back to index

Annie Duke: A Professional Poker Player's Guide to Making Better Decisions


Whisper Transcript | Transcript Only Page

00:00:00.000 | I play tennis a lot, and a ball hits the net court a lot.
00:00:08.120 | Sometimes it plops over to the other side of the court, and sometimes it plops over
00:00:12.480 | to your side of the court.
00:00:14.400 | This can be on very big points that this happens.
00:00:17.920 | When it plops back over to your side of the court, you're super sad, and you feel like
00:00:22.160 | you're really unlucky.
00:00:24.040 | But then what I say to myself is, "But it's gone the other way for me at really important
00:00:28.400 | times as well."
00:00:30.560 | So I try to always remember that.
00:00:32.520 | And the question then I try to focus on is, "Did I pick the right shot?
00:00:38.680 | Should I have done something which was going to have more clearance over the net?"
00:00:43.440 | So then I start to think about, "Was it the right choice?"
00:00:47.240 | I try to get back to the decision itself, as opposed to just the outcome.
00:00:51.040 | Hello, and welcome to another episode of All The Hacks, a show about upgrading your life,
00:00:54.840 | money, and travel, all while spending less and saving more.
00:00:58.000 | I'm Chris Hutchins, and I'm excited to have you on my journey to optimize my own life
00:01:01.320 | by sitting down each week with the world's best experts to learn the strategies, tactics,
00:01:05.760 | and frameworks that shape their success.
00:01:08.040 | Today I'm talking with Annie Duke, a former professional poker player turned bestselling
00:01:11.960 | author and what I'll call a professional decision maker.
00:01:15.600 | Her 2018 bestseller, Thinking in Bets, Making Smarter Decisions When You Don't Have All
00:01:20.000 | the Facts, explored how life is more like poker than chess, and has helped countless
00:01:24.260 | people make better decisions.
00:01:26.140 | And her latest book, Quit, The Power of Knowing When to Walk Away, is all about learning to
00:01:30.080 | quit well, which is a skill that can not only be learned, but that should be learned and
00:01:34.360 | practiced by just about everyone.
00:01:36.060 | We're going to talk about making good decisions, getting better at quitting, but more specifically,
00:01:41.440 | why learning the difference between luck and the quality of your decisions is so important,
00:01:45.480 | and how that can improve your life and your career, why just looking at the outcome of
00:01:49.460 | decisions isn't enough, what forces work against good quitting behavior, and why so many people
00:01:54.480 | struggle with the decision to walk away, how to use kill criteria to determine when to
00:01:58.560 | quit a project, a job, or some other endeavor, and a lot more.
00:02:02.040 | Annie, welcome to the show, thanks for being here.
00:02:09.840 | Thanks for having me.
00:02:11.460 | You say there are only two factors in how life turns out, luck and the quality of your
00:02:15.480 | decisions.
00:02:16.480 | You don't have control over luck, but you do have control over the quality of the decisions
00:02:19.280 | you make.
00:02:20.360 | How can people tell the difference between the two, and more importantly, improve the
00:02:23.560 | quality of the decisions they make?
00:02:25.240 | Yeah, so that's like the $64,000 question.
00:02:29.960 | So let's just first of all say, when you make a decision, there's two forces that are going
00:02:34.480 | to influence that outcome of the decision, right?
00:02:38.480 | One is luck, and one is the quality of the decision itself.
00:02:42.400 | Now luck is a form of uncertainty that exerts itself on the outcome, right?
00:02:47.440 | So I could make a choice that's going to work out 95% of the time.
00:02:53.140 | That means by definition, 5% of the time, I'm going to observe a bad outcome, and I
00:02:58.280 | have no control over when I'm going to observe that 5%, right?
00:03:02.160 | So that's just kind of how it is.
00:03:05.280 | What's important to good decision-making is to accept that luck exists and try to see
00:03:10.720 | it clearly, right?
00:03:12.680 | So what we want to do is know that this is a 95% to 5% good to bad ratio on the outcomes
00:03:20.240 | that I could see and not believe something else.
00:03:23.760 | So if we forecast that it will work out 95% of the time, but it's actually only going
00:03:28.800 | to work out 65% of the time, that would be a problem.
00:03:32.440 | So it's not about controlling luck because you can't definitionally, it's about being
00:03:37.440 | able to forecast appropriately how luck might exert its influence, what the influence of
00:03:42.920 | luck on the outcome might be.
00:03:45.240 | And we know from a lot of work in cognitive bias, for example, that there's all sorts
00:03:50.000 | of different ways in which we don't see luck very well.
00:03:53.800 | The illusion of control, we think we have more control over the way that things are
00:03:56.600 | going to turn out than we do actually, or overoptimism and overconfidence, for example,
00:04:02.720 | are ways in which we will just sort of misassign the probabilities of different outcomes that
00:04:06.960 | might occur in the future.
00:04:08.420 | So that's the luck side, right?
00:04:10.500 | So part of a good quality decision is to be a good forecaster of these things, to start
00:04:15.700 | to see the outcomes in the way that they actually are true to reality, what ground truth actually
00:04:23.200 | is, as opposed to what our brains sort of think they are, right?
00:04:27.240 | So that's piece number one.
00:04:29.680 | Piece number two, the other form of uncertainty that is going to exert itself on the quality
00:04:35.600 | of the outcome that we might observe, and in fact, the quality of the decision that
00:04:39.800 | we're making is just hidden information.
00:04:42.720 | So for most things that we're deciding about, we know very little in comparison to all there
00:04:47.780 | is to be known.
00:04:48.780 | You know, like I said, the stuff we know fits on like the head of a pin, the stuff we don't
00:04:53.260 | know is like the size of the whole universe, right?
00:04:55.700 | So obviously, I'm sure you've had the feeling, Chris, of saying, "I wish I knew then what
00:05:00.420 | I know now."
00:05:01.420 | All the time.
00:05:02.420 | Right?
00:05:03.420 | So that's just hidden information.
00:05:06.340 | You're just sort of seeing the influence of hidden information that you sort of realize,
00:05:10.740 | if I knew then what I know now, I may have made a different choice.
00:05:14.820 | Why?
00:05:15.820 | Because when we're thinking about those forecasts that we make, right, I'm considering a decision
00:05:20.420 | and I'm thinking about what the possible outcomes are, each of those outcomes has a payoff associated
00:05:26.460 | with it.
00:05:27.460 | Each of those outcomes has a probability associated with it.
00:05:31.060 | So when we think about our beliefs, the things that we believe to be true, they're determining
00:05:38.300 | for us how we think any option that we're considering might turn out, right?
00:05:42.660 | What we believe to be true of the world is going to influence what we think the possible
00:05:46.720 | outcomes are of any decision that we might be considering, what the payoffs are associated
00:05:51.820 | with those and what the probability of those are.
00:05:54.620 | And in fact, our beliefs are going to inform other things as well, like what options we
00:06:00.140 | might actually consider, right?
00:06:02.900 | So we're usually considering more than one option.
00:06:05.640 | They're going to inform what we think our goals are.
00:06:08.000 | What is the point that we want to be heading toward, right?
00:06:13.420 | So beliefs change over time, not as much as they should, but they do change over time.
00:06:19.900 | And those beliefs are always going to be imperfect.
00:06:22.100 | There's always going to be inaccuracies in the things that we believe because we're partially
00:06:27.340 | behind the veil of ignorance, right?
00:06:29.460 | We just don't, we don't know a whole lot.
00:06:31.820 | So that's the second way in which we can think about how do we actually improve the forecast?
00:06:36.460 | So we can consider from the option that we're making a decision about going forward, seeing
00:06:42.480 | the luck very clearly, but then we can take a step back and say, well, if beliefs are
00:06:47.000 | at the center of that, if they're the foundation really of every decision we make, that's a
00:06:51.880 | place that we can really start to focus our energy on improving the quality of the beliefs
00:06:56.980 | that we have.
00:06:58.400 | - Is there a common decision you think people make where this kind of really comes to life?
00:07:02.800 | - Well, it's like every decision, but one of the ones that I love to think about is
00:07:07.340 | hiring, right?
00:07:09.720 | So in hiring, we can see all sorts of different things at play, but we can see a basic decision,
00:07:19.020 | right?
00:07:20.020 | You have some idea of what your goal is, what you want to hire for, what your values are,
00:07:24.500 | right?
00:07:25.500 | So that's going to be the rack, the job description, but that's going to have to do with things
00:07:32.140 | like who's the person that you think is going to be right for the role, right?
00:07:35.900 | What's the role that you think you need to be hiring for, which is sometimes not explored
00:07:41.580 | enough.
00:07:42.700 | Then you get lots of resumes in, right?
00:07:45.020 | And you start to try to fit the resumes to what you think that you want.
00:07:49.480 | And those are different options that you're considering, different people that you're
00:07:53.260 | thinking about hiring.
00:07:55.380 | There's other options you're considering, like what is the, what's the comp going to
00:07:59.860 | be?
00:08:02.100 | What's going to be the balance of say equity to cash, the balance of guaranteed comp versus
00:08:11.420 | bonus, things like that.
00:08:12.420 | But we'll put those aside and just think about the options, which are the people that you're
00:08:16.340 | considering hiring into the role.
00:08:19.260 | And then for any of those people that you're considering, whether you know it or not, you're
00:08:24.700 | thinking about what are the chances this person is going to work out?
00:08:28.060 | What's the chances they're not?
00:08:29.980 | Now we don't, we don't often do that, do that explicitly, but it's an implicit in any decision
00:08:35.660 | that we make.
00:08:36.660 | And there's a lot of luck that influences the outcome.
00:08:39.020 | So let's think about that and the hidden information.
00:08:42.620 | In terms of the hidden information, as you're considering each of those, those people to
00:08:47.580 | fill the open rack, well, okay, you have a CV, you have a few references and a couple
00:08:57.800 | of interviews.
00:08:58.800 | Right?
00:08:59.800 | Like they haven't worked at your company for a long time.
00:09:03.460 | You don't really know if they're going to be a good fit.
00:09:07.420 | You don't know really what their work ethic is.
00:09:10.260 | You have a little bit of idea, but you don't really find out most of the things that you
00:09:14.740 | would want to know until after you've decided whether to hire them or not.
00:09:19.300 | And separate from that, all the people that you say no to, you have no idea how they would
00:09:23.940 | have turned out.
00:09:25.140 | Right?
00:09:26.140 | So, so we can see this hidden information problem.
00:09:29.020 | Like I sort of say like with hiring, it's kind of like marrying someone after like a
00:09:32.800 | couple of dates and two friends telling you they're cool.
00:09:37.180 | People do that by the way.
