We Think We're Better Than We Are
Overconfidence is one of the most pervasive and well-documented biases in human psychology. People systematically overestimate their own abilities, the accuracy of their knowledge, and their ability to predict the future.
In one classic study, 93% of American drivers rated themselves as above average — a statistical impossibility. In financial forecasting experiments, analysts were asked to provide 90% confidence intervals for their predictions — ranges within which they were 90% sure the true answer would fall. The true answer fell outside their ranges far more often than 10% of the time. They were far more confident than they were accurate.
Overconfidence manifests in three distinct forms: overplacement (believing you are better than others), overestimation (believing you will perform better than you actually do), and overprecision (believing your predictions are more accurate than they are). In financial contexts, all three appear.
Why Overconfidence Persists
Overconfidence is maintained by several reinforcing mechanisms. When predictions prove correct, people attribute the success to skill. When predictions are wrong, they attribute the failure to bad luck or unusual circumstances. This self-attribution bias means that experience tends to increase confidence without a proportionate increase in accuracy.
Feedback in financial markets is also noisy. A poorly reasoned trade can succeed if a stock happens to rise for unrelated reasons; a well-reasoned trade can fail if unforeseen events intervene. Over a short series of trades, luck and skill are nearly indistinguishable. This makes it genuinely difficult to learn from market outcomes in the way one might learn a craft with cleaner feedback loops.
Overconfidence in Investing
The investment implications of overconfidence are well-documented. Barber and Odean's landmark 2000 study "Trading Is Hazardous to Your Wealth" analysed 66,465 household brokerage accounts. The most active traders — those whose behaviour most reflected confidence in their stock-picking ability — earned an annualised net return 7.1% below that of passive buy-and-hold investors. Trading frequency was inversely related to net returns.
Overconfidence leads investors to concentrate portfolios in their highest-conviction ideas rather than diversifying broadly. It leads to excessive trading (generating costs and taxes). It produces the belief that market timing — moving money in and out at the right moments — is achievable through careful analysis, when the evidence suggests it is not.
The corrective is not to become paralysed by doubt, but to design an investment process that does not rely on the accuracy of individual predictions: diversification, index funds, and rule-based rebalancing all reduce the scope for overconfidence to damage outcomes.