Hindsight is a wonderful thing, but the bias, not so much

In 1972, just prior to President Nixon’s historic trip to China and the USSR, researchers Fischhoff and Beyth asked a large group of students to estimate the probability of various events happening on the trip. For example, that Nixon would meet Mao.

A full six months after the trip was over, the students were then asked to recall their original estimates. What the researchers found, was that when an event did occur, 75% of the students’ recollected probabilities that were higher than their original estimates. And when an event didn’t occur, the students’ recollected probabilities that were lower than their original estimates.

The study showed that the students’ recollections of their estimates, gravitated towards the real outcomes of the trip. It identified for the first time, the hindsight bias, which is the tendency to believe that past events are more predictable than what they actually are.

Since this original study, further studies have shown that the hindsight bias is widespread and difficult to avoid. It occurs across age, genders, cultures and across a range of situations, such as with the weather, sporting victories, election results and investing.

Stock market bubbles

In the five years leading up to October 1987, the stock market had tripled. However, on 19 October 1987, the market dropped 22.6%, which at that time was the biggest one-day percentage fall in US history.

After the crash, Yale University economist Robert Shiller mailed a survey to 3,250 investors, to investigate how they navigated the crash. Responses showed that many investors believed they saw the crash coming. However, despite saying this, there was very little objective evidence in their trading activity to suggest they did.

Though the crash appeared very clear in hindsight, Shiller concluded that the investors had fooled themselves into believing they saw it coming.

Other bubbles such as the GFC were also the subject of extensive hindsight bias, with many market commentators saying how predictable the crash was, despite saying nothing about it before the crash.

Causes of Hindsight Bias

There are several theories as to what causes hindsight bias. One theory is that it is caused by a memory distortion whereby our memory of the actual true event distorts the memory of our predictions.

A second theory is that people are comforted by believing they live in a predictable world. A third theory is that it feels good to believe you were right, so it’s a subconscious instinct intended to promote a positive view of ourselves.

Why is the Hindsight Bias a problem?

The main reason the hindsight bias is a problem is that it leads to overconfidence.

If a person believes (incorrectly) that they predicted past events, it may lead them to think they can also predict future events. This can lead them to take on riskier activity, which can result in poor investment returns.

Given that we believe we’ve ‘known-it-all-along’, the hindsight bias can also stop us from digging deeper into a situation to find out what really happened, thus inhibiting our learning and growth. If we work hard to understand a situation, it will help us to make better decisions in the future.

Also, understanding that some events are difficult to predict can help us to appreciate the complexity of the systems that we work in. Events can often go in different directions, and realising this, can help us with our preparation and risk management.

How to cure the Hindsight Bias

There are a number of ways to fix the hindsight bias. Here are three:

1. Keep a diary. Write down in advance your predictions for the stock market, interest rates, inflation, the AFL premiership winner, an upcoming election, and the Ukraine war. In a few months, revisit your predictions to see how accurate you were. It may surprise you how difficult predicting is.
2. Review all potential outcomes of an event. Pick an event, then map out all of the ways the event could have worked out differently. If we consider only the reasons why something turned out the way it did, it’s likely we’ll overestimate its inevitability.
3. Remember your predictions. Research shows that if we intentionally try to recall our predictions accurately, it can actually help us to avoid the hindsight bias.