by Annie Duke
I learned from this experience that thinking in bets was easier if I had other people to help me. (Even Neo needed help to defeat the machines.) Remember the buddy system from school field trips or camp? Teachers or counselors would pair everybody up with a buddy. Our buddy was supposed to keep us from wandering off or getting into water too deep, and we did the same for our buddy. A good decision group is a grown-up version of the buddy system. To be sure, even with help, none of us will ever be able to perfectly overcome our natural biases in the way we process information; I certainly never have. But if we can find a few people to choose to form a truthseeking pod with us and help us do the hard work connected with it, it will move the needle—just a little bit, but with improvements that accumulate and compound over time. We will be more successful in fighting bias, seeing the world more objectively, and, as a result, we will make better decisions. Doing it on our own is just harder.
Members of our decision pod could be our friends, or members of our family, or an informal pod of coworkers, or an enterprise strategy group, or a professional organization where members can talk about their decision-making. Forming or joining a group where the focus is on thinking in bets means modifying the usual social contract. It means agreeing to be open-minded to those who disagree with us, giving credit where it’s due, and taking responsibility where it’s appropriate, even (and especially) when it makes us uncomfortable. That’s why, when we do it with others, we need to make it clear the social contract is being modified, or feelings will get hurt, defensiveness will rear its ugly head, and, just like Lauren Conrad, your audience won’t want to hear what you have to say. So, while we find some people to think in bets with us, with the rest of the world, it is generally better to observe the prevailing social contract and not go around saying, “Wanna bet?” willy-nilly. (That doesn’t mean we can’t ever engage in truthseeking outside of our group. Our approach just needs to be less head-on, less Letterman-like. More on that later, after we explore communications within the group.)
Out in the world, groups form all over the place because people recognize how others can help us; the concept of working together on our individual challenges is a familiar one. Having the help of others provides many decision-making benefits, but one of the most obvious is that other people can spot our errors better than we can. We can help others in our pod overcome their blind-spot bias and they can help us overcome the same.
Whatever the obstacles to recruiting people into a decision group (and this chapter points out several, along with strategies for overcoming them), it is worth it to get a buddy to watch your back—or your blind spot. The fortunate thing is that we need to find only a handful of people willing to do the exploratory thinking necessary for truthseeking. In fact, as long as there are three people in the group (two to disagree and one to referee*), the truthseeking group can be stable and productive.
It’s also helpful to recognize that people serve different purposes in our lives. Even if we place a high value on truthseeking, that doesn’t mean everyone in our lives has to adopt that or communicate with us in that way. Truthseeking isn’t a cult; we don’t have to cut off people who don’t share that commitment. Our Pilates friends or our football friends or any of our friends shouldn’t have to take the red pill to remain our friends. Different friends fill different needs and not all of them need to be cut from the same cloth. Those different groups can also provide much-needed balance in our lives. After all, it takes effort to acknowledge and explore our mistakes without feeling bad about ourselves, to forgo credit for a great result, and to realize, with an open mind, that not all our beliefs are true. Truthseeking flies in the face of a lot of comfortable behaviors; it’s hard work and we need breaks to replenish our willpower.
In fact, in my poker strategy group, we understood the need to occasionally opt out and off-load intense emotions before engaging in the work of accurately fielding an outcome. If, for example, one of us just got eliminated from a tournament, it was acceptable, every once in a while, to say, “For right now, I just need to moan about my bad luck.” The key was that when we did that, we recognized it was a temporary exception from the hard work we were doing together and to which we would return when the emotional rawness of the moment passed.
We know our decision-making can improve if we find other people to join us in truthseeking. And we know we need an agreement. What’s in that agreement? What are the features of a productive decision-making pod? The remainder of this chapter is devoted to offering answers to those questions. Chapter 5 builds on that by providing a blueprint for rules of engagement within truthseeking groups, how to keep the group from drifting off course, and the productive habits of mind the group can reinforce in each of us.
Not all groups are created equal
A well-chartered group can be particularly useful for habits that are difficult to break or change. This is not a crazy or even novel idea. We are all familiar with how the group approach can help with reshaping habits involving eating, consuming alcohol, and physical activity. The most well-known example of a productive group approach is Alcoholics Anonymous (AA).
The first of AA’s founders, Bill W., initially refrained from drinking through a difficult process that included years of failure, hopelessness, hospitalization, drugs, and a transformative religious experience. To maintain sobriety, however, he realized that he needed to talk to another alcoholic. Bill W. recruited Dr. Bob, the second founder of AA, on a trip to Akron, Ohio. Dr. Bob, considered by family and doctors to be a hopeless and incurable alcoholic, kept Bill W. from drinking on the trip. In turn, Bill W. eventually helped Dr. Bob give up drinking. AA has subsequently helped millions of people get and stay sober, and led to organizations trying the same approach with other difficult-to-tackle habits like narcotics abuse, smoking, unhealthy eating, and abusive relationships. That all sprang from the concept that we can do better with the help of others.
