Site icon Ewing Marion Kauffman Foundation | Kauffman.org

Decisions, decisions: why we shouldn’t take this basic practice for granted

Annie Duke
Rebuild better

I first reached out to author, speaker and former poker player Annie Duke early this year, when we were exploring content for our “The Risk Optimistic” theme, which launched in January. Who better to talk about risk than a woman who’d found success in the high-stakes, male-dominated field of professional poker? As it turned out, Duke was heads-down, editing her new book “How to Decide” (which debuted this week), and then the pandemic hit, making risk seem like a less-palatable topic for a conversation.

When we did get a chance to talk, the world was very different, and the quality of the decisions we make on a day-to-day basis seemed especially poignant. When lives are at stake, and the future course of our country is in question, it adds a different dimension to the kind of decisions we personally make about our health and careers, not to mention on behalf of our families and communities. If we’re to rebuild systems in ways that are to the benefit of all Americans, some very well-considered, deliberate decisions will need to be made.

For Duke, it isn’t enough to talk about decision-making in theory, though. She feels we need practical tools to help us understand what gets in the way of making better decisions, but also that give us methods to break down the process, so that we can do it in a more rational, conscious way.

Here’s our conversation, edited for length and clarity:

Keith Mays: You’ve done so many things in your life. How do you describe yourself when you meet new people?

Annie Duke: I describe myself as an author and decision strategist. But, in terms of my journey, I guess it’s “lapsed academic, turned poker player, turned lapsed poker player, turned decision strategist.” And, avid tennis player.

Mays: You have this new book, “How to Decide.” You’ve been obviously thinking about decision-making for quite a while, and this builds on your previous book (“Thinking in Bets”), as well. Why were you compelled to write this particular book, and more generally, why do you think it’s important that people have tools for decision-making?

There are two things that determine how your life turns out…. One is luck and one is the quality of your decisions.

Duke: The reason that I write books is generally because I’m trying to understand my own thoughts about things better. There’s nothing better than the process of writing to help you sort of butt up against the limits of the things that sound so incredibly eloquent and rational in your head, and when you put them down on paper, you sort of realize “that’s really horrifying. Let me try to think about this better.”

In “Thinking in Bets,” I was trying to take this cut that had to do with really looking at a uncertainty, and examining that, and the way that that really interferes with your ability to learn from your environment, and make really good decisions and understand the feedback that the world is giving you. So, after I wrote that book and worked through those thoughts, I would say that the most common question that people would come up and ask me was, “Okay, but how do I do that? What would a good decision process look like?” And I said, “Oh, maybe I should do that.”

There are two things that determine how your life turns out. There’s only two. One is luck and one is the quality of your decisions. So, we can’t do anything about luck. You should certainly know it exists. But if something’s going to happen 2% of the time, it’s going to happen 2% of the time, and you don’t have any control over that.

What you do have some control over, though, is your decisions and the quality of your decisions.

What that means is that we want to be thinking about decision quality in a couple ways: one is we want to make sure that we can figure out ways to improve the accuracy of the decisions that we’re going to make through really creating a good process that’s going to produce more accurate decisions. Two, we want to make sure that process is repeatable. So, in the same sense that I can take a screwdriver, and I can reliably get a screw in the wall using that tool, and I can do it repeatedly, I want to make sure that I’ve got tools in my decision toolbox, so that I can know when I use this tool, or when I’m applying this decision process, that I can expect to have a particular type of decision or a certain quality of decision come out of that.

When you start to think about, for example, gut decision-making, that’s like the opposite of a good tool. I can’t really examine it, I can’t really check back in with it, because I don’t really know how it works and I certainly can’t hand it to you. I can’t say, “Here’s my gut. You go use that.”

Our brains are constructing our reality for us all the time, and it’s no different with decisions.

Mays: Let’s back up to how you got interested in decision-making. Was that from your academic work? Or, did that somehow come out of poker? I know you use poker as an analogy and there is a lot of literal high stakes decision-making in poker, but how did you come to thinking about decision-making in business terms?

