Illusion of Validity

People remain confident in their ability to predict outcomes even when given explicit, credible evidence that their basis for prediction has no actual predictive value. Named and studied by Daniel Kahneman and Amos Tversky in their 1973 paper "On the Psychology of Prediction." In one test, participants were given pairs of student ability scores and explicitly told all score pairs were equally predictive of future performance — yet participants consistently favored pairs with consistent scores (e.g. two similar grades) over pairs with inconsistent ones, ignoring the explicit instruction that consistency carried no extra predictive weight. The same pattern holds with financial decisions: given two hypothetical stocks that both end at the same final value — one via a smooth, steady 10% rise, the other via a volatile 50% rise then 33% fall — and told explicitly that past performance won't predict the future, most people still prefer the steadier-looking stock.

A costly real-world instance. On CNBC's Mad Money in March 2008, host Jim Cramer confidently told a viewer asking about Bear Stearns's stability: "Bear Stearns is fine... don't move your money from Bear... don't be silly." Days later, Bear Stearns collapsed and was sold in a government-brokered rescue, its shares down roughly 90%. Cramer's unwavering confidence despite genuinely limited information is cited as a real-stakes example of the illusion of validity (see Jim Cramer Bear Stearns Illusion-of-Validity Case). Related theoretical work by Einhorn and Hogarth (1978) frames the illusion as a persistent overconfidence in judgment that resists correction even after repeated feedback of error.

Discussed in

• Episode 278 — 278-real-world-examples-of-cognitive-biases

Related

• Jim Cramer Bear Stearns Illusion-of-Validity Case
• Overconfidence (Above-Average Effect)
• Tom Bowden-Green
• Luan Wise