00:09:38.900 | And what do we all think when they do that?
00:09:40.660 | Wow.
00:09:41.660 | You're kind of nuts.
00:09:42.660 | You know, when they're like, oh, I'm in love and it's, you, your first date was a week
00:09:46.780 | ago.
00:09:47.780 | You're heading to Vegas.
00:09:48.780 | Really?
00:09:49.780 | But that's kind of what hiring is.
00:09:52.100 | It's like heading a loping to Vegas after a week.
00:09:55.060 | Right?
00:09:56.060 | It's all part of the hidden information.
00:09:57.060 | And then what happens is that you find things out after the fact and you say, I wish I had
00:10:00.420 | known that before I hired them.
00:10:03.900 | And then in terms of luck, you know, when you hire, when you hire anybody most the,
00:10:10.880 | the probability of the person working out, it depends on sort of how you cut the data
00:10:15.000 | is around 50, 50.
00:10:18.460 | And you know, that's just a lot of luck, right?
00:10:20.860 | Like you sort of think they're going to be a fit.
00:10:22.660 | Did they turn out to be a fit?
00:10:24.300 | Did something happen in their personal life?
00:10:26.040 | Like after you hire them that you couldn't have known about beforehand.
00:10:30.140 | So on and so forth.
00:10:31.140 | So it's just a very kind of noisy, uh, it's a noisy decision.
00:10:36.600 | Now you can do things to improve the quality of the decision to increase the probability
00:10:40.740 | of getting a good outcome to somewhere between 60 and 65%, but that's about as far as you
00:10:45.140 | can push it because it's just too uncertain.
00:10:48.380 | So hiring is one of those places where it's like super clear that you can see the problems
00:10:51.960 | with this kind of decision making investments.
00:10:54.860 | Investing is another, but there's all obviously a huge influence of luck on the way that your
00:10:59.100 | investments turn out.
00:11:00.220 | Just look at the pandemic.
00:11:01.220 | Nobody had control over that.
00:11:02.780 | Right.
00:11:03.780 | And it had all sorts of effects on the stock market supply chain issues, which I don't
00:11:08.340 | control.
00:11:09.340 | You said there are a few things you can do.
00:11:10.340 | It's like, I now feel like I totally get why the decision just might not, I might be outside
00:11:15.820 | of my control, but it seems like there's a part of it that is in your control.
00:11:19.060 | Oh, there is.
00:11:20.060 | Yeah.
00:11:21.060 | And so what's, that's the piece that I, let's dive into.
00:11:24.180 | So let's talk about what are the things that you can do.
00:11:27.080 | So the first thing you can do is make yourself familiar with a concept called base rates.
00:11:30.900 | This is really, really important.
00:11:32.840 | So if we go to that first thing that I said, that we're not very good at understanding
00:11:36.600 | what the probability of any event occurring is, know your base rates.
00:11:41.400 | What's a base rate?
00:11:42.480 | A base rate is how often something occurs in a situation similar to the one that you're
00:11:46.520 | considering.
00:11:47.520 | Simple.
00:11:48.520 | All right.
00:11:49.520 | So I'll give you a really simple example of a base rate.
00:11:54.680 | So let's just take your morning routine, right?
00:11:57.040 | You're trying to decide when to wake up because you need to decide how long does it take me
00:12:01.760 | to get ready in the morning so that I know when I need to get into my car in order to
00:12:06.120 | get to work on time.
00:12:07.120 | So this is something that people do, you know, every morning, at least before the pandemic,
00:12:11.600 | before we're all online, there's a whole bunch of base rates involved in there, right?
00:12:15.680 | So one is how long does it, on average, take me to get ready?
00:12:19.840 | Okay.
00:12:20.840 | So I could figure that out.
00:12:23.280 | I could track that for you over a certain period of time and figure out on average how
00:12:28.200 | long it takes you to get ready.
00:12:29.360 | I could even figure out a lower bound and an upper bound.
00:12:32.280 | That's going to tell you sort of like something about risk, what's the volatility.
00:12:37.440 | And then you can sort of figure out for yourself how much risk you want to take on, like how
00:12:41.120 | important is it for you to get to where you're going exactly on time?
00:12:45.340 | You would want to be at a sort of account for the upper bound of how long it takes you
00:12:49.520 | to get ready.
00:12:50.520 | Um, if you have some slack in there, you could account for the lower bound of how long it
00:12:55.140 | takes you to get ready, but we can figure out on average, how long does it take you
00:12:58.120 | to get ready?
00:12:59.120 | And in fact, when you're deciding when to set your alarm, that's included in your decision,
00:13:04.280 | just not explicitly.
00:13:05.280 | Right.
00:13:06.280 | Then when we think about, okay, I got to get in my car and figure out, you know, when do
00:13:12.080 | I need to be in my car in order to go to work?
00:13:14.200 | That's another base rate that we're considering, which is on average, how long does it take
00:13:17.820 | me to get to work on the route that I'm considering taking?
00:13:21.060 | Right.
00:13:22.060 | Okay.
00:13:23.060 | So that's for you personally.
00:13:24.060 | We could figure out those base rates, but there's all sorts of base rates in the world
00:13:26.780 | as well.
00:13:27.780 | Right.
00:13:28.780 | What?
00:13:29.780 | Like I gave you one just a second ago, depending on how you cut the data, uh, your chances
00:13:33.740 | of success, a successful hire about 50, 50, that's a base rate.
00:13:38.860 | So when I'm sitting here thinking, um, you know, in that, what we talked about, like
00:13:44.400 | being overoptimistic, illusion of control, overconfidence, those kinds of things.
00:13:49.360 | When I'm saying I'm a great person at hiring, when I talk to someone, I can look into their
00:13:54.280 | eyes and I can figure out whether they're going to be a good fit or not.
00:13:58.960 | And I'm great at this.
00:14:00.920 | So you say to me, well, what do you think the probability is when you hire for this
00:14:04.520 | role that it's going to work out and I say, oh, I think it's going to work out 90% of
00:14:08.920 | the time.
00:14:10.280 | Well, that's too far away from the base rate to be sane.
00:14:15.040 | So what base rates do is they give us a starting point for the forecast that we're going to
00:14:18.640 | make, right?
00:14:19.640 | That forecast of what's the probability that things will work out one way or another.
00:14:23.720 | Um, and if I know that the base rate is about 50, 50, and I'm estimating that I'm going
00:14:29.640 | to successfully hire 90% of the time, I need to discipline myself to that base rate.
00:14:36.600 | Now maybe I am better than the average bear, but I can't be that much better than the average
00:14:42.560 | bear.
00:14:43.560 | Right?
00:14:44.560 | That's the thing.
00:14:45.560 | So, so maybe I can say like, Ooh, I'm great.
00:14:47.160 | I'll be 60%, but notice how far that is away from 90%.
00:14:51.040 | So it's just disciplining you to the base right now.
00:14:53.400 | I actually do research on this with Phil Tetlock who wrote super forecasting along with his
00:14:57.480 | collaborator and wife, Barb Mellors.
00:15:00.000 | And this is the single most important thing we found in getting people to be better estimators
00:15:04.120 | of sort of probabilities and outcomes in the future is to teach them about race rates and
00:15:08.520 | start to discipline their guests, give them that starting point of a base rate.
00:15:12.520 | So that's the first thing that you can do for anything you're considering, try to figure
00:15:15.600 | out what's the base rate, uh, what's the appropriate reference class, right?
00:15:19.840 | So if I want to figure out how long it's going to take me to get to work and I live in Philadelphia,
00:15:24.160 | I shouldn't ask how long it takes people to get to work in Cincinnati.
00:15:27.240 | It doesn't have anything to do with me.
00:15:28.520 | Right.
00:15:29.520 | So you want to get the right group that you're comparing to.
00:15:31.760 | Um, so like if I'm thinking about hiring, I could look at, uh, averages for all hires,
00:15:38.080 | but if say, um, uh, a SAS startup, I can look at average turnover at SAS startups, right?
00:15:44.960 | And that gives me a really good base rate to start from.
00:15:47.500 | So that's piece number one is know your base rate.
00:15:49.840 | So that has to do with seeing that look really clearly.
00:15:53.200 | Piece number two is improve the quality of the information that you're getting.
00:15:57.400 | And the way to do that, honestly, the best way that you can do that is to start gathering
00:16:02.880 | independent viewpoints on the problem that you're considering.
00:16:06.120 | Okay.
00:16:07.120 | So we can go back to the hiring example.
00:16:10.120 | When you're thinking about creating the job description, um, gather, figure out a group
00:16:17.200 | of people that are going to be on the hiring committee and are appropriate to be giving
00:16:22.280 | input into what you think that role requires.
00:16:25.160 | So let's say we're talking about like a C-level executive, let's say it's like a CRO, right?
00:16:31.120 | We would, you know, want maybe the CMO to be involved.
00:16:35.360 | Certainly the CEO, um, you may want the chief product officer involved, you want customer
00:16:40.960 | success involved, uh, you want people who are doing enablement, so on and so forth.
00:16:45.480 | So you want to get the people who are sort of like in that function.
00:16:48.520 | And then the leaders that that person, the other leaders, that person is going to be
00:16:52.060 | interacting with.
00:16:53.060 | So you figure out who the people are that you think should give input on this.
00:16:57.320 | And when it's a personal decision, because I, you know, like, let's say I'm trying to
00:17:00.320 | decide if I should buy a house.
00:17:02.160 | Do I build a group of people that I get feedback from?
00:17:05.560 | Or how do you apply that to a normal kind of non-professional decision where there's
00:17:09.680 | teams involved?
00:17:10.680 | Yes, a hundred percent, you should find people that you get opinions from, but here's the
00:17:16.840 | really important thing.
00:17:17.840 | So what I was about to say is you need to ask them independently and asynchronously.
00:17:23.920 | So you just send out to them, write a job description for C level.
00:17:28.120 | You would have them write a job.
00:17:29.120 | You know, you would do a little more work, write a job description, what, uh, describe
00:17:33.480 | this person to a recruiter, and you're asking them to do this all independently.
00:17:37.620 | What are the near term challenges, near term opportunities, far term challenges, far, you
00:17:42.240 | know, things like that.
00:17:43.240 | So you're just asking for, tell me independently what you think this role should be.
00:17:47.960 | Now, why are you asking them independently?
00:17:49.960 | Because otherwise they're going to influence each other.
00:17:51.920 | Okay.
00:17:52.920 | So when you're asking for yourself, your own self, you want to use that same principle
00:17:58.080 | of how do I get independent information?
00:18:01.320 | And that's to say, um, um, you give them a bunch of facts.
00:18:06.200 | I'm thinking about buying a house.