Duke: It was a collision of a lot of weird things, is basically my answer. I was studying cognitive science and cognitive science is really understanding how our brains construct our reality. A tree is not green. There’s a material that reflects light that our brains process as green. We know that greenness is not a quality of the leaf itself because there are people who are colorblind who show us that.

Our brains are constructing our reality for us all the time, and it’s no different with decisions. There are things that are true of the world that if you were omniscient, and we were also fortune tellers I suppose, then we would know exactly what reality was – but we don’t.

And then, instead of going into academics, I took this left turn and started playing poker, which was really spurred by an illness that made me have to take a year off. It was during that year that I played poker, and that’s very fast-paced, high stakes decision-making, where there’s a tremendous amount of uncertainty, but the rules are very well defined.

I figured out pretty quickly that there were a lot of great things about that learning environment, which had to do with the rapidity of the feedback. I do something, and then two seconds later you raise me, or you fold, or you call it, or whatever.

Annie Duke playing poker.

And that poker was also actually a pretty unkind learning environment. I saw how that affected not just other people’s decision-making, but my own: the traps that I was falling into. I became really interested just in a practical sense, not in an academic sense, of “how do I sort of solve these problems and how do I think about two things, one is how can I make this better for myself?” And then, “if I really understand the places where people’s decision-making goes sideways, how can I use that to build better models with my opponents and figure out what are the strategic and tactical ways that I could take advantage of that?”

And then in 2002, another accident happened, which is a friend of mine got asked to speak to a hedge fund, talk to their traders about how poker might inform the way that you would think about risk. He knew that I taught classes, and I was used to getting up in front of large groups of people and speaking, and so he recommended me.

That was the first time, about eight years into my poker career, that I started thinking in a much more explicit way about the way that poker and cognitive science could inform each other.

Once that happened, I ended up getting recommended from that one talk to other ones, and so I developed a speaking business from that and then started getting asked to do consulting. In 2012, I retired from poker, and then started working on thinking, and sort of ended up back fully in academics. So now I do consulting, and I also do academic work in research in the space, which is really honestly my first love, anyway.

Mays: I would like to delve more into some of the specific topics in the book. One is about bias and decision-making. You’ve talked about how we personally self-sabotage, but can you talk more broadly about how the extreme polarization of views and beliefs in this country might impact our decisions?

Our beliefs, in the large sense, are our identity. Because what are we, but the things we believe? And I don’t just mean religious beliefs – I mean like the things we think are true: the facts that we believe, the knowledge that we have.

Duke: There are so many ways to think about this. It is very well-replicated and very well-documented in science, (that we believe) we interact with information in some kind of objective way. It turns out mostly that’s not the way it works, that what we engage in is “motivated reasoning.”

What happens is that we have beliefs and those beliefs really in the large sense are our identity. Because what are we, but the things that we believe? And I don’t mean just religious beliefs – I mean like the things we think are true: the facts that we believe, the knowledge that we have.

Think about any individual beliefs that you have as the thread in the fabric that’s your identity. The way that we reason about the world is to protect that fabric – we don’t want it to tear. We reason about the world in a way that is motivated to keep our identities intact, and what that means is that we have to keep our beliefs in. So, we’re essentially doing kind of two things at the same time. One is we’re really seeking out information sources and people who will certify the beliefs that we have. That’s confirmation bias.

The other thing is that if we do collide with corrective information, then we’ll swat it away. Meaning if I read a political article that agrees with me, I will read the headline and say, “That’s the smartest article I ever read.” And if I read something that disagrees with me I have a dissertation about why it’s biased and the data that they were using was they were cherry picking and data mining and they weren’t thinking about it from this angle and so on, and I can tell you all the reasons why that’s ridiculous, and you can see that we’re applying a different standard to those two things.

We get into this circular reasoning where our beliefs are blind in the way that we’re processing information with the purpose of reinforcing the beliefs that are already in the driver seat.