But, while a group can function to be better than the sum of the individuals, it doesn’t automatically turn out that way. Being in a group can improve our decision quality by exploring alternatives and recognizing where our thinking might be biased, but a group can also exacerbate our tendency to confirm what we already believe. Philip Tetlock and Jennifer Lerner, leaders in the science of group interaction, described the two kinds of group reasoning styles in an influential 2002 paper: “Whereas confirmatory thought involves a one-sided attempt to rationalize a particular point of view, exploratory thought involves even-handed consideration of alternative points of view.” In other words, confirmatory thought amplifies bias, promoting and encouraging motivated reasoning because its main purpose is justification. Confirmatory thought promotes a love and celebration of one’s own beliefs, distorting how the group processes information and works through decisions, the result of which can be groupthink. Exploratory thought, on the other hand, encourages an open-minded and objective consideration of alternative hypotheses and a tolerance of dissent to combat bias. Exploratory thought helps the members of a group reason toward a more accurate representation of the world.
Without an explicit charter for exploratory thought and accountability to that charter, our tendency when we interact with others follows our individual tendency, which is toward confirmation. The expression “echo chamber” instantly conjures up the image of what results from our natural drift toward confirmatory thought. That was the chorus I heard among some groups of players during breaks of poker tournaments. When one player brought up how unlucky they had gotten, another would nod in assent as a prelude to telling their own hard-luck story, which, in turn, would be nodded at and assented to by the group.
Lerner and Tetlock offer insight into what should be included in the group agreement to avoid confirmatory thought and promote exploratory thought. “Complex and open-minded thought is most likely to be activated when decision makers learn prior to forming any opinions that they will be accountable to an audience (a) whose views are u
nknown, (b) who is interested in accuracy, (c) who is reasonably well-informed, and (d) who has a legitimate reason for inquiring into the reasons behind participants’ judgments/choices.” Their 2002 paper was one of several they coauthored supporting the conclusion that groups can improve the thinking of individual decision-makers when the individuals are accountable to a group whose interest is in accuracy.
In addition to accountability and an interest in accuracy, the charter should also encourage and celebrate a diversity of perspectives to challenge biased thinking by individual members. Jonathan Haidt, a professor at New York University’s Stern School of Business, is a leading expert in exploring group thought in politics. Haidt, in his book The Righteous Mind: Why Good People Are Divided by Politics and Religion, built on Tetlock’s work, connecting it with the need for diversity. “If you put individuals together in the right way, such that some individuals can use their reasoning powers to disconfirm the claims of others, and all individuals feel some common bond or shared fate that allows them to interact civilly, you can create a group that ends up producing good reasoning as an emergent property of the social system. This is why it’s so important to have intellectual and ideological diversity within any group or institution whose goal is to find truth.”
In combination, the advice of these experts in group interaction adds up to a pretty good blueprint for a truthseeking charter:
A focus on accuracy (over confirmation), which includes rewarding truthseeking, objectivity, and open-mindedness within the group;
Accountability, for which members have advance notice; and
Openness to a diversity of ideas.
An agreement along these lines creates a common bond and shared fate among members, allowing the group to produce sound reasoning.
None of this should be surprising to anyone who recognizes the benefits of thinking in bets. We don’t win bets by being in love with our own ideas. We win bets by relentlessly striving to calibrate our beliefs and predictions about the future to more accurately represent the world. In the long run, the more objective person will win against the more biased person. In that way, betting is a form of accountability to accuracy. Calibration requires an open-minded consideration of diverse points of view and alternative hypotheses. Wrapping all that into your group’s charter makes a lot of sense.
The charter of the group must be communicated unambiguously, as Erik Seidel made clear to me. I had met Erik when I was a teenager, but when I started running into him at poker tournaments, it was the first time we were interacting in a business setting. Early on in my career, I saw Erik during a break in a tournament, and started moaning to him about my bad luck in losing a big hand. In three sentences, he laid out all the elements of a productive group charter. “I don’t want to hear it. I’m not trying to hurt your feelings, but if you have a question about a hand, you can ask me about strategy all day long. I just don’t think there’s much purpose in a poker story if the point is about something you had no control over, like bad luck.”
When you think about a charter for truthseeking interactions, Erik Seidel pretty much nailed it. He told me the rules of being in a pod with him. He discouraged me from confirmatory or biased thought like “I got unlucky.” He encouraged me to find things I might have control over and how to improve decisions about those. I knew he would hold me accountable to these things in future interactions. We would explore diverse ideas because he insisted that be the focus of our interactions.