00:18:08.840 | I assume you're currently renting in that case, maybe I'm currently renting.
00:18:12.580 | This is the location I rent in.
00:18:14.480 | This is, uh, how much money I pay in rent and utilities.
00:18:19.640 | Uh, this is how much I'm making.
00:18:22.280 | This is the estimate that I have of my job stability.
00:18:25.080 | Uh, here's the area that I'm thinking about, um, buying in, um, whatever you give them
00:18:32.040 | the facts.
00:18:33.040 | Now notice nowhere have I said, but here's my concern, right?
00:18:38.360 | Like renting gives me all sorts of flexibility.
00:18:41.000 | If I get a house and I'm kind of tied down to that house and what if I don't like it?
00:18:46.180 | And uh, I'm thinking that maybe, um, given the percentage of my total net worth that
00:18:52.580 | I would have to put into the house, it's too much, which is the way that I would normally
00:18:55.700 | ask.
00:18:56.700 | You know, like if you're sort of left to your own devices, that's the way you ask things
00:19:00.260 | in the simplest sense of, if I say to you, like, I thought Queens gambit was like such
00:19:04.140 | a great show.
00:19:05.140 | What did you think?
00:19:06.140 | Right?
00:19:07.140 | Like you've done that before.
00:19:08.140 | Yeah.
00:19:09.140 | We all do that.
00:19:10.140 | We all are very confident.
00:19:11.140 | But what instead of what you want to say is, did you watch Queens gambit?
00:19:15.460 | And you say, yes, I assume.
00:19:17.460 | Yes.
00:19:18.460 | And then I say, what did you think of it?
00:19:20.500 | And I completely withhold my opinion from you.
00:19:23.940 | Now this is really hard to do.
00:19:25.660 | It's incredibly unnatural.
00:19:27.340 | We think about our opinion as being important data for the other person to have.
00:19:32.500 | But the issue is that the opinion that I'm about to offer you is actually the information
00:19:37.300 | that I'm trying to get from you.
00:19:40.060 | So if I give you my opinion first, I'm not going to get your opinion in a way that's
00:19:45.100 | actually going to be helpful for me.
00:19:47.220 | So instead, what I want to do is just offer you up the facts that you need in order to
00:19:51.500 | give me a good opinion and then just get your opinion back in it in a uncontaminated way.
00:19:58.700 | So when I say I'm currently renting and this is how much my rent is, and this is my family
00:20:02.380 | situation, and this is how much money I have saved, and this is how much I make, and I'm
00:20:09.620 | thinking about buying a house, I would just love your opinion on that.
00:20:14.620 | Tell me what you know.
00:20:15.620 | Tell me what you think.
00:20:16.620 | So I've given you the facts that you need, and now you can tell me all sorts of things
00:20:21.140 | about what you think.
00:20:22.940 | And what you want to do is get a good range of people who have kind of differing schools
00:20:27.580 | of thought, if possible, and find out what they think.
00:20:33.980 | And then explore that.
00:20:36.260 | And then you'll have collected all the data.
00:20:37.900 | You can go back to them and say, well, I had another opinion that was kind of saying this.
00:20:43.380 | I'd love to hear what you think about that.
00:20:46.100 | But you've allowed them to express themselves in a way where they don't have to disagree
00:20:49.740 | with you.
00:20:50.740 | They don't know what you already think, because mostly when we're having conversations, we're
00:20:54.100 | just trying to affirm the other person.
00:20:57.340 | So even if I did disagree with you, or you disagreed with me and you thought The Queen's
00:21:01.620 | Gambit was a terrible show, you would say it in a super nice way.
00:21:07.300 | Instead of me just saying, what did you think of The Queen's Gambit?
00:21:09.020 | And you're like, oh, that's horrible.
00:21:10.860 | Yeah, it's much more stressful on the other side.
00:21:13.080 | Someone asks you a question with no opinion, you're like, oh, I don't know what to say.
00:21:15.900 | Do I want to?
00:21:16.900 | I don't want to offend them.
00:21:17.900 | Right.
00:21:18.900 | You have to give people permission to do that.
00:21:19.900 | Say, look, I'm really trying to just get your opinion here, so I'm not going to tell you
00:21:22.500 | what I think.
00:21:23.500 | Because this is a really tough decision for me, and I need your help.
00:21:27.500 | Understanding the base rate so you kind of have a sense of what you're expecting, which
00:21:30.580 | I guess we can get into how that might affect resulting in the future, knowing that you
00:21:36.020 | had an opinion and you weren't as far off and you don't judge your decision incorrectly.
00:21:41.220 | And then asking people with kind of a more objective manner to get their feedback.
00:21:46.420 | Any other big things to make a better decision?
00:21:50.180 | Yeah.
00:21:51.180 | So, well, do you want personal or business?
00:21:53.860 | I think personal is probably going to be much more relevant.
00:21:57.100 | Right.
00:21:58.100 | So improve the quality of your decisions as much as possible.
00:22:02.700 | Try to write things down that you're deciding about.
00:22:07.260 | Get other people, trusted advisors, particularly people who have different points of view with
00:22:12.340 | you are really important.
00:22:14.980 | Here's a really important one.
00:22:16.020 | You have to give permission.
00:22:19.020 | So when we're trying to seek information from other people, we think that permission is
00:22:25.300 | implied for them to say whatever to us, but we have to actually offer that permission
00:22:30.980 | and they have to be willing to give it to us.
00:22:33.420 | So I'm sure this has happened to you.
00:22:37.860 | You like fire someone or you break up with somebody or whatever.
00:22:42.780 | And as soon as you do that, you have a whole bunch of friends who say, "Oh, I'm so glad
00:22:47.100 | you did that.
00:22:48.100 | I thought you should have done that like six months ago."
00:22:50.540 | Has that one happened to you?
00:22:52.980 | That one hasn't because my wife and I started dating, oh gosh, 2004.
00:22:56.900 | So it's been a long time since I've had that conversation.
00:23:00.420 | I'm sure it has, but boy, has it been a while.
00:23:02.540 | But like, or you fire someone or you quit your job or whatever.
00:23:07.940 | And people say, I bet you might've said it to somebody when they break up, like, "Oh,
00:23:12.140 | I'm so happy because really wish you would've done that sooner."
00:23:15.500 | So this is like a really common thing that happens, you know, "Oh, I'm so happy you quit
00:23:22.800 | your job.
00:23:23.800 | Like you were so miserable every single day.
00:23:25.020 | I knew you should have done that like a long time ago."
00:23:28.020 | And your thought when people say that to you is, "Okay, but like, why didn't you tell me?"
00:23:33.420 | And the answer is because there wasn't any explicit permission given.
00:23:38.360 | So in order to really find out what somebody thinks, you have to give them permission to
00:23:42.780 | say what they think.
00:23:43.780 | Yeah.
00:23:44.780 | I saw this so much in venture capital that I would be in meetings with other venture
00:23:49.060 | investors and they'd give feedback to founders saying, "Oh, I think it's like pretty good,
00:23:54.220 | but it's not a good fit."
00:23:55.220 | And then after they'd turn to me and be like, "God, that company's horrible.
00:23:57.620 | It's terrible."
00:23:58.620 | I was like, "We should have just told him that."
00:24:02.180 | And I think everyone feels bad telling them that.
00:24:05.660 | And now every time someone asks me for feedback on their startup, I'm like, "How honest do
00:24:09.140 | you want me to be?"
00:24:10.140 | Like, "Do you want me to tell you to quit?"
00:24:12.900 | So I guess if you're on the other side, you can also invite someone to invite you to be
00:24:18.980 | more honest and give critical feedback.
00:24:20.540 | Yeah.
00:24:21.540 | So I actually do that with my friends.
00:24:23.060 | I'm like, "Are we having a conversation where you want advice or where you're just offloading
00:24:30.020 | your emotions?"
00:24:31.020 | Because I want to know, right?
00:24:33.220 | Because sometimes they don't really want your advice.
00:24:36.580 | So I think that's really important.
00:24:37.860 | I mean, I think that's just really important.
00:24:39.620 | I think Ron Conway, who founded SV Angel, is such a good example of doing this really
00:24:47.620 | well, being a very good decision coach to somebody.
00:24:50.860 | So he would actually tell the companies what he thought of them.
00:24:55.500 | But he would sit down and he'd basically say like, "I think we can admit this isn't working."
00:25:00.100 | As opposed to, "It seems pretty good.
00:25:03.820 | Just keep going."
00:25:04.820 | He'd be like, "Look, it's not working."
00:25:08.060 | And invariably the founders would push back on him and say, "No, but I know we can turn
00:25:13.100 | it around."
00:25:14.100 | So it wasn't so much disagreement that it wasn't working, because I think everybody
00:25:16.300 | knows when things aren't working very well, but it was always, "No, but we can turn it
00:25:19.780 | around."
00:25:20.780 | So he was trying to coach them into letting go.
00:25:26.100 | And founders are gritty by nature.
00:25:28.720 | Of course they don't want to let go.
00:25:30.540 | So when they said, "No, but I think I can turn it around," his tactic was not to disagree
00:25:36.240 | with them, which I think would have just caused a rift, but to say, "Okay, tell me exactly
00:25:43.620 | what turning it around looks like."
00:25:47.020 | So let's think about the next two months.
00:25:50.540 | So you're saying you're going to turn this around.
00:25:53.060 | What exactly does that look like?
00:25:54.340 | And try to figure out what those benchmarks are.
00:25:57.300 | Now he would sit down and do that with the founder so that the founder now owned that
00:26:01.740 | along with Ron Conway.
00:26:03.780 | And this is a really good trick for decision-making is that often your decision-making is at its
00:26:08.020 | worst when you're trying to make the decision right then, like when you're actually facing
00:26:11.980 | the decision down.
00:26:13.340 | And this is something that really improves decision quality is to think in advance.
00:26:17.220 | As much as possible, do advanced planning on your decisions.
00:26:19.860 | It's much more efficient.
00:26:21.140 | It creates much better decisions.
00:26:22.340 | So that's essentially what he's doing in this moment.
00:26:24.260 | He's saying, "Let's think about two months ago, two months from now, rather.
00:26:28.140 | So we're going to think about two months from now or three months from now, the end of the
00:26:31.900 | next quarter.
00:26:33.660 | Let's think about what this is going to look like when you've turned it around."
00:26:36.620 | And then basically just set a set of benchmarks.
00:26:40.220 | And then in three months, you can sit down with them and say, "But you haven't hit them."
00:26:44.980 | And we agreed that that's what it looked like.
00:26:46.580 | And that actually makes those decisions a lot cleaner.
