We think we understand why things happened, and can predict what will happen. That overconfidence can get us into a lot of trouble. (More)

Thinking, Fast and Slow, Part III – Overconfidence (Non-Cynical Saturday)

This week and next, Morning Feature looks at Daniel Kahneman’s book Thinking, Fast and Slow. Thursday we saw the two ways we think, fast and slow. Yesterday we examined our mental shortcuts and biases. Today we explore overconfidence. Next Thursday we’ll consider how we make choices. Next Friday we’ll meet our two selves, experience and memory. Next Saturday we’ll conclude with how our improving model of human thinking should shape public policy.

Daniel Kahneman is a Professor Emeritus of psychology and public affairs at Princeton University’s Woodrow Wilson School of Public and International Affairs. In collaboration with Amos Tversky, Dr. Kahneman was a pioneer in the psychology of judgment, decision theory, and behavioral economics. He was awarded the 2002 Nobel Prize in Economics for his work in prospect theory, and has taught at universities in Israel and across the U.S.

“I told you so!”

We’ve all said that at some point in our lives, usually right after some variation of: “I shouldn’t say ‘I told you so, but’….” Something went wrong, and we knew it would happen. We remember saying so, but the person or people didn’t listen. And it’s not just us.

As Dr. Kahneman notes, many economists now insist they knew the economic collapse of 2008 was coming. Most are generously reinterpreting what they said or wrote. A handful did predict it, but Dr. Kahneman argues not even they could say “I knew it,” in the strictest sense of that term. Simply, the collapse had not yet happened and might not have. Leaders in Washington and Wall Street would have needed to be exceptionally brilliant or exceptionally lucky to prevent it, but brilliant insights and lucky escapes do happen. Events are almost never inevitable.

Yet we often see past events as inevitable, what Dr. Kahneman calls hindsight bias. We look at the past for causes in conditions, trends, and human decisions. Having found what feels like a good explanation, we tend to conclude that an event was certain to happen, given the factors we identified. But the key words in the previous sentence were “feels like.” If a story we’ve heard or developed is plausible and coherent, System 1 will believe it. Plausibility and coherence are subjective, and System 1 judge them by comfort rather than critical analysis. Once System 1 has chosen to believe, System 2 will more often defend than critique that belief.

Skill or Luck?

The result, Dr. Kahneman writes, is that we grossly overestimate the probability of past events. They happened, ergo they were inevitable. Studies show we also revise our memories, imagining ourselves as having predicted or at least raised the possibility of those events.

Having done that, we tend to be confident we can predict future events, at least if we are skilled in that field. Brokers can tell you which stocks will rise or fall. Guidance counselors can tell you which students will succeed or fail in college. And a military psychologist could tell the Israeli army which officer candidates would perform well in action. That psychologist was Dr. Kahneman, and he writes that he was very confident of his predictions … even after a statistical review found they were little better than random guesses.

The term illusion of skill describes situations where we apply difficult, complex processes to predict random or unknowable events. Because we’ve learned how to apply that process, we feel we must be more skilled than someone who has not, and that our predictions must be more reliable than guesses. We estimate the reliability of our predictions from inside factors like how well we understand and apply the ‘expert’ process. We tend to dismiss outside factors – randomness or unknowability inherent in the event itself – even when we know about them.

Add it all together and we are overconfident in our predictions, overestimating the role of skill in our wins, explaining away our losses. Worse, the more confident we are, the more others will look to us for advice and leadership. Experts don’t get invited to cable TV shows to say: “There’s no way to be sure.” We don’t rally behind people who say: “This is my best plan, but we’ll have to get at least a little lucky.”

So it’s all luck?

No. Skill does matter. In fact, Dr. Kahneman spent several years in “adversarial collaboration” with a colleague who was convinced that skilled intuition was reliable. (The humorous and revealing subtitle of the paper they eventually published was “A failure to disagree.”) They finally agreed on conditions where skill – expert use of a complex process, even intuitively – was very reliable.

Those conditions include minimal randomness, lots of practice, and clear and prompt feedback. Part of learning to drive is learning how to brake while turning. Random events can happen – a brake line could rupture, or you could hit an invisible patch of oil or ice – but they’re rare. In learning, you got lots of practice braking while you turned, and you got clear and prompt feedback: creeping around a corner when you braked too much, or whew! (or oops!) when you braked too little. You very quickly developed skill, and your System 1 intuition began deciding when and how much to brake. System 1 could also tell you that something felt wrong – with the vehicle or the situation – even if you could not (yet) identify what.

Those conditions also exist in other areas of life, but not as often as we suspect. More often, random or unknowable factors play a larger role, we don’t get as much practice as we think, or feedback on our decisions is vague and delayed. Did a brilliant plan work that year, or was the previous year so exceptionally awful that returning to average looked brilliant? Did sales rise across that industry? Did a competitor’s mistake push business to that company? For most non-trivial events, it’s difficult to identify and calculate the statistical salience of every factor.

Confidence wins

But if you devised the plan, you’re likely to be confident your skill made the difference, and to follow the same ‘expert’ process again. And if you seem confident enough, other people will often credit you with expertise. They may even ask you to lead them, convinced your skill and intuition are the stuff of which legends are made. If events work out well enough next time – again, without you or they or anyone else knowing all the factors in play and how much each mattered – you might even get a multimillion dollar compensation package and a book deal to write How to Be as Brilliant as Me (though the publisher may change the title).

Someday, someone might even invite you to host a major party’s presidential candidate’s debate. Heck, you might be one of those candidates. And maybe you should be.

Or maybe it’s overconfidence, and you’ve been less Expert than Lucky.


Happy Saturday!