Quit: The Power of Knowing When to Walk Away

Former professional poker player and author of Thinking in Bets, Annie Duke has another interesting book called Quit: The Power of Knowing When to Walk Away.  Through entertaining stories and discussion of scientific research, she makes a strong case that people aren’t good at deciding when to quit relationships, jobs, and many types of life goals.

In one interesting story, “Blockbuster, when presented with the opportunity to acquire Netflix, refused.”  We now know that Blockbuster would have been better off focusing on streaming and giving up on renting physical copies of movies.  Of course, if they had tried to do both, their own executives responsible for renting movies might have killed off streaming to protect their own jobs and bonuses.

Other interesting and tragic stories concern those who failed to give up on climbing Mount Everest when it became too dangerous, and the many people who have finished marathons on broken bones “Because there’s a finish line” they couldn’t abandon.

“If you feel like you’ve got a close call between quitting and persevering, it’s likely that quitting is the better choice.”  This is because people have a tendency to stick with failing paths too long.  So, by the time you think it might be time to quit, that time is likely long past.  “Quitting on time usually feels like quitting early, and the usually part is specifically when you’re in the losses [and still hoping to come out ahead by persevering].”

It’s hard to tell which scientific research studies of behavioural biases are likely to stand up to further testing and which won’t.  Apparently, NBA teams stick with high draft picks too long when they’re not working out.  This certainly seems plausible.  However, it’s not obvious that the reason is behavioural bias.  NBA teams don’t think with a single mind.  An exec who knows a player isn’t working out but is at risk of losing his job in response to fans’ wrath might appease the fans at the team’s expense.  We see CEOs with massive stock option grants make choices good for them but do long-term harm to their companies all the time.

One possible test of the plausibility of behavioural biases is to see if they are evolutionarily useful in some way.  For example, many people are uncomfortable flipping a fair coin to either lose $100 or win $200.  In the distant past when such gambles had higher stakes, it usually made more sense for a starving person to eat a meal in hand than to abandon that meal for a 50/50 chance to chase down 3 meals.  It’s plausible that this risk aversion is baked into our DNA in some way.  I find it easier to believe in the existence of a behavioural bias when it appears to be a useful genetic adaptation that is simply being misapplied in a modern context, such as the low-stakes coin flip bet with positive expectation.  When research studies find behavioural biases, I’d be interested to hear evidence that the bias had some usefulness for survival or procreation.  However, these are just my thoughts and not a failing of Duke’s book.

Daniel’s Kahneman’s endorsement of Quit is accurate: “Brilliant and entertaining.”  I can recall things I should have quit sooner.  This book might have helped.