00:26:49.780 | And you can do that with your personal decisions as well, right?
00:26:52.860 | Like if you ever...
00:26:53.860 | We're all in these situations, like you have a relationship where it's just not going well,
00:26:58.980 | but you feel like you can turn it around.
00:27:00.820 | You feel like you have so much time invested and you do love the person, but you're like
00:27:04.500 | unhappy, but you know you can make it change.
00:27:08.340 | There's kind of two things to do.
00:27:11.020 | Don't face the decision down right then.
00:27:13.520 | Figure out, "But what does change look like?"
00:27:17.260 | So at that moment, say, "How long am I willing to tolerate the status quo, the way that the
00:27:22.980 | relationship is right now or worse?
00:27:26.820 | What do I need to do, do I think, in order to turn this around?"
00:27:30.260 | And then however long that is that you're willing to tolerate the status quo, figure
00:27:34.620 | out at the end of that time period, say it's six months, what does change look like?
00:27:40.780 | What will it look like when things have actually turned around here?
00:27:44.340 | You can do this for a job, right?
00:27:46.020 | A job that you're miserable in, what do I need to do?
00:27:48.740 | What will it look like that it's actually turned around?
00:27:50.740 | What will have changed?
00:27:52.260 | And the reason that you need to do that is because when you're actually facing down the
00:27:56.740 | decision, that's when you're most likely to be rationalizing things away.
00:28:00.300 | That's when your forecasts are going to be really inaccurate.
00:28:03.540 | That's when the things that you want to be true of the world are going to influence your
00:28:08.500 | actions the most.
00:28:10.540 | It's when everything kind of goes to crap.
00:28:13.980 | And the way that I try to sort of put it to people to get them to understand this is that
00:28:18.180 | you can say all you want that you don't want to eat sugar, but when there's a cupcake right
00:28:21.500 | in front of you, it's really hard to say no.
00:28:24.260 | And that's true, not just of like cupcakes or pizza or whatever that are, you know, it's
00:28:29.060 | sitting right in front of you.
00:28:30.060 | It's also true like when you're facing a breakup or you're facing having to fire an employee,
00:28:37.180 | which isn't fun, or you're facing wanting to quit your job, right?
00:28:43.720 | Or breakup with a friend or whatever.
00:28:47.060 | These things are very hard to do because it's that moment where we're worried about all
00:28:51.060 | the time we've put into something, whether it's that moment where you go from in a state
00:28:56.780 | of failing to having failed, because as long as you keep going, you might turn it around
00:29:02.540 | and not have to actually sort of put that failure on the books.
00:29:08.580 | You know, like if you buy a stock and you bought it at 50 and it goes to 30, as long
00:29:12.100 | as you hold, you could get back to 50.
00:29:14.060 | Yeah.
00:29:15.060 | But selling now is kind of accepting that.
00:29:17.460 | It's accepting that.
00:29:18.860 | And that's really hard for us.
00:29:20.180 | So that's why we want to think about both.
00:29:23.940 | How do we ask for permission for people to tell us the truth?
00:29:30.020 | How do we explicitly ask people if they want the truth, right?
00:29:34.300 | So we need to have both of those things going on.
00:29:37.620 | And then how do we also figure out how the person can receive the truth?
00:29:42.660 | And that's true whether we're coaching somebody else or we're thinking about how we can do
00:29:46.860 | this for ourselves.
00:29:48.180 | And one of the best ways for us to receive the truth is for us to think about the truth
00:29:52.580 | as something in the future.
00:29:54.780 | So in the investment example, you know, this is a very common thing.
00:29:59.700 | I have some stock and I know I should probably sell it, but now that I want to make sure
00:30:04.580 | it goes up, in advance, obviously it would have been great if I had said, "You know what?
00:30:08.700 | If this gets to this point, I'm going to sell it."
00:30:10.980 | And you could even go as far as to set up the order so that you force yourself to do
00:30:14.420 | it.
00:30:15.420 | Yes, you can put in a stop loss order.
00:30:16.420 | Yes.
00:30:17.420 | If you didn't do that, how do we get past that point?
00:30:21.740 | What can you tell someone or coach someone through it?
00:30:24.500 | Just do it now.
00:30:25.500 | Just do it.
00:30:26.500 | But sometimes that's hard to just do it.
00:30:27.740 | No, no.
00:30:28.740 | I don't mean sell it now.
00:30:29.740 | I mean, put in the order now.
00:30:31.260 | Yeah.
00:30:32.260 | But let's say I'm in the moment.
00:30:33.260 | What does it mean to be trading at it in a week?
00:30:35.460 | Okay.
00:30:36.460 | So you're saying if you're worried now, then set it for a week from now.
00:30:39.940 | Don't try to make the decision today.
00:30:42.220 | Try to make a smaller future decision, even though you're dealing with it in the present.
00:30:46.700 | So look, in an ideal world, you would do it right now.
00:30:49.380 | Yeah.
00:30:50.380 | Right?
00:30:51.380 | And one of the ways that you can do it, one of the ways that you can get to doing it right
00:30:57.220 | now is to ask yourself what you could do with that $30.
00:31:03.620 | So this is a helpful tool.
00:31:05.740 | So you bought it at 50, it's now at 30.
00:31:08.500 | You don't want to sell it because you're worried about losing the $20.
00:31:15.260 | But that $20 is already lost.
00:31:16.780 | So this is something called the sunk cost fallacy.
00:31:19.980 | It's not just about stocks.
00:31:21.440 | It's about like, I'm in a job and I want to quit, but what about all the time and effort
00:31:25.700 | I've put into the job and the onboarding and the training, and then I'll have just wasted
00:31:30.780 | all of that.
00:31:33.740 | It's true of relationships.
00:31:34.740 | It's true of employees that we've hired.
00:31:37.140 | So it's basically true of everything.
00:31:39.860 | It's true, like if you buy a concert ticket, you've now paid for the concert ticket.
00:31:45.500 | There's planning that's gone involved when it's freezing cold and the driving conditions
00:31:53.660 | aren't even safe for you to go to the concert.
00:31:55.620 | You still go because you don't want to have wasted the ticket, even though it's a ridiculous
00:31:59.100 | decision to go.
00:32:00.820 | Okay.
00:32:01.820 | So we know that this is a problem.
00:32:02.940 | So one of the things to do is to shift the frame and say, instead of saying, but I don't
00:32:08.020 | want to lose my $20, say, what could I do with that $30?
00:32:11.860 | So that can be helpful in the moment because what that does is it focuses you on the opportunity
00:32:16.540 | cost of the capital or the opportunity cost of your time, so on and so forth.
00:32:21.780 | What could I use that $30 for?
00:32:24.580 | What could I use this time for?
00:32:27.260 | That kind of assumes that you've made the decision that you know you need to sell it,
00:32:30.620 | you're just struggling with it.
00:32:32.340 | But you highlighted in quit that, let's say sometimes I used to tell people, "Oh, if you're
00:32:38.140 | trying to...
00:32:39.140 | " Let's say you work at Google and you have a ton of Google stock, and I would often tell
00:32:42.180 | people as a financial advisor, I'd say, "You probably don't want all of your money tied
00:32:45.580 | in the same company."
00:32:47.300 | And people would say, "Okay, well, I'm trying to figure out if I should sell it."
00:32:49.780 | And I would ask them, "Well, let's pretend there was a mistake and it all got sold.
00:32:53.900 | Would you want to buy it all?"
00:32:55.520 | And you said in the book that that kind of like Jedi mind trick doesn't really work.
00:32:59.340 | So I'm very curious what we should do instead or maybe why it doesn't work.
00:33:02.740 | - There's two things that you can do instead.
00:33:06.820 | One is advanced planning, which we've just talked about.
00:33:10.700 | So advanced planning combined with a pre-commitment contract.
00:33:14.260 | So essentially you want to set up kill criteria.
00:33:17.020 | What could I observe in the future that would make me want to sell this?
00:33:22.220 | So sometimes you'll discover, "Oh, that's the state of the world today."
00:33:27.220 | And you may sell, but just saying like, "If I were fresh to this decision, what would
00:33:34.340 | I do?"
00:33:35.340 | Doesn't really help you.
00:33:38.060 | So the first thing is that you can think about it.
00:33:41.100 | You can do this advanced planning, you can sell kill criteria.
00:33:43.860 | Kill criteria are just what we've been talking about.
00:33:47.900 | What are the signals that I'm going to see in the future that would tell me that I should
00:33:50.660 | walk away?
00:33:53.100 | This is actually going to get you to walk away sooner.
00:33:56.340 | You won't be perfect, but it will help you do it.
00:33:59.380 | So if you talk about like a stop loss order, for example, people cancel them all the time.
00:34:03.500 | But the thing is not everybody does.
00:34:06.220 | And I might cancel some stop loss orders, but not others.
00:34:08.700 | And that means at least I'm selling some things sooner than I otherwise would have.
00:34:13.580 | So that's good.
00:34:14.900 | So a little bit of progress goes a long way, but we want to always be thinking about kill
00:34:19.260 | criteria.
00:34:20.660 | So that's like I'm so miserable in my job, but I don't want to quit because maybe I have
00:34:26.260 | equity in it or whatever, but it's like the cost benefit doesn't look good to me right
00:34:30.580 | now.
00:34:31.860 | What would have to happen over the next three months in my work that would tell me that
00:34:36.340 | things I've turned things around?
00:34:37.900 | What would I see that would tell me things are still bad?
00:34:41.380 | Write that list of things down and commit to an action associated with any of those
00:34:46.420 | things.
00:34:48.020 | So I think one of the simplest examples of a kill criteria is when people mountain climb
00:34:52.980 | there, something called the turnaround time.
00:34:56.700 | So when you're climbing Everest, they have turnaround times for each day's climb.
00:35:00.660 | So there's base camp, you know, camp one, two, three, and four, four is when you would
00:35:07.300 | actually head towards the summit, but you have to sort of go up and down in between
00:35:10.940 | those camps in order to get acclimated to the altitude.
00:35:15.260 | And for any climb that you do on the day, including summit day, they have a turnaround
00:35:18.820 | time, which is sad because they don't want you to descend in darkness.
00:35:23.520 | So for example, on summit day, when you leave camp four, the turnaround time is 1 PM.
00:35:28.840 | And what that means is that if I'm not at a certain point at 1 PM, then I must turn
00:35:34.580 | around.
00:35:35.580 | Now, why do you set that those times?
00:35:37.820 | Well, because of like summit fever, right?
00:35:41.100 | Because we know that people will continue in very bad circumstances up to the summit,
00:35:46.740 | you know, even when visibility is bad or the climbing conditions aren't good, or there's
00:35:50.460 | a snow storm, there's famous books written about it.
