thinking in bets by annie duke (summary)

in

We went into 2021 at Animalz determined to create a 5 year plan. I jumped into research, but didn’t get too far into reading articles before questioning my approach. Would a better way be to talk to people in my network to hear how they think about long term planning, what has/hasn’t worked and build a reading list off of their hard won recommendations?

One of the people I reached out to was my friend Daniel Zarick, co-founder of customer success tool Arrows.to. When we met, he talked about driving the business forward by making a series of bets. Him and his co-founder worked in sprints on each bet, leaning into what was working every 2ish weeks and comfortably letting go of anything that didn’t work on a similar timeframe.

With each bet, they paused to consider possible positive and negative outcomes, building reflection into their process and validating or tweaking their thinking at the end of each sprint. The approach made sense applied to an agile, early stage company, could it work for us?

To learn more, he recommended that I read Thinking in Bets by Annie Duke. Notes organized into:

About the author, Annie Duke

  • graduated from Columbia University with degrees in English and psychology

  • got her master’s at UPenn did doctoral work in cognitive psychology

  • brother played in the World Series of Poker and brought her into the game

  • Annie played poker for 20 years and won over $4 million in tournaments

  • Spoke to a group of traders at a hedge fund about the transferable skills between poker and securities trading which helped her begin to bring together what she learned from her studies and poker about decision making

Why read this book

Read this book to improve your objectivity and decision making skills! While I found the first few chapters repetitive (which is why I summarized them all in a single block below) there were several immediately applicable concepts for my work and life that will make me a better thinker both in doing and facilitating work.

Notable one liners and advice from Thinking in Bets

  • “The quality of our lives is the sum of decision quality plus luck.”

  • Get comfortable with not knowing and “I’m not sure”

  • “Thoroughly conscious ignorance is the prelude to every real advance in science.” - James Clerk Maxwell

  • In poker, a good player will fold about 80% of the time. That means spending 4/5 games watching and learning in a very low risk way. 

  • Through watching, you start to see other ways to play beyond the cookie cutter how to list you start from as a novice

  • “Many people say I’m the best women’s soccer player in the world. I don’t think so. And because of that, someday I just might be.” - Mia Hamm

Life is poker, not chess

Thinking in bets helps you make decisions “objectively, accurately and open mindedly.” Our lives are uncertain, and we should make decisions with that always in mind. Thinking in bets helps you get comfortable with the unknown and to acknowledge all that you don’t know, promoting curiosity and helping you move towards objectivity by exploring uncertainties versus relying on existing beliefs. While a more experienced person may be better and recommending the appropriate strategy or tactic for a given situation, external forces will always play a role in the outcome of decisions. Great decisions focus less on right vs wrong and more so on navigating through all the possibilities in between, knowing that we are always better for or against ourselves.

All decisions we make are bets. Our bets are driven by our beliefs, which we should be continuously refining. Often we’ll accept a belief quickly without questioning it which can lead to really poor decision making. In the book, annie says we think we form beliefs by:

  • hearing something

  • thinking about it to determine whether it is true or false

  • forming a belief

But she corrects us to say that in reality we typically:

  • hear something

  • believe it be true

  • maybe/sometimes refine that belief later

Great decision making comes from being open minded, always seeking out new information and being willing to be unsure or wrong. A good decision or set of decisions may not create a good outcome, and vice versa. Often, people focus on resulting (the outcome) instead of the decision and the way a decision is made. The more we can recognize flaws or successes in our decision making, the greater our learning opportunities.

Useful definitions:

  • Bet: a decision about an uncertain future. When we make a bet, we make a choice for a specific future self and not for other potential future selves.

    • Merriam Webster’s definitions:

      • “a choice made by thinking about what will probably happen”

      • “to risk losing something when you try to do or achieve something”

      • “to make decisions based on the belief that something will happen or is true”

  • Game theory: “the study of mathematical models of conflict and cooperation between intelligent rational decision-makers”

  • Learning: when you get lots of feedback tied closely in time to decisions and actions. Eg poker: bet, win/lose very quickly

  • Experience: “it’s not what happens to a man, it is what a man does with what happens to him.” - Aldous Huxley

  • Betting on a future: belief + bet = set of outcomes. Should be a continuous loop.

  • Motivated reasoning: using emotionally biased reasoning to make justifications/desired decisions.

  • Self serving bias: taking credit for the good stuff, making excuses for the bad stuff. Make sure your cue, routine, reward loop includes reflecting on successes and failures…even when you succeed to continue to improve. Self critique is an essential part of improving and overcoming the self serving bias. Phil Mickelson and Phil Ivey are both really good at this.

  • Fielding outcomes: “any outcome that is the result of our decision making is in the skill category. If an outcome occurs because of a force beyond your control, luck is the main influence.” When we field outcomes, we’re deciding which bucket an outcome belongs in, luck or skill. We often put our own bad outcomes in the luck bucket and good outcomes in the skill bucket but do the opposite for those around us.

  • Decision quality: how (well) is/was a decision made?

