about when you say, like, the remembering self, is, that's actually, you know, in Thinking, Fast and Slow, obviously, it's a discussion of the inside and outside view, right? So getting to the remembering self is actually a way to get to the outside view. In other words, to get outside of the things that are driving the decision in the moment, where you're driven by your own perspective and your own biases and...... kind of what you want to be true of the world, um, to outside of yourself, by imagining yourself as an outside observer, 'cause that's really what you're doing when you're thinking about the remembering self is you're trying to imagine how somebody else would view this decision from the outside. So, uh, one of the most valuable decision tools that we have is to get to the outside view, to think about how, how would somebody else view this, the, the problem that I'm considering. Um, and we know that two people with the exact same data can come to very different conclusions. People apply different mental models. They have different perspectives. They may have different information than you have. They could have a different fact that would be very helpful to you. And so, uh, we'd like to, in order to become better decision makers, to live less in the inside view and more in the outside view. We wanna, we wanna be living the outside view, which I suppose is another maxim. All right. So, so how do you get to the outside view? Well, one of the ways is to go find out what other people think, because, um, a lot of the outside view, by definition, kind of lives in other people's heads. Uh, they've got a lot of the information you don't have, and they have a lot of perspectives that you don't have, that's, you know, separate and apart from base rates, which you can go google. So, one of the problems that we have is that when we communicate to each other, um, human beings really like to be members of a tribe, uh, and part of what creates cohesion in a tribe is, uh, that you believe the same things to be true. Uh, we see this in politics. Um, and so human beings are kind of like agreement machines. W- uh, if you've ever been in a meeting, which I'm sure you have, you'll notice that a lot of the meeting is spent on areas where people agree. That makes us feel good. It makes us feel like we're, like, on the same page and a team player. But that's true just, like, interpersonally as well, right? So, you know, we've all been at, like, cocktail parties where someone says something that you totally disagree with and then you're like, "Don't speak up" (laughs) 'cause it's like, you don't wanna be rude. So, um, what you always need to be thinking about is that your beliefs can really infect other people. The, uh... So when I tell you what, what I believe to be true of the world, I am much less likely to find out what you think is true of the world. When I tell you what I believe the best course of action is, you are much less likely to tell me what you think the best course of action is, not with any fidelity. And the worst thing I could do is if I'm asking you about something that happened in the past, and I'm trying to figure out whether I should have turned left or right in my video game, if I tell you I died, then I've ruined the whole thing, right? So you need to hide that information from the other person. And actually, what your goal is, by not telling them all the things that you think or, you know, that, that are, um, or like the outcome, is that I'm trying to put you into the same state of knowledge that I was in at the time that I made the decision. Because when I tell you my belief, there's sort of a couple of things that could happen. One thing is that you could believe a different thing, but you could think I'm wrong, and so you don't speak up. Another thing is that, um, you could believe a different thing, but either you don't wanna embarrass me or you wanna move the meeting along, or you don't have enough skin in the game to, like, tell me or whatever. So you just suppress it for that reason, not, not because you think you're wrong. You actually think you're right, but you don't actually speak up. And the third th- thing that can happen is if I'm speaking and I tell you what my opinion is, that your opinion can shift while you hear me. And so by the time that you actually speak up, you've come to a place that agrees with me more, even though you didn't start in that place. This is obviously particularly problematic for people who are perceived to be leaders or subject matter experts. So you can imagine if I were asking for your advice on a poker hand that I played, and I told you how I played it, I said, "The, uh, the person in front of me raised," and I tell you all sorts of details about the person that you would need to know, in other words, the knowledge that I had at the time. And then I said, "I looked down, and I had an ace queen, and so I raised him back because I'd seen him play and I felt like, uh, he was probably pretty weak because he'd been playing a lot of hands. And so I felt like, A, I probably had the best hand, and B, I thought there was a good chance that he was gonna fold," and so on and so forth, which is really how we communicate to people because somehow we think that my opinion's really valuable data for you to have. What do you think is gonna happen now? If you didn't think that... If your, your instinct was to call, you'd go, "I must be wrong." And now I actually don't get to hear that opinion from you.

    28. CWChris Williamson1:11:33

        Interpersonally, that's a really interesting challenge for people to come up against, I think. It's something that I learned, uh, during the process of this show, and I brought up a couple of times with, uh, communication experts, that allowing a short question to sit during a conversation is really uncomfortable.

    29. ADAnnie Duke1:11:52

        Mm-hmm.

    30. CWChris Williamson1:11:52

        And I noticed when I was talking to people that there would be this sort of weird after-effect echo thing. I'd say a question, and then (imitates echoing) 　 and then ask a question, (imitates echoing) and have this... What's this trai- what's all that? What was all of this, like, extra words that I needed to say at the end of it? But I think sometimes, especially if you were speaking to someone who perhaps was an authority in a particular subject field, or even just someone whose r- opinion you respected generally, you would ask a question and then clarify... I don't kn- just do something that dampens the impact of sitting in silence, because we hate silence, and if it's silent-

Episode duration: 1:30:11