00:35:53.500 | John Krakauer, Into Thin Air, talking about a year when a whole bunch of people got injured
00:35:59.320 | and like four people died, because they just continued up to the summit.
00:36:03.300 | So those turnaround times are meant to make it much more likely that you're actually going
00:36:06.700 | to turn around when it's no longer safe to be climbing.
00:36:11.060 | Knowing that when you're facing that decision down, you're not going to be particularly
00:36:14.640 | good at it.
00:36:15.640 | So that's piece number one is think in advance and set turnaround times for yourself basically.
00:36:21.800 | Thing number two is get a good coach.
00:36:24.680 | So when you're coaching people about Google and their stock, you're giving them an outside
00:36:30.560 | perspective that's more likely to get them to do it today.
00:36:34.580 | So notice they're not saying for themselves, what would I do in this situation?
00:36:38.880 | That's not really the important thing that's going on there.
00:36:42.560 | The important thing is they're getting somebody's perspective, who's been there, done that,
00:36:46.440 | who kind of thinks about equity, who's offering them a perspective that's really important
00:36:51.480 | where they're getting advice that would be similar to the advice they would give somebody
00:36:57.460 | if they could get out of their own head.
00:37:00.760 | So that is really helpful is to get yourself a quitting coach, which is what you're saying
00:37:05.360 | to them there, right?
00:37:07.400 | So that's offering them that perspective that allows them also to sort of get out of the
00:37:11.480 | moment.
00:37:12.520 | What Daniel Kahneman says, the Nobel Laureate is, don't make a decision when you're in it.
00:37:17.600 | Okay?
00:37:18.600 | So notice both of these things are trying to get you to not be in it.
00:37:22.620 | One is to get you to think about the future.
00:37:24.200 | The other is to have somebody looking in from the outside who can allow you, because they're
00:37:28.800 | not in it, to see the way it looks to somebody else.
00:37:34.200 | Or we could think about it a different way, the way it would look if you were standing
00:37:37.140 | in their shoes.
00:37:38.480 | Now, the best thing to do is to combine the two, which is what Ron Conway does.
00:37:44.860 | So he acts as a quitting coach.
00:37:46.540 | He tells them things aren't really working out here, or I think this would be a good
00:37:51.520 | decision.
00:37:52.520 | But when they disagree, he then turns to kill criteria and coaches them through that.
00:37:56.520 | - And is there something to look for in a good quitting coach, like a particular profile
00:38:01.800 | of a friend or a colleague?
00:38:03.640 | - Yeah.
00:38:04.640 | I love the way that Kahneman put it to me.
00:38:06.960 | Find somebody who loves you, but doesn't care much about hurt feelings in the moment.
00:38:12.480 | So his quitting coach is Richard Thaler.
00:38:15.080 | He has a Nobel Laureate for his quitting coach.
00:38:16.920 | - Yeah, I assume most people listening don't have that luxury.
00:38:20.320 | - No, I don't.
00:38:22.800 | But what's he saying there?
00:38:23.800 | Find someone who loves you, but doesn't care much about hurt feelings in the moment.
00:38:26.600 | You need someone who cares about, like who really has your best interest at heart.
00:38:29.720 | Really important.
00:38:30.720 | - But this is not, to be clear, it's not your hype person.
00:38:33.340 | It's not the person you call who's always like pumping you up and giving you compliments.
00:38:37.440 | It's almost like the exact opposite of that.
00:38:40.000 | - Someone who really cares about how you do in life.
00:38:45.600 | Who really is rooting for you, but isn't concerned about hurt feelings in the moment.
00:38:52.920 | Because why don't we tell somebody the things that we see?
00:38:56.180 | Because we're afraid they'll be sad, right?
00:38:58.800 | Who wants to deliver the news to somebody?
00:39:01.260 | When you're talking about the VCs who are like, "Yeah, it's pretty good.
00:39:05.100 | I think you're getting there with the product market fit."
00:39:07.500 | And then the person walks out of the room and they're like, "That's just a dead company
00:39:12.040 | walking."
00:39:13.040 | Right?
00:39:14.040 | And it's like, "Oh, hold on.
00:39:15.040 | Why didn't you tell them?
00:39:16.040 | Why?"
00:39:17.040 | Because they don't want to hurt their feelings.
00:39:18.460 | They don't want to be like the Debbie Downer character from SNL.
00:39:24.160 | But the problem is they're doing a tremendous amount of harm to those founders.
00:39:28.500 | That's the problem is we think that we're sparing them harm.
00:39:32.780 | We think that we're being kind by not telling them the truth right now.
00:39:36.860 | But what we're actually doing is harming them.
00:39:39.540 | Because the founders aren't getting feedback that they need to get, right?
00:39:43.340 | So there's all sorts of things that could come from that feedback.
00:39:46.760 | One is they could stop.
00:39:49.340 | They could return the capital to the investors.
00:39:51.660 | That's a good thing.
00:39:52.660 | That's capital that can be used for other opportunities that might be better, right?
00:39:57.660 | They can free up their time.
00:39:59.940 | Anybody who's a founder is, for the most part, super smart, super driven, really gritty,
00:40:07.740 | right?
00:40:08.740 | And their time is incredibly valuable.
00:40:11.460 | And if they're spending time on something that isn't worth that time, then you are doing
00:40:17.200 | them harm by not letting them know that.
00:40:19.380 | So they could switch and turn their attention to something better, something more worth
00:40:24.580 | their time and their brainpower.
00:40:27.620 | And then the other thing is that it's a real disservice to their employees.
00:40:31.500 | Because the people who are working at a startup, as you know, Chris, are generally working
00:40:36.480 | for very low cash comp, but lots of equity.
00:40:39.700 | So once you determine that the equity isn't worth their time, you ought to free them up.
00:40:45.260 | And what's interesting is that founders will often say, "I owe it to my employees to keep
00:40:50.860 | going."
00:40:51.860 | And it's like, no, but once you've decided that the equity isn't worth their time, you
00:40:56.040 | owe it to them to let them go so that they can go spend their time on something that
00:40:59.940 | might actually be world-changing, where the equity might really matter, right?
00:41:04.180 | Because there's only two reasons that people are willing to do this, because they want
00:41:07.620 | to change the world or they think the equity is going to be worth a lot, or both.
00:41:12.140 | So once you've figured out that none of those things are true, let them go.
00:41:15.220 | So when the VC is trying to spare their feelings, look at all the harm that's coming from that,
00:41:21.380 | right?
00:41:22.620 | You're allowing someone to be a dead company walking for however long until, you know,
00:41:28.140 | I guess they can't raise another round.
00:41:31.500 | That's months and months and months and months, sometimes years, that you're trapping people
00:41:36.340 | in a losing endeavor when you see something else.
00:41:39.260 | And so when you're looking for a quitting coach, this has to do with like that, first
00:41:42.860 | of all, that permission giving.
00:41:44.180 | You have to be clear with the person that you want to hear the truth.
00:41:47.540 | And then you need to go find someone who really has your long-term best interest at heart,
00:41:52.340 | is willing to say the hard things to you because they love you.
00:41:55.820 | - Yeah, I have a few people like that.
00:41:58.220 | For some decisions and then for others, it's a little tough.
00:42:02.780 | I briefly want to go back in time to talk about base rates for a second, 'cause I just
00:42:08.740 | thought as I was-
00:42:09.740 | - Okay, 'cause I love base rates.
00:42:10.740 | It's like my favorite topic, for real.
00:42:12.220 | - Yeah, so when I think about base rates, it reminds me of kind of the whole premise
00:42:16.020 | of thinking in bets, which is that, and really brought to life, the idea of asking someone
00:42:21.860 | to want to bet, right?
00:42:22.860 | Like the fact that we think of these beliefs, we have these ideas that we think are so certain
00:42:27.580 | and we're so confident, I'm gonna hire someone, they're gonna be the best, is really just
00:42:32.020 | thinking about everything in life as a bet and challenging yourself to put certainties
00:42:36.660 | around things, even mundane, silly decisions, kind of the best way to rewire how this all
00:42:42.260 | works in your head?
00:42:43.260 | - Oh, I think so, absolutely.
00:42:45.340 | So here's the thing.
00:42:46.340 | When I say, "Do you want to bet?"
00:42:48.260 | All I'm doing is making you examine the assumptions that are going into your statement.
00:42:53.340 | - Yeah, one for my wife and I is always just like remembering something.
00:42:56.380 | I'm like, "Oh, I'm totally sure that our daughter's preschool is starting next week."
00:43:03.140 | And like-
00:43:04.140 | - Right, do you want to bet?
00:43:05.140 | - My wife didn't ask that, but if she had, I would've been like, "Oh, I don't know if
00:43:07.380 | I want to put all my money on the line."
00:43:10.020 | - Right, because what it makes you think about is like, there's a couple things.
00:43:13.620 | Well, what is the person who's betting me know that I don't know?
00:43:16.900 | So in some sense, when I say, "Want to bet?"
00:43:18.460 | I'm acting a little bit as a coach to you, because I'm telling you that I have a perspective
00:43:22.900 | that is different from yours.
00:43:24.100 | If I'm willing to bet, that means that I'm saying that I believe something different
00:43:27.900 | than you.
00:43:28.900 | Okay, so it's automatically making you start to think like, "Well, what's a different way
00:43:33.260 | to look at this?"
00:43:36.320 | And then you have to start examining like, "Well, exactly how sure am I of this?"
00:43:42.220 | So this is, my husband and I talk like this all the time, where generally where it is,
00:43:50.020 | it's like, "Well, what's the price?"
00:43:52.060 | That's what we end up doing, which is just a way to say, "What's the probability?"
00:43:54.980 | So we don't think about bets as just like dollar for dollar.
00:44:01.620 | He might say that, let's say that it's an election, and you say, "I live in Pennsylvania."
00:44:11.460 | So let's say you say like, "I'm sure John Fetterman's going to win the election."
00:44:15.260 | And someone says, "Want to bet?"
00:44:18.740 | You could immediately say, "Yes," but then the response would be, "Well, what's the price?
00:44:22.940 | What are you willing to, are you willing to lay me a price?
00:44:25.460 | Are you giving me a price?"
00:44:26.880 | So in other words, as an example, if I say to you, "I'll bet $10 for every dollar that
00:44:32.460 | you bet," I'm telling you that I'm very certain that I'm right, because I'm risking a lot
00:44:36.860 | compared to what I can win.
00:44:38.740 | But if I say, "No, no, no, no, no."
00:44:41.440 | So the implied thing, it's like if I say I'm 90% sure of something, I should be willing
00:44:47.040 | to lay you nine to one, $9 for every one, right?
