A Champion Poker Player on How to Make Better Decisions
Most people don’t understand how to be good decision-makers, but they could get better, says Annie Duke, a decision strategist and retired professional poker champion.
Duke won more than $4 million in tournament poker before retiring from the game in 2012. She is the co-founder of The Alliance for Decision Education, a non-profit whose mission is to improve lives by empowering students through decision-skills education. She recently was interviewed in a live YouTube session by Don Moore, who studies overconfidence and is a professor at the Haas School of Business at the University of California, Berkeley.
The talk was wide ranging and covered decision-making on the COVID-19 pandemic, writing books, investing and playing poker. But no matter what the challenge, she said, the common thread in making good decisions is thinking probabilistically. That means putting aside biases, accepting evidence, embracing uncertainty and weighing the range of likely outcomes before deciding on a course of action. The result is not guaranteed, but you will have made the best decision possible based on the information that you have.
That is easier said than done and often is not what happens when people make decisions. Instead, people sometimes use their beliefs, which are tied to their identities, to guide their decisions and infect others, Duke said.
In fact, she sees parallels between people refusing to wear masks, in spite of evidence that they slow the spread of COVID-19, and the rejection of evidence brought to light nearly 200 years ago by Dr. Ignaz Semmelweis, who got fired for telling doctors to wash their hands to stop spreading germs. She devotes a chapter to his story in her new book, How to Decide: Simple Tools for Making Better Choices, to be released in September.
Semmelweis had been trying to figure out why so many mothers of newborns were dying of childbed fever. Another colleague contracted childbed fever himself after performing an autopsy, and Semmelweis hypothesized that washing hands between patients would reduce the number of childbed fever deaths. Semmelweis instituted a policy of hand washing, and the percentage of mothers dying from the fever dropped from 16% to 2%.
“You would think, OK, that's it. Yay, he solves that problem. But no, the doctors got very upset about this when the hand washing turned out to work,” Duke said, “because what they felt was that the implication was that doctors’ hands were dirty, and doctors were not dirty people; they were gentlemen. They rejected the evidence right before their eyes and in fact fired him.”
Duke noted with the coronavirus pandemic, knowledge is imperfect, there’s a lot of uncertainty and the information landscape is shifting very rapidly. For example, whether you get the virus or not and the timing of when a vaccine is developed involves a lot of luck. People find that the uncertainty is disconcerting, and they believe that when the virus goes away, they will be more secure and better able to predict the future again, Duke said. But people won’t be able to predict the future even after the pandemic. To make good decisions, people will still need to consider probabilities because that is how the world works – there is always a lot of uncertainty, she added.
Duke did point to examples of high-stakes impaired decision-making in the pandemic that have produced outcomes that could have been avoided. Because of the way the virus spreads, most of the valuable data will be lagging, Duke noted, but people look at what is happening in the moment and demand action. One of the most astonishing examples, she said, happened in Florida a couple of months ago, when people were demanding the state’s governor, Ron DeSantis, receive an apology for being criticized for his response to the coronavirus. At the time, Duke said, the outcomes were looking pretty good and his policies that encouraged the opening of businesses and activities were reinforced. But as virus cases have spiked in the state, his process for making decisions does not look good after all, she concluded.
Duke said people think they have a lot more control over their success than they really do; success is usually a combination of luck and skill. In deciding whether to write her previous book, Thinking in Bets: Making Smarter Decisions When You Don’t Have All the Facts, she considered that there were a wide range of outcomes on how well the book would sell. A lot of amazing books aren’t commercial successes. Sales might be juiced if someone influential happened to read the book and talk it up on their platform. Plus, she’d have to devote a couple of years to the process of writing it.
Duke ultimately decided to write the book because it was a way to learn and work through her own ideas, making the decision to write the book a good one, regardless of the commercial outcome. “I think that the framing of ‘I want to write this book for myself’ makes it a good decision,” Duke said. “But let's say that I wrote the book because I said I want to write a bestseller so that I can be a famous author. Even if my book ends up being a bestseller, and I end up being a famous author, that doesn't make that decision high quality. It's very hard for us to actually disconnect those things in our heads, and that's one of the ways that our experiences can interfere with learning. The simplest example is the trope, ‘I drive better when I'm drunk’ because I got home safely. But that is actually a poor decision.”
Duke stressed that making a good decision does not mean necessarily choosing the option that has the greatest chance of success. In fact, sometimes it is better to make a choice that has a small chance of succeeding. The potential payoff justifies the decision, if you can tolerate the downside risk and afford the loss that could occur.
She used the early days of Amazon as an example. “It was the Jeff Bezos point: ‘If you're gonna invest in me, are you OK if you never see this money again?’”
In cases like that, Duke said, a small percentage of the time you will win and the gain is a small amount of money, and an even smaller percentage of the time you might make so much money that you don't know what to do with it, which is what ended up happening for Bezos and some early Amazon investors.
Duke added one of the best ways to innovate is to change your frame of mind in thinking about decisions. “We tend to approach the world thinking that when we make choices, it will be the only choice that we ever get to make, like ‘I've got to spend a month researching whether I should take piano lessons because once I do, I could never stop. You ask everybody, ‘Do you think I should? So you think I'd like piano?’ You are trying to get some sort of certainty about whether you would like it. That's actually the thing that stops you from having innovative and spectacular things happening because you're unwilling to engage in something until you're certain that it's going to be OK, until you're certain that it's going to work out.”
That approach is very counterproductive to good decision-making, she said. It’s better to recognize when you're in a situation where it's very easy to revert either back to an option that you've rejected or to a new option, and that leads to good decision-making because it allows you to make lots of little experimental choices, she said.
Most human activities and environments in which we are making decisions are similar to playing poker, Duke maintained. Decisions and outcomes are not correlated well because we have incomplete information. “If I lose a hand of poker, I'm not exactly sure what that means about my decision-making comparisons. I can make a perfectly good decision and because of a turn of a bad card I can certainly lose, or I can make terrible, horrible awful decisions and still win,” Duke said.
Good poker players try to assess the quality of information they have, independent of the actual outcome, to decide whether to raise their hand. That kind of thinking can be applied outside of poker, but it’s not easy, it’s more aspirational, Duke explained.
“I do believe that, while you're never going to get there 100%, small gains create big results in the long run,” Duke said. “So if you can improve the way that you're thinking about your outcomes, and what the quality of the feedback is that you're creating for yourself, then it's going to compound over time. That's the way that you can actually get better results in your life.”
Dorothy Hinchcliff is the editor at Advisor Perspectives.