Overconfidence effect
Our subjective certainty routinely outruns our actual accuracy.
What it means
The overconfidence effect is the systematic tendency for people's confidence in their own judgments to exceed their objective accuracy. Researchers distinguish three faces of it: overestimation (thinking your performance is better than it is), overplacement (thinking you are better than most others), and overprecision (setting confidence intervals far too narrow). The effect is driven partly by selective attention to confirming evidence, by the ease with which a single plausible story comes to mind, and by failures of metacognition that leave people poorly calibrated about what they do not know. It is among the most robust biases in the literature, though its magnitude depends on task difficulty — hard tasks inflate overconfidence while very easy tasks can produce underconfidence (the hard–easy effect). It matters because overconfident forecasts underlie cost overruns, market bubbles, medical misdiagnosis, and strategic blunders, and because confidence is socially persuasive even when it is unwarranted.
Examples
Asked for a 90% confidence range for a factual quantity, most people set ranges that capture the truth far less than 90% of the time — a clear case of overprecision.
Ask a room of drivers to raise a hand if they are above average behind the wheel and nearly every hand goes up. They cannot all be right, yet nobody lowers theirs.
Two developers promise the feature by Friday because a clean version of the work springs instantly to mind. The broken API and the reviewer on holiday never get imagined, so Friday becomes March.
First described in Lichtenstein, Fischhoff & Phillips (1977).