  • Outcome quality: the result of your decisions. A bad outcome doesn’t necessarily mean that your decision was bad. Good decisions + luck drive our quality of life.

The buddy system

It’s really hard to think in bets alone because we all have blind spots. Annie got better at playing poker by recapping her games with her brother and other expert pokers where she was forced to not complain about luck but instead talk about the quality of her decisions. She would often do this without mention of outcomes, so they could break down her success by strength/weakness of decisions alone. She’d focused specifically on places where she thinks she may have messed up or experienced confusion while playing. This allowed Annie to become really good at focusing on her internal locus of control vs. all of the things outside of her control.

Not everyone is going to be down to seek out truth with you. And that’s ok! Different groups play different roles in your lives. But if you’re trying to overcome ignorance or bias, an accountability group will make it a lot easier to see the flaws in your thinking. This made me think about:

  • Sustained Dialogue, a group at UVA designed to bring together a variety of views to create space for people to understand other sides and work through unconscious bias

  • Stepping into Discomfort, a group led by Caitryn Megan focused on challenging each of us to do the work to understand our own privilege and perspective and read and listen to other perspectives to learn and grow.

AA came to be because founder Bill W realized he couldn’t beat alcoholism on his own. The group created accountability, reminding him and the millions that have followed in his footsteps that he would have to answer to others in his group for his decision to stay on or fall off the wagon.

A group helps overcome “our tendency to confirm what we already believe.” Well, usually. Sometimes, a group can become an echochamber. This has been proven amongst federal judges of similar political leanings and even research communities, which is why it’s helpful to “focus on accuracy over confirmation.” Remembering that we can’t make a decision or seek truth without hearing multiple perspectives will help you escape the echo chamber.

Annie shared a study on a betting market that gets scientists to bet “on the likelihood of experimental results replicating” and found that they were more accurate when they had money on the line. Winning a bet incentivizes people to be focused on the right things: truth vs getting along with other people.

Dissent to win

Great decision making requires commitment to truth seeking. Truth seeking is possible with an openness to a diversity of thought that brings together multiple viewpoints and encourages discussion. When we are held accountable by a group of people with different opinions who can see our blind spots, it’s less likely that we’ll incorrectly attribute bad decision making to bad luck or allow ourselves to slip into self serving bias. You can encourage openness by:

  • Expressing uncertainty

  • Lead with agreement using “and” instead of “but” language to force open mindedness and calibrate beliefs

  • Ask to engage in truth seeking vs focusing on emotions/luck (not always possible, we all just need to vent sometimes). This allows us to focus on internal locus of control.

CUDOS

Communism: share data openly with everyone in the group.

  • “Secrecy is the antithesis of this norm; full and open communication its enactment.” - Robert Merton

  • Share what might make your data/viewpoint invalid

  • Rashomon effect: beware that people will have different accounts of the same event based on different perspective and facts.

  • Experts share data to move us towards accuracy.

  • Detail drives decision quality

    • Eg Vince Lombardi spent 8 hours unpacking a single play to a room full of pro football players, waking everyone in the room (inc hall of fame coach John Madden) up to what “pro” and “expert” really means

  • Goal is to provide as much accurate, detailed, objective evaluation of fact vs opinion

Universalism: use frameworks consistently to analyze data to limit bias.

  • Don’t dismiss an idea just because you don’t like who/where it came from

  • When we’re closed (I don’t/do like this idea/message), we miss opportunities to be curious and to learn.

  • Instead of jumping to conclusions, look for something good when you hear dissenting views/opinions

  • Groups can make it more possible to create necessary open mindedness

Disinterestedness: maintaining objectivity and staying wary of the potential for conflict of interest.

  • We naturally try to confirm our beliefs rather than question them, to not admit what we don’t know or when we’ve made a mistake.

    • Eg when New England Journal of Medicine published research from 3 Harvard scientists saying “fat is bad” that opened the door to a whole other suite of health problems for America (and the world). That study was sponsored by the sugar industry so of course it was biased.

  • If reflecting on a strategic decision, it’s better to focus on everything other than the outcome. Knowing the outcome biases everyone trying to analyze the decision. We should also leave out our own opinion when possible. Both sway the audience.

Organized skepticism: creating a safe space to encourage dissenting views to be heard.

  • Looking at the world “asking why things might not be true” vs assuming that they are as presented

  • We should “lean over backwards” (Richard Feynman) trying to see how we could be wrong.

  • Many companies have a feedback loop to encourage anonymous feedback knowing it will make them stronger if they can surface employee innovations, ideas, alt strategies and points of view.

    • Works as long as the rules of engagement are clear and people can’t be nasty, destructive or dismissive.

Adventures in mental time travel

We can create accountability by evaluating decisions with the help of both past and future versions of ourselves.

  • Looking ahead to explain decisions to other people in our truth seeking group (friends, team members, family, bosses, etc)

  • In poker, you know that every decision has consequences. Great players are able to bring their longterm strategic goals into their short term decision making. Eg capping at a certain level of loss, knowing they’ll have to explain decisions to an accountability group,

  • Most decisions don’t have immediate consequences. If we consistently make the same types of decisions based on unchecked beliefs we will feel the weight of consequences over time.