This becomes a problem because a lot of our reasoning isn’t so much about “what do I believe,” but, what group do I belong to? And “the other group is bad and evil, and I can’t believe the things that they say.” And then you start to get into this problem of not examining the information objectively, just using the proxy of “did it come from a Republican or Democrat?”

And we do that not just in politics, by the way. You can see that in science, you can see that in investing – because people tend to become group-ish in terms of investing – and so on, and so forth.

A lot of our reasoning isn’t so much about ‘what do I believe,’ but, what group do I belong to? And, ‘the other group is bad and evil, and I can’t believe the things that they say.’ And you start to get into this problem of not examining the information objectively, just using the proxy of, ‘did it come from a Republican or a Democrat?’

Mays: At the Kauffman Foundation, we do storytelling about entrepreneurs and students that highlights themes of persistence and succeeding by staying the course despite barriers. So, I’m interested in your idea of quitting as a tool to decide what you like and don’t like.

Duke: I hope you’re talking about people who succeed by quitting.

Mays: Well, maybe that’s the question. The idea of quitting seems like a counter-narrative to popular ideals, like perseverance and grit. Is what you are suggesting a contradiction of popular notions?

Duke: It is not a counter-narrative. It’s that people took that narrative too far. When we think about stick-to-itiveness or grit, the thing that I would say about that is, of course anything that you’re eventually going to succeed at, you had to have stuck to it. So, the example for poker is that if I stuck with every single hand, I would surely lose. I need to be cutting my loses, and I need to do it quickly at the point that I first recognize that I may have a negative expectancy.

There are a few values of quitting. One is that we can undo, we can unwind decisions that aren’t working out well, which is incredibly important to do. The idea is that you want to stick to things that are going to have a positive expected value and are going to advance you to your goals, but you want to quit things that aren’t. And the power to recognize the difference between those two is really important. So, most of what you’re going to be doing is quitting. Because particularly when you’re deciding under uncertainty, with very little information, you’re going to have a lot of false starts. And I wouldn’t even call them false starts, I’d say starts that then new information reveals itself, and you want to pull off of those as quickly as possible to shift course, to pick up maybe an option that you rejected in the past, or a new option that has presented itself to you that is going to be a better way.

So, why in the ’80s did IBM get so crushed by Microsoft and Apple? Because IBM was gritty in that sense. But, they weren’t agile, they weren’t thinking about what is working and what isn’t and what are the things where we want to just sort of cut our loses and say, “Yes, this is the way that we’ve done it in the past, but now we need to go find a new way to do it and we should stop doing that thing.”

No. 2 is if you don’t think about quitting in advance, it’s very hard to get a lot of information from the world.

When you recognize that your goal is to try stuff out, but then quit quickly when it doesn’t work, this allows you to get really, really experimental with your choices because you’re not thinking about of choice that you make as permanent anymore.

When you recognize that your goal is to try stuff out, but then quit quickly when it doesn’t work, this allows you to get really, really experimental with your choices because you’re not thinking about of choice that you make as permanent anymore.

I can’t get myself an answer which is 100%, about whether I’m going to like playing the trombone, if I’ve never done it. So, a lot of times you’ve got to try. And then you find out, “Okay, how’s it working out?” And if it doesn’t work out, you’ve gotten the information that’s going to make your future choices better.

Mays: One of my least favorite phrases – and, it’s been particularly overused recently – is, “In these uncertain times.” And, of course, the future is never certain, though it perhaps feels a little more unpredictable right now. I’d like to know what advice you’d give to parents who are putting their kids out into the world over the next few years. So, given a hypothetical high school freshman, I’m wondering if you could walk through how she and her parents might very intentionally work toward making decisions about her higher education, her career, that sort of thing with the presumption of ultimate success in life.

Duke: Separate apart from COVID or no?

Mays: Well, COVID obviously plays into that. It adds extra uncertainty on top of …

Duke: By the way, not much. I mean I think that’s one of the keys, is that we think that this decision-making environment is incredibly different, and it’s like extra. That’s it, extra.