00:44:51.020 | But as soon as I'm like, "What?"
00:44:53.580 | When you say, "Okay, I'll take the nine to one," and I go, "Wait, no, no, no, no, no,"
00:44:59.500 | like even money, then what I'm saying actually is actually I think it's 50/50, and I wasn't
00:45:05.260 | really thinking about it.
00:45:06.660 | So it's like a good way to kind of be like interacting.
00:45:12.060 | And I can tell you that I do this with little things and big things and whatever.
00:45:16.100 | It's actually kind of fun once you start like, "Well, what price do you want to lay?"
00:45:20.460 | And we always bet $1.
00:45:21.460 | It's really fun.
00:45:22.460 | But I think that the thing that's really important about the betting, this is so crucial to good
00:45:32.360 | decision-making, is that it forces you to make the things that are implicit in the decision
00:45:39.260 | you're making explicit, okay?
00:45:42.920 | So as an example, the simplest example is one of the things that is always implicit
00:45:49.840 | in the decision you're making is your forecast of the probability of different things occurring.
00:45:56.840 | In other words, whether you know it or not, base rates are part of the decision-making.
00:46:04.400 | So when I say to you, "I'm going to meet you for dinner at seven o'clock, and I promise
00:46:12.680 | I'll be on time," and you say, "Want a bet," implied in that is that I must have a tendency
00:46:19.620 | to be late.
00:46:21.520 | And the question is, how much of a tendency do I have to be late, and how willing am I
00:46:25.000 | to say to you, "Yes, I'll bet you that I'm going to be on time," right?
00:46:30.120 | So when I say things like that, and then you bet back, those are all things that have to
00:46:34.680 | do with base rates, right?
00:46:38.200 | You could say something like, "I know the stock market is going to go up this year."
00:46:42.200 | Well, that's a prediction of the future.
00:46:44.400 | And if I know the base rates, I may bet you on that, depending on what you're willing
00:46:48.280 | to lay me, right?
00:46:49.600 | Because the stock market goes up actually somewhere between, like, it depends on how
00:46:52.800 | you measure it, 65% and 75% of the years, it will go up.
00:46:57.400 | So if you say, "I'm 100% sure it's going to go up this year," I might bet you.
00:47:01.160 | Because now what I'm doing-
00:47:02.160 | I would take that bet, because there are actually options out there I could use to arbitrage
00:47:07.640 | your confidence.
00:47:09.520 | That is true.
00:47:10.960 | That is very true.
00:47:13.400 | But I'm not involved in the arbitrage, so you could hedge away.
00:47:16.680 | I'm just going to take my bet, and then maybe I'll hedge it as well.
00:47:21.760 | But the point being that when somebody says, "I want to bet," it's like when you say, "I
00:47:27.360 | think this person I'm going to hire is going to work out 90% of the time," and I bet you,
00:47:31.640 | it makes you actually go through and start thinking, "Well, wait a minute.
00:47:34.080 | Why do I believe that?"
00:47:36.960 | Remember one of the things that we want to do in order to improve decision-making is
00:47:40.880 | to start thinking about what's the quality of the beliefs that we have that go into any
00:47:44.560 | decision that we're making.
00:47:46.840 | And part of that is making the implicit explicit.
00:47:50.680 | And this is one of the things that's so incredibly important.
00:47:54.760 | Even talking about a hiring example, where you send out to a bunch of people, "What do
00:47:58.880 | you think the job description should be?
00:48:00.960 | What do you think the person," if you were describing them to a recruiter, and you're
00:48:05.680 | getting that all independently, what you're finding out is, in an explicit way, when they
00:48:13.400 | would otherwise just go and interview somebody, what are the things that they think are important
00:48:17.880 | for filling this role?
00:48:20.800 | And now you've made that explicit in a way that you can, first of all, examine it, which
00:48:24.360 | is incredibly important.
00:48:26.680 | We have to be able to examine our decisions.
00:48:30.480 | And that means that we have to actually make all this stuff that's in our gut explicit
00:48:36.160 | in a way where you can examine it like an object.
00:48:39.220 | You can walk around it.
00:48:40.400 | You can have discussions about it.
00:48:43.320 | I can see that you have a different perspective, and when we see that we have a dispersion
00:48:47.560 | of opinion, we can dive into that dispersion to see where the gaps are, to see why you
00:48:53.200 | believe something different about the world than I do.
00:48:56.200 | And this is the way that we start to learn really fast.
00:48:59.760 | It's the way that we start to get our decisions to be much more accurate.
00:49:03.640 | But there's a side effect of this, of this making things explicit, which is that it allows
00:49:09.600 | us to close feedback loops a lot better.
00:49:12.560 | So now, when we do actually start to get some sort of set of outcomes, when we see an outcome,
00:49:17.280 | we can actually now go back and say, what did I believe at the time when I made this
00:49:22.800 | decision?
00:49:24.280 | And we can understand, here were my beliefs at the time.
00:49:28.040 | Here's what I found out after the fact, right?
00:49:32.160 | Let me go back and figure out, did I already have these outcomes in my set of possibilities?
00:49:39.960 | Do I feel like I had the probabilities right that those were going to occur?
00:49:44.240 | Was there stuff that I learned after the fact that I didn't know beforehand?
00:49:47.960 | Could I have known it beforehand?
00:49:50.160 | Could I know it going forward?
00:49:52.520 | So as an example, if I'm just thinking about when should I leave for work, right?
00:50:01.520 | And I wouldn't write this down because it's a more mundane decision, but I'm actually
00:50:07.880 | explicitly thinking about when should I leave for work?
00:50:10.320 | How long do I think it's going to take for me to be here?
00:50:12.600 | How bad is it if I'm late, right?
00:50:15.800 | And I decide in an explicit way, it's not very bad if I'm late today.
00:50:20.600 | I don't really have any important meetings for the first hour and a half of the morning.
00:50:25.460 | So I've got a little bit of leeway.
00:50:27.440 | So I'm really going to leave according to what the average time it is for me to get
00:50:32.120 | to work.
00:50:33.480 | So now I go and there's some huge accident on the highway and I ended up being 45 minutes
00:50:40.400 | late to work, okay?
00:50:42.560 | So do we look back on that and say that that was a mistake, which is what we have a tendency
00:50:48.000 | to do, right?
00:50:49.000 | Do we look back and say that was so stupid, why did I leave so late?
00:50:52.320 | Well, in this particular case, it wouldn't make sense to do that because you took it
00:50:55.400 | into account, right?
00:50:57.920 | You sort of knew there's some possibility of bad things happening on the road.
00:51:01.560 | You don't really have control over whether there's an accident on the road or not, and
00:51:05.880 | that you had a lot of leeway because it wasn't a big deal if you were late that morning.
00:51:10.200 | So you took all of that into account and then a bad thing happened, but that doesn't mean
00:51:13.700 | that the decision was bad to take that risk on, right?
00:51:18.120 | Whereas if I'm deciding to leave for the airport, I may say, well, I can't miss my flight.
00:51:24.220 | This is the only flight that they have today to the location I'm going.
00:51:27.920 | So now you may leave a lot of time for just that type of thing, like bad traffic or an
00:51:33.440 | accident.
00:51:35.520 | So what are you sacrificing for that?
00:51:37.520 | Well, you may end up spending an extra hour hanging out in the airport, but you make sure
00:51:41.360 | that you get to your plane on time.
00:51:43.880 | Now just as in the reverse case, when there is no traffic and the roads are totally clear
00:51:50.160 | and you get to the airport an hour early and you find yourself sitting around bored in
00:51:55.560 | an airport lounge, you also shouldn't think it was a mistake because in an explicit way,
00:52:01.240 | you thought about what your risk tolerance was.
00:52:03.360 | You thought about the base rates, how long it could take you to get to work, and you
00:52:08.640 | took that all into account when you made the decision.
00:52:11.960 | And this is what we need ultimately in order to be good decision makers.
00:52:15.760 | If you think about in advance the likelihood of the outcome happening, that makes it much
00:52:20.200 | easier to reflect on it.
00:52:22.360 | This happened to me in a situation where we were trying to, I won't go into the details
00:52:27.680 | of it because I could go down a rabbit hole, but we were trying to do something tax efficiently
00:52:32.360 | when we were buying a home and it involved taking some money out of the market.
00:52:35.280 | And I was like, "Okay, we're going to have to take a good chunk of our investments out
00:52:38.840 | of the market for about 60 days."
00:52:42.860 | And data showed about what the return on average is for 60 days.
00:52:47.220 | And so we knew about what we thought we would lose out in those 60 days, and we knew the
00:52:53.840 | upside of going through this entire process to buy the house in this specific way.
00:52:58.800 | And the only X factor was that in those 60 days was our last presidential election.
00:53:03.960 | So we were like, "Oh, we're a little anxious about this because there's this thing that
00:53:07.320 | could really move the markets."
00:53:10.000 | And so at the end of the day, I think time will tell if we stay in this house for more
00:53:14.880 | than seven years, it was a good decision.
00:53:17.720 | But if we don't, I still think it was a good decision.
00:53:20.920 | And even the wording I just used is probably wrong.
00:53:23.440 | If we stay in this house for seven years, it will have been the optimal outcome.
00:53:28.000 | But no matter what, it was the right decision.
00:53:30.600 | And I just have to keep harping on myself when people say, "Oh, is that a bummer?
00:53:34.200 | You might have lost out."
00:53:35.200 | It's like, "Well, I made the best decision with the information I had at the time, and
00:53:39.360 | I can feel good about that as hard as it is."
00:53:41.720 | That's still...
00:53:42.720 | I don't know if you have any tips for getting over that emotional angst of like, it's really
00:53:46.640 | easy to know you made the right decision.
00:53:48.320 | But if the right decision made you lose money, it's still hard to deal with that.
00:53:52.480 | Yeah.
00:53:53.480 | And I mean, the other thing is I assume that you took that into account, right?
00:53:56.400 | There are base rates for how long people stay in their homes as well, right?
00:53:59.980 | So you can say like, "In the worst case scenario, this is how long we would have to be in the
00:54:04.680 | house.
00:54:05.680 | What do we think the probability is that we'd be in the house this long?
00:54:07.880 | In the best case scenario, this is how long we..."
00:54:09.880 | So you take all of that into account, right?
00:54:13.560 | One of the things that I do, again, is to always change the frame.
00:54:16.560 | Well, how would I feel if it had worked out?
00:54:18.680 | Yeah.
00:54:19.680 | I guess in the last presidential election, there's other factors, but just from a financial
00:54:23.880 | standpoint, how would I feel if it worked out?