    • Eg eating candy vs an apple, not brushing your teeth, procrastinating, rejecting new ideas (leading to a dearth of new ideas.

We tend to favor our present self over future self. For better dec decision making, consider your future self.

    • eg when we stay up too late ignoring how our morning selves will feel

    • Eg when children have the option of a treat now vs 2 treats later

    • Eg taking lump sum retirements at a discount (savings for retirement plan provider) vs annuity plans

Move regret in front of decisions. Create regret before vs after the fact. 

  • How would you feel today if you’d made this decision 10 minutes/months/years ago?

    • eg How will you feel if you retire and don’t have sufficient savings?

    • eg How will you feel if you actually *aren’t* sober enough to drive

    • Bring that regret forward for better perspective.

  • Stop being a “ticker watcher of your own life”

    • Eg Berkshire Hathway stock looks great zoomed out, not so great zoomed in to individual days of the 2008 financial crisis

    • Being too zoomed in makes us reactive, focused on avoiding negative emotion or sustaining positive emotion. Can sustain detrimental self serving bias.

Be aware of tilt

    • Often, where we end up matters less than how we got there. We tend to focus on the most recent changes and then react emotionally to those changes, losing perspective.

    • When we’re in reactive tilt mode were thinking with our amygdala instead of our higher, logical brain

    • Asking questions, breathing deeply, sleeping on it etc can help you overcome tilt. Asking people within a truth seeking pod can help too.

    • Take the long view.

Proactively anticipate what will take you out of your logical mind and create strategies to keep you grounded and objective

  • Know the triggers that lead you to temptation and create decision interrupters (eg if you’re on a diet, don’t keep your favorite chocolate in the house)

  • Create a “Ulysses contract” eg automatically diverting money into a 401k or 529, forcing you to think with your longterm self before your long term self

  • Use a “decision swear jar” that you pay into when you:

  • Say you’re 100% sure about something without checking your beliefs

  • Act overly confident

  • Are overly focused on the role luck plays in outcomes

  • Unproductively criticize others/their ideas

  • Zoom too far in on a moment

  • Get stuck in an echo chamber

  • Become overly self critical

  • Bias listeners

  • Stop others from sharing opinions

Reconnaissance

  • Use second order thinking to understand risks and prepare for as many potential outcomes as possible

  • Belief -> bet -> multiple potential future selves factoring in probability for scenario planning

  • Scenario planning within a truth seeking group becomes very powerful when you have a wide range of view points represented.

  • Don’t be surprised by an ever changing future.

  • This can be used in a wider range of scenarios

    • Eg sales

Work backwards vs forwards

  • What’s the goal? Work backwards from there. Combining possible positive and negative futures helps you move towards reality and improves the likelihood that we achieve our goals.

  • Pre-mortem (we do this with our new customers at Animalz!) https://hbr.org/2007/09/performing-a-project-premortem

    • Pre-mortems encourage organized skepticism

  • Backcasting from “we achieved our goal!”

    • What events occurred?

    • What decisions were made?

    • What went right?

      • Do low probability events need to occur to get there? If so, goal should be adjusted.

      • Eg losing weight, work backwards

  • Backcast from “we failed to achieve our goal” with a pre-mortem, encouraging everyone to look for ways that things might go bad. “Ok we failed, why did we fail?”

    • Delays in new product

    • Exec turnover

    • Competition

    • Sales/growth slow

Treat time as a tree

  • Trunk is past

  • Branches are possible futures

  • Branch -> trunk = present

  • Present can be a chainsaw eliminating other possible futures. Can lead to hindsight bias also known as creeping determinism “I knew it all along” or “whatever happened was bound to happen”

  • Stop being an “outcome junkie:”

  • Maintain optionality

Key questions from Thinking in Bets

  • How likely is that to occur?

    • Keep in mind all the information that you don’t know and the blindspots you have when making a decision.

  • How sure/unsure are you?

    • Determine based on available info and experience making bets

    • It’s rare that we are 0% or 100% certain, usually we are somewhere in between.

    • When we admit that we don’t know or aren’t sure we have the opportunity to challenge and improve our beliefs to become more accurate.

  • Wanna bet?

    • See degrees of uncertainty

  • Why did we (fail to) achieve our goals?

    • Review all of the small decisions and key results

  • How did luck factor in? How did skill factor in?

    • It can be really hard, even impossible, to work backwards from outcomes.

  • Why might my belief not be true?

  • What other evidence might be out there bearing on my belief?

  • Are there similar areas I can look toward to gauge whether similar beliefs to mine are true?

  • What sources of information could I have missed or minimized while trying to reach my belief?

  • What are the reasons someone else could have a different belief, what’s their support, and why might they be right instead of me?

  • What other perspectives are out there as to why things turned out the way they did?

  • How would I feel about this if I heard it from a different source?

  • What are the consequences of each of my options in 10 minutes, 10 months, 10 years? - Suzy Welch