So essentially, thing No. 1 is this: we want to be figuring out, how do we improve this problem that we all have as decision makers, which is that the stuff we know basically fits on the head of a pin, and the stuff we don’t know is like the size of the universe? We’re always making decisions with very, very limited information. What we want to do is increase the information that we have for making the decision, and some of that information, by the way, is just about “what are my own preferences,” and, “what are my goals? What am I really trying to accomplish here? What are my resources?” And then getting better and more educated about our ability to think about what the future might hold for us, which is essentially an exercise in forecasting.

The way to do that is to decision stack. To say, “I’ve got this big thing on the horizon. What are things that I do beforehand that will help me to gather information where I’m not taking on a lot of risk in the information-gathering endeavor?” So, we can think about some ways to do that. One of them is “quit-to-itiveness.” The easier something is to unwind and revert, the more of that thing we want to be doing. Because what it means is that we’re essentially “de-risking.” If things don’t go well, we can stop and go back, and try something else.

So, renting versus buying. Renting is much easier to unwind than buying is in terms of a house. Dating before you marry. Those are naturally occurring examples of decision stacking, but we can think about how to apply that to a variety of decisions. That’s one of the ideas behind internships, where you get to try a bunch of different careers out before you figure out what you want to do.

We want to think about how we can be getting all these kind of reversible decisions in advance. The other thing you want to think about is the long-term impact of the decisions that you make. You want to be prioritizing reversible decisions that have lower impact if you happen to get it wrong, and this information gathering to clarify for you what you want to do.

I can’t find anybody who was taught anything about how to make a decision in K-12. They might have taken it as an elective at some point in college, but we’re not teaching kids this.

Mays: At the Kauffman Foundation, we have a Real World Learning initiative, which is a way of rethinking the high school experience to emphasize giving students skills that will help them be successful in careers and to be self-sufficient individuals in life. That can be through teaching an entrepreneurial mindset, project-based learning, internships, and professional skills. What do you think about the balance between practical, pragmatic skills, and academics, and how important do you think it is for decision-making to be taught?

Duke: I would prioritize decision-making above almost anything else. You want to give people the opportunity to explore what they like and what they don’t like and taking a variety of different classes is really helpful in the endeavor. But, in terms of what you’re giving people as the core curriculum, I would really prefer that people are thinking about how this is actually going to impact the rest of somebody’s life. That’s actually why I co-founded a nonprofit called The Alliance for Decision Education, which is really about the ways that in K-12, we can start getting decision education in there. I can’t find anybody who was taught anything about how to make a decision in K-12. They might have taken it as an elective at some point in college, but we’re not teaching kids this.

I think it’s like in ninth grade or so everybody’s taking trigonometry. And there’s a lot of things that are not great about trigonometry as something that you’re teaching kids. Thing number one is that people are rarely raising barns or navigating across the seas without a computer anymore. And, it’s the type of math where there’s a right and wrong answer, which is not the type of math that I think is really useful for most of the type of decision-making that we have to do in our lives, where we’re dealing with things that are much more subjective, where outcomes aren’t guaranteed and you can’t take a look back and say yes or no. If I had my druthers, in ninth grade you would not be doing trigonometry, you would be doing statistics and probability.

Trigonometry does not make our democracy healthier. Statistics and probability actually make our democracy healthy, which is what I really care about. We have to have a society that manages to stitch itself together in a way that’s functioning and cohesive. Teaching people how to make decisions would be really good for that.

Trigonometry does not make our democracy healthier. Statistics and probability actually make our democracy healthy, which is what I really care about.

We have to have a society that manages to stitch itself together in a way that’s functioning and cohesive. Teaching people how to make decisions would be really good for that.

No. 2 is that I think that as a society we have very strange, for me, values about what we value as somebody’s skillset or knowledge set. Works that are trades – being an electrician, being a plumber – I feel that as a society we don’t put the same kind of value on those, as say, someone who’s a philosopher.