00:54:28.600 | So one of the things that I always try to think about is, "Oh, but those things go my
00:54:34.320 | way a lot too."
00:54:35.320 | Right?
00:54:36.320 | So I think I'll take a really simple example where I play tennis a lot, and a ball hits
00:54:46.040 | the net court a lot.
00:54:48.480 | And sometimes it plops over to the other side of the court, and sometimes it plops over
00:54:53.640 | to your side of the court.
00:54:55.440 | And this can be on very big points that this happens.
00:54:58.960 | And when it plops back over to your side of the court, you're super sad, and you feel
00:55:03.160 | like you're really unlucky.
00:55:05.200 | But then what I say to myself is, "But it's gone the other way for me at really important
00:55:09.560 | times as well."
00:55:10.560 | So I try to always remember that.
00:55:13.680 | And the question then I try to focus on is, "Did I pick the right shot?
00:55:19.840 | Should I have done something which was going to have more clearance over the net?"
00:55:24.580 | So then I start to think about, "Was it the right choice?"
00:55:28.400 | I try to get back to the decision itself as opposed to just the outcome.
00:55:32.920 | And we do this all the time.
00:55:34.980 | You're talking about a pretty high stakes decision that has to do with financial planning
00:55:38.800 | in-house, but you're taking equal risk by taking it out of the market.
00:55:43.880 | And the thing is that you weren't taking it out of the market because you were guessing
00:55:48.480 | out the outcome of the presidential election.
00:55:50.720 | Now, that I would say that's not so smart, right?
00:55:54.160 | We just know that from the base rates, that that's not trying to time the market doesn't
00:55:58.040 | really work out for people very much.
00:56:00.180 | But you're taking it out for a different reason where you understand what the risks are, that
00:56:04.120 | sometimes it's going to go up, sometimes it's going to go down.
00:56:07.400 | You understand how long you would have to stay in your house in order for it to be a
00:56:12.480 | reasonable decision, so on and so forth.
00:56:16.680 | And you decided that you were willing to take on that risk, I don't have any issues with
00:56:20.920 | that.
00:56:21.920 | Now, that's not to say that you wouldn't close the feedback loop and go back and say, "Would
00:56:25.320 | it have made more sense for us to wait until the presidential election had played out?"
00:56:34.640 | And the answer to that might've been, "But this is the house that we really want.
00:56:38.400 | We've been looking for a long time, and it's kind of go time.
00:56:42.480 | We can't wait."
00:56:43.480 | Okay, right, none of that sounds bad.
00:56:48.760 | The only thing that would be bad is that I had lots and lots and lots of time to wait,
00:56:53.920 | and I didn't need to do this at a moment of high uncertainty.
00:56:57.680 | Then I would say, "Well, maybe we should talk about that issue."
00:57:01.080 | But outside of that, you've thought it through.
00:57:04.360 | But this tendency, when you said, "Then it will have been a mistake, been a bad decision,"
00:57:10.640 | it's so strong with us.
00:57:12.040 | Let me ask you this.
00:57:13.480 | You've gone to a restaurant, and you're looking at the menu, and you're trying to figure out
00:57:20.960 | what to order.
00:57:21.960 | Let's say it's a restaurant you've never been to.
00:57:24.960 | You're trying to figure out what to order.
00:57:26.520 | This is really high uncertainty, like you've never been to this restaurant.
00:57:30.080 | It's got good Yelp reviews.
00:57:32.880 | You're trying to figure out what to order, and you narrow it down to a couple of dishes.
00:57:37.540 | Then at some point, you've just got to order.
00:57:42.440 | You do, and you get the food, and it's disgusting.
00:57:46.760 | It's yucky.
00:57:47.760 | What do you immediately say to yourself, "Oh, I made such a big mistake, I should have ordered
00:57:52.000 | something else"?
00:57:53.000 | What do you mean, you should have ordered something else?
00:57:56.560 | You know what your preferences are, you know what you like.
00:58:00.680 | What I try to tell people is, "Look, you know what you know, and then there's a lot of luck.
00:58:07.260 | You know what you know, there's a whole bunch of stuff you don't know, and then there's
00:58:09.560 | a lot of luck."
00:58:10.560 | Even with something like ordering off a menu, I say, "Just look, here's the deal.
00:58:16.340 | You should just divide the menu up into stuff you like and stuff you don't like."
00:58:21.760 | It's easy for me.
00:58:22.760 | I'm a vegan.
00:58:23.760 | It's like, "What's vegan?
00:58:24.760 | What's not vegan?"
00:58:25.760 | I can divide it up that way, and then I could do further things, like, "What has eggplant
00:58:29.840 | in it?"
00:58:30.840 | Then I get it narrowed down to whatever is available for me to eat off the menu, and
00:58:36.720 | then honestly, you should flip a coin, like literally, just, "Okay, here's the things
00:58:43.220 | that are in my choice set.
00:58:46.000 | I'm just going to flip a coin to try to figure it out."
00:58:48.680 | Now, if you're not comfortable doing that, I'll do this sometimes.
00:58:51.720 | I'll just say to the waiter, "You decide."
00:58:54.600 | It's the same thing as flipping a coin, like, "Oh, here are these two things.
00:59:00.200 | Surprise me."
00:59:01.200 | Yeah.
00:59:02.200 | I did an episode with Patrick McGinnis, who's kind of notable for coining the term FOMO,
00:59:06.120 | and he just looks at his watch.
00:59:07.440 | He comes up with two things.
00:59:08.440 | He says, "If the second hand's on the left, it's this.
00:59:10.720 | If it's on the right, it's this."
00:59:11.880 | It's his version of flipping a coin.
00:59:13.600 | Or you can just tell, like I say, you can tell the wait person to just decide for you,
00:59:20.440 | and then you get what you get, and it's fine.
00:59:22.640 | But that strategy, so I call it the menu strategy, is actually a way to make your whole life
00:59:27.880 | much more efficient.
00:59:28.880 | It's like you're trying to decide what to watch on TV, like figure out what your choice
00:59:32.960 | set is and flip a coin.
00:59:35.680 | The key to doing that is that when you don't like the thing you're watching, quit it.
00:59:43.720 | This is the thing about this type of strategy.
00:59:45.640 | It's like you have to be willing to abandon it.
00:59:49.280 | The problem is that when we're approaching a decision like, "Should we start this series?"
00:59:55.580 | We think we have to finish the series.
00:59:58.480 | We think we have to finish an episode.
01:00:01.520 | We think all of these weird things about it, and so then we get really caught up in trying
01:00:05.400 | to do research, and we're like, "Whatever."
01:00:07.920 | I just look at Rotten Tomatoes and look at the things that are high rated, or things
01:00:12.240 | that my friends have recommended, and then I just do it, and if I don't like it, I stop.
01:00:18.080 | But you have to be able to do the thing on the back end in order to do it, but you can
01:00:21.120 | do this with everything.
01:00:23.280 | There's two houses that you love, and you can't decide between them.
01:00:26.480 | Okay, if you can't decide between them, by definition, that means that your forecast
01:00:30.520 | of the probability of you loving either one and being happy in them is equal.
01:00:35.000 | So flip a coin.
01:00:37.360 | You can really do it with big decisions.
01:00:39.520 | The difference between a big decision and a small decision is that the bar for sorting
01:00:47.080 | out options you would consider versus options you wouldn't is higher.
01:00:52.920 | For what you'd order off a menu, it's not that big a deal.
01:00:55.440 | You eat three meals a day, I assume, whatever.
01:01:00.820 | It's a very low bar for getting to a coin flip.
01:01:03.720 | For buying a house, it's a pretty high bar for getting to a coin flip.
01:01:07.840 | That's all that you need to think about is what's the threshold.
01:01:11.680 | I love this idea, because I have an example I've mentioned in the past of we were going
01:01:15.140 | to Greece, and we found three hotels, and they were all pretty similar, and we were
01:01:19.560 | struggling.
01:01:20.560 | We were like, "Well, does this room look nice in this room?"
01:01:23.360 | Yeah, and it turns into months.
01:01:25.760 | The easy answer was once you've gotten to a point that you're pretty sure all of them
01:01:29.680 | would be fine, just pick one and move on.
01:01:32.680 | But it brings up the bigger question, which is you've written a couple books on decision
01:01:35.760 | making, and then you decided that this one decision of quitting and walking away was
01:01:41.120 | different enough, I assume, that you wrote an entire another book on it that just came
01:01:45.040 | out.
01:01:46.040 | I'm curious what made quitting such a hard decision.
01:01:49.640 | What characterized it so differently that it was worth the follow-up to dig into?
01:01:55.040 | Basically, here's the deal.
01:01:59.480 | What allows us to make decisions under uncertainty is that we have the option to quit.
01:02:04.720 | Just basic.
01:02:07.280 | Imagine if when you hired someone, you could never fire them.
01:02:12.560 | How hard of a decision that would be.
01:02:15.480 | Imagine if as soon as you went on a date with somebody, you had to marry them.
01:02:21.740 | How hard would it be to choose somebody to go on a date with?
01:02:26.320 | Imagine if you rented an apartment, you couldn't get out of the lease, like ever.
01:02:31.940 | These things would become very difficult.
01:02:35.520 | Decision making under uncertainty is really hard, but what allows us to be able to do
01:02:39.680 | it, to be able to choose things when we know so little, is that when we find out new stuff,
01:02:45.960 | whether it's about the world or it's about our own preferences, "Oh, it turns out I don't
01:02:50.480 | like this neighborhood," whatever, that we can quit.
01:02:54.720 | It's a really, really, really important decision.
01:02:58.260 | Having this option to quit is incredibly valuable, but what's interesting is that what the science
01:03:03.600 | tells us is that we're really bad at it.
01:03:08.480 | There's a lot.
01:03:09.480 | Obviously, grit is a really, really popular concept.
01:03:12.720 | People think about grit as character building, in fact, as almost synonymous with character.
01:03:20.560 | It's true that grit's really important for getting you to stick to hard things that are
01:03:24.360 | worthwhile.
01:03:25.360 | That's incredibly important.
01:03:27.000 | We have to not quit things just because they're hard, otherwise we won't ever be successful
01:03:31.400 | at anything because anything worth doing is going to be hard at some point.
01:03:36.440 | That's for sure true.
01:03:38.120 | I think everybody should read Grit, and it's a great book, and Angela Duckworth is brilliant,
01:03:43.160 | but I was missing the other side of the conversation, which is that grit also gets you to stick
01:03:48.640 | to hard things that are not worthwhile.
01:03:52.160 | That's a really huge problem.
