Placing your bet: What overconfidence can teach us about decision-making

Much of the language that we use to describe decision-making comes from the world of sports betting and gambling. Phrases like "the stakes are high," "let's roll the dice," "play your cards right," and "all in" occur as frequently in boardrooms as in casinos.

In Decision Mojo™, a learning lab focused on decision-making, we encourage leaders to consider some decisions as placing a bet  in which probabilities are at play and the outcome or payoff is not guaranteed.

Whether placing a bet or making a decision, it's helpful to accurately assess the probability of success. Overconfidence can help us in some walks of life, but in the world of gambling it can lead to financial ruin. In the world of decision-making it can be equally dangerous. So, when a recent study from Stanford University revealed how accurately sports bettors predicted the future outcomes of their bets, it caught our attention.

Overconfidence at play

Turns out that people who place bets on sports tend to be overly optimistic about how much money they're going to win. In the study, the average sports bettor predicted they would break even but actually lost 7.5 cents for every dollar wagered. This research aligns with the overconfidence trap we explore in Decision Mojo™, which is the tendency to overestimate one’s knowledge, skill, or predictive accuracy.

One suggested antidote to this trap is forcing ourselves to look objectively at past results and seeing whether our confidence level is warranted. One would hope that confronting the historical record – how our decisions or predictions in a particular situation didn't always work out in the past – would temper our tendency to be overly optimistic about how a similar decision might work out in the future.

To test this theory, the authors of the study on sports betting designed an experiment they called the "history transparency intervention." In this experiment, before placing a bet, sports gamblers viewed summary information about the net negative returns from their past wagers. If the human mind were purely rational, this kind of intervention would surely correct their over-optimism.

But what the Stanford researchers discovered was sobering.

"Even when bettors know their past losses, they remained optimistic about the future," said Matthew Brown, a lead author of the study. "We found that people more or less understood the amount of money they had lost in the past, but they just thought the future would be different."

Why the overconfidence?

Psychologists have posited several factors that contribute to overconfidence. One is attribution error, where people give themselves credit for positive outcomes while attributing negative outcomes to external factors, regardless of the true cause. Think of the sports bettor who remembers vividly all the bets that paid off. These winning moments reinforce the gambler's view of their betting talent. At the same time, the same sports bettor easily dismisses or forgets the losses and writes them off as bad luck.

Another factor is social desirability, where people seek to present themselves in a positive light, causing an overly optimistic view of their actual level of skill or talent.

Whatever the cause, the overconfidence trap goes beyond the world of sports betting to be one of the more insidious traps that befall leaders in organizations. The more senior the leader, and the more supposed success the leader has experienced in the past ("I just got promoted, I must be good."), the more susceptible the leader is to the overconfidence trap.

It isn't necessarily bad to have a bit of overconfidence in some situations. For example, even though most entrepreneurial ventures fail, we need entrepreneurs to believe that their venture will succeed, otherwise there would be very few new companies! However, we want to be aware of when overconfidence could be leading us to make decisions we later regret.

Business implications: When overconfidence proves costly

In business, overconfidence manifests in various ways that can prove detrimental to organizations:

• Strategic planning failures: Executive teams frequently overestimate their ability to forecast market changes and underestimate competitive threats.
• Project management disasters: Leaders routinely underestimate timelines and costs. How often do major projects really come in on time and under budget?
• M&A disappointments: Despite evidence that most mergers and acquisitions fail to deliver anticipated value, companies continue to pursue them with overconfident expectations of synergies. (“This one will be different!”)
• Talent assessment errors: Leaders often overrate their ability to judge character and potential, leading to costly hiring mistakes.

Practical techniques for combating overconfidence

Reducing the likelihood of overconfidence negatively impacting a decision often requires a multi-pronged approach. In Decision Mojo™ we teach useful techniques for doing this, including:

1. Seek disconfirming evidence: Because we have a natural tendency to interpret facts in a way that confirms existing beliefs (the confirming evidence trap), it's helpful to challenge ourselves to actively seek "disconfirming evidence" as well. For example, before investing in a stock, read a report from a short-seller or an analyst who is bearish on the company. Go out of your way to uncover evidence that runs counter to your current thinking.
2. Encourage diverse perspectives: Gather insight and intelligence from people with different backgrounds, experiences or viewpoints from your own. Encourage people on your team to speak up, especially when their opinions or predictions are counter to your own.
3. Ask yourself what you don't know: When approaching a decision, focus as much on what you don't know as what you do know. This simple technique is often a powerful antidote to thinking that you know more than you do.
4. Keep a decision journal: This is critical for tracking your past predictions. When making a decision, write down your rationale for the decision and your prediction for how it will play out. This helps avoid hindsight bias and failure to learn appropriately from past decisions. It also gives you an objective record to reference when determining the success of past similar predictions and decisions.

The pre-mortem: A 5th powerful technique

Consider adding the "pre-mortem" technique to your decision-making toolkit. (This technique was first described by Gary Klein). Before finalizing an important decision, gather your team and ask them to imagine it's one year in the future and the decision has turned out to be a complete disaster. Have everyone write down all the reasons why they think it failed. This exercise forces people to consider potential pitfalls and weaknesses in the plan, counterbalancing natural optimism.   It also enables the team to pre-emptively identify strategies to prevent those reasons from derailing the decision.

The Stanford study highlights one of the most dangerous and challenging aspects of decision-making – our ability to fool ourselves and thus be overconfident when we shouldn't be. Becoming a better decision-maker means accepting this powerful aspect of human nature and approaching decision-making with a sense of humility. As the physicist Richard Feynman famously said, "The first principle is that you must not fool yourself—and you are the easiest person to fool."

Sources:

• Wall Street Journal - You Like to Bet on Sports? Here's a Reality Check. https://www.wsj.com/finance/investing/sports-betting-study-5d9a0044
• Do Sports Bettors Need Consumer Protection? Evidence from a Field Experiment https://mattbrownecon.github.io/assets/papers/jmp/sportsbetting.pdf
• Annie Duke – Thinking in Bets https://www.amazon.com/Thinking-Bets-Making-Smarter-Decisions-ebook/dp/B074DG9LQF/
• Gary Klein – The Power of Intuition https://www.amazon.com/The-Power-Of-Intuition and Streetlights and Shadows https://www.amazon.com/Street-Lights-and-Shadows
• Richard Feynman – CalTech Commencement Address 1974 https://calteches.library.caltech.edu/51/2/CargoCult.htm