I think that’s really a shame, because I think that we’re telling people that your value comes from college. That you have to go to college in order to be a valuable member of society, and I think that we got that really wrong. And I think we’re sending people very bad messages about whether they value themselves, because I think people internalize this kind of thing, and so I would like to see that idea about how you become a valuable and productive member of society change.

People have different needs in terms of how quickly they’re starting to produce. If you have somebody who’s living in poverty, delaying their ability to produce, I think, does really big damage to them in terms of allowing them to bring themselves out of poverty. And, so, I love, as an option, the idea of apprenticeship to sort of bypass, so that you can start to produce wealth for yourself and the people around you and your family, which for a lot of people is really like an emergency.

I think that that kind of thing is amazing, and I’d love to see that moving into high school as well for people that choose that for themselves.

Mays: Which is all fine and good. But I think the hard part is, how do we change the mindset of that parent and that high school freshman, that not going to college could be a perfectly honorable path? Even knowing that could mean not having the burden of college debt, starting on a career and accruing wealth sooner, earlier than a college student, parents seem to be thinking, “That’s great, but not for my kid.” So how do we kind of make that mental shift in our society?

I think that we’re telling people that your value comes from college. That you have to go to college in order to be a valuable member of society, and I think that we got that really wrong.

Duke: I wish that I knew how to solve this. I think that decision education would be helpful, but I think that this goes back to earlier in this conversation and when we were talking about motivated reasoning and the idea that we like to certify our identities.

I think that in a broader society sense what ends up happening is that the people who are degreed, who are credentialed in that way, are the ones who end up with the levers of power.

I think in the ’40s or ’50s it was like 10% of the population had college degrees, and we had a perfectly good, functioning society. We were the wealthiest country in the world. And then there started to be this weird slippage where all a sudden, jobs that used to just need a high school diploma, with some apprenticeship or training or whatever, you all the sudden need a college degree.

I’m not discounting the network effects of going to college. Of course, those things are really important, but like all decisions, you need to balance those out with what your other considerations are. If I’ve got an 18-year-old who’s living in poverty, I want to get them out being productive as quickly as possible and start allowing them to make money. And I don’t want to tell that kid that you’re less than somebody who got to go to college right away.

Mays: I was thinking about the societal pressures that cause people to make certain life and career decisions. There are unspoken rules about what careers are appropriate because of your race, or because of your gender. For example, I was surprised to learn recently that on the list of the wealthiest Black Americans, almost all of them were in the entertainment or sports fields. On the gender side, the STEM fields can be unwelcoming to women.

I wanted to ask about your experience. My quick Google search said that only 3% of professional poker players are women. What kind of advice would you give to that girl who’d like to enter a male-dominated field, helping her visualize both the potential positive and negative outcomes, and how to deal with them?

Duke: In poker, that 3% number is so remarkable because when I first started playing it was 3%, and today it’s 3%. So, it really hasn’t changed. I think there’s a lot of reasons why that is, but it’s a hard place to be a woman, I’m not going to lie. Remember, poker has no HR department – you can’t make an HR complaint. There’s a lot of sexualizing that occurs at the table. A lot of anger toward you. It’s a highly emotional game, and people are losing amounts of money, that it’s tough on them.

And that emotion is going to get released somewhere and it’s not going to get released at the dude who’s going to go beat them up. There’s resentment. And then there’s just like that kind of subtle stuff that happens like, “Why don’t you smile more?” I’m not kidding.

I’d love to figure out how to change those environments and again, I don’t know how. Hopefully just by existing I was part of some change. I think that society is changing in terms of noticing the subtle ways in which sexism expresses itself a little bit more.

I’m not saying that society doesn’t have to let you in. There was a time when I wouldn’t have been allowed to sit. I was very lucky I had a seat at the table. It was unpleasant, but I had a seat there. And so, we have to create those seats for sure. But at least the first women that are going to take those seats are going to need to love what they do.


Rebuild better: Success that ensures the opportunity of prosperity for all can’t be achieved within broken systems; true success will only come from systems we rebuild better.

Exit mobile version