01:03:54.760 | We can't think about grit as a virtue and quitting as a vice or a character flaw, because
01:03:59.920 | I can show you lots of circumstances where grit is a character flaw, where sticking to
01:04:05.460 | something too long is bad, like Muhammad Ali boxing for years after he already had symptoms
01:04:12.200 | of Parkinson.
01:04:15.200 | When he was really suffering physically, he should have walked away from the game.
01:04:22.400 | We know that it's very context dependent.
01:04:26.360 | Sometimes quitting is the best thing that you can ever do.
01:04:29.160 | If you're in a job with a toxic boss, or you're in a toxic relationship, or you're at the
01:04:34.360 | top of Everest, and it's past the turnaround time, and there's a snowstorm coming in, clearly
01:04:39.560 | sticking to it is dumb in that situation.
01:04:42.800 | You need to realize that they're the same decision.
01:04:46.600 | One is not good, and one is not bad.
01:04:48.480 | It's very dependent on context.
01:04:50.120 | Then, when you start to look and dive into the scientific literature, it turns out that
01:04:55.920 | there's just a tremendous cognitive forces that make it very hard for us to walk away
01:05:03.740 | from things, not at the right time anyway, all the way from sunk costs, which we've talked
01:05:08.920 | about a little, which is this tendency to stick to things because you're worried you'll
01:05:12.560 | have wasted the time, or the money, or whatever that you put into it, because you want to
01:05:16.820 | get that back somehow, like when you hold on to a stock, or you stay in a job, or whatever.
01:05:23.360 | There's issues that have to do with endowment ownership over things.
01:05:26.480 | We value the things that we own more than the things that we don't, including our beliefs,
01:05:32.160 | by the way.
01:05:33.600 | We don't like to let go of those things that have to do with something called sure loss
01:05:38.520 | aversion, which Daniel Kahneman talks about, which is, as we just said, you have a loss
01:05:44.000 | on the books.
01:05:45.120 | If you quit, it turns it ... I'm sorry, you have a loss on paper rather.
01:05:49.000 | If you quit, it turns into a realized loss.
01:05:53.160 | We don't like that.
01:05:54.680 | That doesn't feel good to us.
01:05:58.320 | As Richard Thaler says, we don't like to close mental accounts in the losses.
01:06:03.800 | We don't like to close those accounts and eat that loss, even though, obviously, just
01:06:11.960 | like a stock portfolio, our decisions are a portfolio, our opportunities are a portfolio.
01:06:18.120 | We ought to be trying to spend our time on the ones that are going to move us toward
01:06:21.680 | our goals the most, that are going to cause us to gain the most ground.
01:06:26.000 | We open an account for a particular thing we're doing, whether it's a marathon, or climbing
01:06:31.320 | Everest, or a job, or a relationship, or whatever.
01:06:34.920 | We don't think about that time, that effort, that all of that is fungible across all the
01:06:39.880 | possible opportunities we have.
01:06:41.840 | Instead, we just think, "If I quit now, I won't have hit my goal, and then I'm going
01:06:45.580 | to be a failure, and I'm going to have lost, and so we won't quit."
01:06:50.200 | There's identity, cognitive dissonance, status quo bias.
01:06:54.320 | These are all in the book, omission/commission bias.
01:06:57.560 | It's like there's so much stuff that we just carry around with us, this cognitive debris
01:07:03.080 | that makes it really hard for us to walk away.
01:07:06.560 | Here's the thing, why I'm so passionate about this topic, is that we think that quitting
01:07:12.440 | will slow our progress down.
01:07:15.680 | We think it may actually cause us to lose ground, but that's not true.
01:07:22.880 | When you quit at the right time, it actually gets you to where you want to go faster, because
01:07:28.640 | it stops you from sticking to something that isn't actually causing you to get there, to
01:07:33.880 | gain ground toward your goal, and allows you to switch to something that will allow you
01:07:39.480 | to gain ground toward your goal.
01:07:42.440 | That's the thing that we miss, because we don't see the other opportunities that the
01:07:46.320 | thing we're doing is stopping us from engaging in, because there's a cost to that time.
01:07:53.320 | There's a cost to that money that we're spending on whatever it is that we're into, or that
01:07:58.320 | we're pursuing, the path that we're on, and we won't let go of it, because if we haven't
01:08:03.760 | reached our goal, we think we're a failure.
01:08:07.440 | Just like if you sell that stock at 30 and you take the $20, whatever, you lost the $20,
01:08:12.560 | so what?
01:08:13.800 | You can put that $30 into other investments that are actually going to get you to financial
01:08:17.640 | well-being more quickly.
01:08:20.680 | Holding on to a stock that isn't worthwhile, because you already lost $20 in it, is actually
01:08:26.360 | slowing your progress down, because you can't go put those resources into something that
01:08:30.400 | will get you to financial well-being more quickly.
01:08:33.160 | That's true whether it's a stock, or a job, or a relationship, or a race you're running.
01:08:37.040 | I don't really care.
01:08:39.120 | If the world tells you you should quit, you ought to do that, because that's the thing
01:08:43.600 | that allows us to decide under uncertainty in the first place.
01:08:46.680 | Let's start exercising that option, so that we can more efficiently and more quickly get
01:08:51.440 | to where we want to go.
01:08:52.600 | I love it.
01:08:53.600 | There's so much more about quitting the book.
01:08:55.440 | I'm going to rapid-fire ask you a few things to see if you could just overview quick one
01:08:59.760 | or two minutes.
01:09:00.760 | Okay.
01:09:01.760 | I love the Astro Teller example of the monkey and the pedestal, and can you just walk people
01:09:08.380 | through that?
01:09:09.500 | We want to be efficient in the things that we're pursuing.
01:09:13.160 | We want to get to understand, should we continue, or should we quit as quickly as possible?
01:09:18.360 | Here's a mental model that will help people do that.
01:09:20.440 | It's called Monkeys and Pedestals, from Astro Teller, as you said, over at X at Google.
01:09:25.840 | Imagine you're trying to train a monkey to juggle flaming torches while standing on a
01:09:29.160 | pedestal in a town square.
01:09:31.160 | This will be a money-making opportunity.
01:09:35.160 | What's the thing that you should do first?
01:09:36.680 | Should you build the pedestal first, or should you try to train the monkey to juggle the
01:09:39.820 | flaming torches?
01:09:40.820 | Well, the thing about the pedestal building is that you already know you can do it.
01:09:45.340 | We've been building pedestals forever.
01:09:47.920 | It doesn't really help you.
01:09:49.780 | It's low-hanging fruit, but it's not something that's actually going to help you to accomplish
01:09:55.860 | your goal, because you could just turn a milk crate upside down even.
01:09:59.700 | If you spend time building the pedestal, it's really kind of wasted time, because it's not
01:10:04.100 | unlocking the bottleneck in the system, which is, "Can I actually train this monkey to juggle
01:10:08.040 | the flaming torches?"
01:10:10.180 | His point is that when you approach this project, you should say, "What are the monkeys?"
01:10:14.980 | Ask yourself this.
01:10:15.980 | What are the monkeys?
01:10:16.980 | What are the pedestals?
01:10:17.980 | What are the things I know I can accomplish?
01:10:18.980 | What are the things that are going to unlock the whole system, but are really hard, and
01:10:22.500 | I don't know if we can do it?
01:10:24.820 | Attack those things you don't know first, and that's going to do two really important
01:10:28.820 | things for you.
01:10:29.820 | One, it's going to get you to understand whether you can actually do this thing or not really
01:10:34.340 | quickly.
01:10:35.340 | Two, and this is really important, is it's going to reduce those sunk costs, because
01:10:42.060 | all that time you spend building pedestals is time and effort and money that you're putting
01:10:47.700 | into something that's then going to cause friction when you butt up against the monkey
01:10:52.460 | that's hard to train.
01:10:54.760 | Train the monkey first.
01:10:56.760 | This is the opposite of the way that most people do things.
01:10:59.420 | How many times have people approached a project and said, "Well, what's the low-hanging fruit?
01:11:03.900 | Let's try to solve for that."
01:11:05.740 | The problem with the low-hanging fruit is that those are all pedestals.
01:11:08.980 | Don't do the low-hanging fruit until you know whether you can actually grow the whole tree.
01:11:12.340 | Can you get to the fruit at the top of the tree?
01:11:14.780 | That's what you really care about, so I love that mental model.
01:11:17.260 | Yeah.
01:11:18.260 | I'm going to link in the show notes.
01:11:19.260 | A friend of mine, Aaron Battalion, wrote this post about knowns and unknowns.
01:11:23.860 | He pushed me so hard in my last startup.
01:11:25.580 | He was like, "Figure out what that known unknown is.
01:11:28.300 | What's the thing that you don't know how to do that you know you don't know how to do?"
01:11:32.860 | In his case, ultimately, the thing that he pushed me to focus on that, sorry, Aaron,
01:11:37.700 | I didn't follow at the beginning, was the thing that we realized halfway through running
01:11:41.400 | the company was the problem.
01:11:43.260 | We ended up shutting the company down with $4 million in the bank because we realized
01:11:49.860 | the thing wasn't going to work and we didn't want to keep going.
01:11:52.900 | You probably could have done that with $5 million in the bank.
01:11:57.340 | This is what AstroTeller is trying to say is like, "Look, you're going to spend some
01:12:01.700 | time trying to solve for the monkey or figure out if you can solve the monkey or monkeys."
01:12:07.100 | You might spend $2 million doing that, but if you can get to that answer in $2 million
01:12:12.420 | instead of $9 million, again, that's a savings of $7 million, not a waste of $2 million.
01:12:18.560 | That's why this is such an important mental model for people to use.
01:12:21.700 | There were a bunch of other great mental models on all kinds of different aspects of quitting.
01:12:26.820 | Where can people find out more about the book and everything else you're working on?
01:12:30.220 | People can always go to annieduke.com.
01:12:33.060 | I have a newsletter.
01:12:34.060 | Lately, it's been a lot about quitting, not surprisingly.
01:12:38.220 | I'm super into the topic, as you can imagine, or you can tell.
01:12:43.460 | And then Twitter @annieduke.
01:12:46.020 | And then you can obviously buy the book at any online or physical retailer near you.
01:12:52.140 | I got an advanced copy.
01:12:53.380 | I appreciate that.
01:12:54.380 | I really enjoyed it.
01:12:56.040 | Sometimes I don't get through the whole book before I talk to someone.
01:12:58.220 | This time, I was staying up late.
01:13:00.180 | My wife's like, "We got to wake up our kids in the morning."
01:13:02.140 | So I really enjoyed it.
01:13:03.840 | Thank you so much for writing it.
01:13:05.100 | And thanks so much for being here.
01:13:07.060 | Thank you so much for having me.