Skip to main content

Overconfidence Bias

Being more confident than accuracy warrants

Judgment

What is it?

Overconfidence bias is one of the most studied cognitive biases and takes three forms: overestimation (thinking we perform better than we do), overplacement (thinking we're better than others), and overprecision (excessive certainty in our beliefs). Many people rate themselves above average on desirable traits, in numbers too high for all of them to be right. Professionals are not immune: overconfidence has been documented among doctors, lawyers, and executives. Entrepreneurs often show high confidence in their own venture's odds, which may help explain both why they start businesses and why many fail. In calibration studies, answers people say they are 90% sure of turn out to be right noticeably less often than that. The bias resists feedback partly because we tend to explain failures as bad luck and successes as skill. Overconfidence leads to inadequate preparation, poor risk assessment, thin contingency plans, and the illusion that complex problems have simple solutions. Calibrating confidence requires tracking predictions over time and deliberately considering scenarios where you could be wrong.

Example

Believing a project will take 3 months when similar projects took 6+ months. Rating yourself as an above-average driver. Being 99% sure of an answer that turns out wrong.

References

Fischhoff, B., Slovic, P., & Lichtenstein, S. (1977). Knowing with Certainty: The Appropriateness of Extreme Confidence. Journal of Experimental Psychology: Human Perception and Performance, 3(4), 552-564.

Lichtenstein, S., Fischhoff, B., & Phillips, L. D. (1982). Calibration of Probabilities: The State of the Art to 1980. In D. Kahneman, P. Slovic, & A. Tversky (Eds.), Judgment Under Uncertainty: Heuristics and Biases (pp. 306-334). Cambridge University Press.

Moore, D. A., & Healy, P. J. (2008). The Trouble with Overconfidence. Psychological Review, 115(2), 502-517.

How to Prevent It

Doxa uses AI and can make mistakes. How it's built

Question

What is my track record on similar predictions?

Question

What assumptions am I making that could be wrong?

Question

How wide should my confidence interval really be?

Question

What do I not know that could affect this outcome?

Question

Would an expert in this area agree with my assessment?

Technique

Seek feedback from people who will challenge your views.

Technique

Use base rates from similar past situations.

Technique

Keep a prediction log and review your accuracy regularly.

Technique

Express estimates as ranges rather than single numbers.

Technique

Conduct pre-mortems to identify overlooked failure modes.