Because I was lucky enough to be part of a group with a truthseeking charter, there was no question that my poker decision-making improved. When I could consult them on in-progress decisions, like whether to move up in stakes or bankroll management or game selection, their advice reduced the number of errors I was making. Likewise, access to their range of strategies and experiences improved the quality of my thinking and decisions on a continuing basis. When I had questions or didn’t understand why something happened, they would see things I didn’t. When they had questions or needed advice on a hand, I wasn’t just helping them work through a decision they made but would often get insights into my own game. Those interactions led to improvements in my game I would have overlooked or, at best, figured out on my own only after making a lot of costly errors.
Even better, interacting with similarly motivated people improves the ability to combat bias not just during direct interactions but when we are making and analyzing decisions on our own. The group gets into our head—in a good way—reshaping our decision habits.
The group rewards focus on accuracy
We all want to be thought well of, especially by people we respect. Lerner and Tetlock recognized that our craving for approval is incredibly strong and incentivizing. In most laboratory situations, they noted, study participants expected to explain their actions to someone they’d never met and never expected to meet again. “What is remarkable about this literature is that—despite the prevalence of these minimalist manipulations—participants still reliably respond as if audience approval matters.” It’s great to get approval from people we respect, but we crave approval so badly, we’ll still work to get it from a stranger. A productive decision group can harness this desire by rewarding accuracy and intellectual honesty with social approval.
Motivated reasoning and self-serving bias are two habits of mind that are deeply rooted in how our brains work. We have a huge investment in confirmatory thought, and we fall into these biases all the time without even knowing it. Confirmatory thought is hard to spot, hard to change, and, if we do try changing it, hard to self-reinforce. It is one thing to commit to rewarding ourselves for thinking in bets, but it is a lot easier if we get others to do the work of rewarding us.
Groups like AA demonstrate how a supportive group can provide the reward for doing the hard work of changing a habit routine, just by its approval. For engaging in the difficult work involved in sobriety, local AA groups give tokens or chips celebrating the length of individual members’ sobriety. The tokens (which members often carry or customize as jewelry) are a tangible reminder that others acknowledge you are accomplishing something difficult. There are chips for marking one to sixty-five years of sobriety. There are also chips given for every month of sobriety in the first year. There is even a chip given for being sober for twenty-four hours.
I experienced firsthand the power of a group’s approval to reshape individual thinking habits. I got my fix by trying to be the best credit-giver, the best mistake-admitter, and the best finder-of-mistakes-in-good-outcomes. The reward was their enthusiastic engagement and deep dives introducing me to the nuances of poker strategy. It was also rewarding to have these intelligent, successful players take my questions seriously and increasingly ask for my opinions. In contrast, I felt disapproval from them when I acted against the charter and complained about my bad luck, or expected them to confirm how great I played simply because I was winning.
While I never got close to attaining the goal of a pure focus on accuracy, my group helped me to give a little more credit than I otherwise would have, to spot a few more mistakes than I would have spotted on my own, to be more open-minded to strategic choices that I disagreed with. That moved me, even if just a little bit at a time, toward my goal of getting closer to the objective truth. And that little bit had a huge long-run impact on my success.
When I started playing poker, “discussing hands” consisted mostly of my complaining about bad luck when I lost. My brother quickly got sick of my moaning. He laid down the law and said I was only allowed to ask him about hands that I had won. If I wanted him to engage with me, I had to identify some point in those hands where I might have made a mistake.
Talking about winning (even if we are identifying mistakes along the way to a win) is less painful than talking about losing, allowing new habits to be more easily trained. Identifying mistakes in hands I won reinforced the separation between outcomes and decision quality. These discussions also
made me feel good about analyzing and questioning my decisions because of the approval I got from Howard and the players I looked up to. I used that approval as evidence that I understood the game and had promise as a player. When they complimented me for finding alternative approaches in my winning hands or understanding the contribution of luck, that felt terrific. In time, I could expand this approach to identifying learning opportunities in any hand I played, not just the winning ones.
Once we are in a group that regularly reinforces exploratory thought, the routine becomes reflexive, running on its own. Exploratory thought becomes a new habit of mind, the new routine, and one that is self-reinforced. In a Pavlovian way, after enough approval from the group for doing the hard work of thinking in bets, we get the same good feeling from focusing on accuracy on our own. We internalize the group’s approval, and, as a matter of habit, we begin to do the kind of things that would earn it when we are away from the group (which is, after all, most of the time).
“One Hundred White Castles . . . and a large chocolate shake”: how accountability improves decision-making
David Grey is a high-stakes poker player and professional gambler, and a good friend. After a night at a racetrack and a bowling alley in New Jersey, David and a bunch of other bettors were hungry. It was late. Someone suggested White Castle. A discussion broke out about how many burgers the biggest eater in the group, Ira the Whale, could eat.
When they got Ira the Whale to say he could eat 100 burgers (remember, White Castle burgers are small), most of the group, not surprisingly, wanted to bet against him. David was an exception. “I was a young guy, just getting started. Fifty dollars was a big win or loss for me. There was about $2,000 out against Ira the Whale. I bet $200 on him because I thought he could do it.”