Ask a large group of people whether their driving is above average, and an overwhelming majority say yes.
The arithmetic does not permit this. Roughly half of any group is below average at anything.
It gets more uncomfortable with expertise. Studies have found doctors expressing around 90 percent confidence in a diagnosis when they turn out to be right closer to half the time.
This is not stupidity. It is a stable feature of how humans assess themselves, and it has a specific and expensive effect on money.
Confidence Is Fine. Overconfidence Is Not.
Confidence lets you act. Without some, you would never make a decision at all.
Overconfidence is different. It means believing your information is better than it is, your reasoning sharper than it is, and your judgment more reliable than the evidence supports.
With money, it produces a specific behavior: acting more than you should.
If you believe you understand something others have missed, the rational response is to act on it. Multiply that across millions of people all believing the same thing, and you get an enormous amount of activity — most of it based on a conviction that cannot be true for everyone holding it.
The Evidence Is Unusually Clear
Behavioral economist Terrance Odean examined this directly.
Studying roughly 10,000 brokerage accounts across 1987 to 1993, he found investors were selling and repurchasing close to 80 percent of their holdings each year.
Then he compared what they bought against what they sold.
The stocks they bought consistently trailed the market. The stocks they sold consistently beat it.
Not once — across four-month, one-year, and two-year periods. Each decision to swap one holding for another destroyed value on average, and people were doing it constantly.
Odean and Brad Barber followed with a study of over 66,000 households, published under the title Trading Is Hazardous to Your Wealth. The finding was straightforward: the most active traders had the worst results, and those who traded least earned the highest returns.
These studies are decades old now, and worth reading as landmark evidence rather than current measurement. But the mechanism they identified has not changed, and trading has only become easier since.
The Bias That Prevents You Learning
Here is the part that makes overconfidence self-sustaining.
Every outcome you experience has two possible explanations: something you did, or something outside your control. Skill or luck.
Deciding which is which is how you learn. Get it right, and experience teaches you something. Get it wrong, and you draw confident conclusions from noise.
People sort these outcomes in a predictable and unhelpful way. Good results go in the skill bucket. Bad results go in the luck bucket.
Annie Duke, the poker player and decision researcher, gives a memorable example. Phil Hellmuth, the most decorated player in World Series of Poker history, once said on camera after being eliminated: if it were not for luck, he would win every one.
Poker players found this funny because it is such a pure statement of the bias. Every loss is variance. Every win is ability.
Duke’s point is that most people simply have the sense not to say it aloud.
Why This Is Expensive
The cost is not the arrogance. It is what happens to your learning.
Attributing bad outcomes to luck means never examining the decisions that produced them. There is nothing to review, because the world was at fault.
Attributing good outcomes to skill means reinforcing decisions that may have been poor and happened to work. You repeat them, with more confidence and more money.
Over enough repetitions, you end up systematically confident about approaches that do not work, and blind to the ones that do.
This connects directly to the distinction we drew in investing versus speculating: judging decisions by results rather than reasoning. Overconfidence is what makes that error feel like learning.
The Amplifier: Reacting Too Fast
One further pattern makes it worse.
Research by Richard Thaler and others found people place too much weight on a small number of recent events, treating them as a trend. The most recent piece of information becomes, in the mind, a signal about the future.
People also respond asymmetrically — overreacting to bad news and adjusting slowly to good.
Combine the two. You believe you read the data better than others, you fix on the most recent thing you saw, and you respond sharply to anything negative.
That is a fairly complete description of how someone talks themselves into selling at the worst possible moment while feeling entirely rational.
What Actually Helps
Knowing about overconfidence does not reduce it. Being told you overestimate your driving does not change your estimate.
What works is building in friction and evidence.
Write down your reasoning before the outcome. Not the decision — the reasoning, and how confident you are. When you review later, you can assess the thinking on its own terms rather than reconstructing it through the result. This is the single most effective correction available, and almost nobody does it.
Watch your own language. Duke suggests treating certain phrases as warning signs — “I know,” “I’m certain,” “there’s no way,” “it always happens like this.” Absolutes signal more confidence than the evidence usually supports. Noticing the word creates a moment to reconsider the thought.
Ask what would change your mind. If nothing would, you are not holding a view — you are holding a commitment. Being specific about what evidence would count against you is uncomfortable and clarifying.
Interrogate your wins, not just your losses. Most people review what went wrong. Almost nobody reviews what went right to check whether they actually deserved it. That is where the reinforcement of bad decisions happens.
Act less. The most reliable correction, and the least satisfying. If overconfidence produces excess activity, and excess activity produces worse results, then doing less is a direct remedy — which is why the checking habits in why watching your money too closely costs you matter so much.
Beyond Investing
The pattern is not confined to markets.
A side income that worked may have been timing rather than judgment. A raise you received may have owed as much to a good year for the company as to your case for it. A budget that held may have held because nothing went wrong, not because the budget was sound.
The useful question after anything working out is not “what did I do right?” It is “would this have worked if the circumstances had been slightly different?”
Sometimes yes. Sometimes the honest answer is no, and that is the more valuable thing to know.
The Bottom Line
Overconfidence is close to universal, it does not respond to being pointed out, and it is expensive because it stops you learning from your own experience.
The correction is not humility as an attitude. It is structure: write your reasoning down before you know the outcome, notice when you speak in absolutes, define what would change your mind, examine your successes as carefully as your failures, and act less often.
None of that makes you calibrated. It makes you slightly less confident than you would otherwise be — which, on the available evidence, is an improvement.
Browse our other topics on the Explore BlurbMoney page.