Two urns, same odds on paper, and almost everyone still prefers one
Urn A has 50 red balls and 50 black balls. Urn B has 100 balls total, red and black in an unknown ratio. You win money if you draw a colour you bet on in advance. Most people, offered a bet on red, prefer to bet using Urn A over Urn B, and also prefer Urn A when betting on black. What's actually wrong with that pair of preferences?
Reveal the answer
It's logically inconsistent: if you prefer Urn A for red, you must believe Urn B is more likely to be black-heavy, which should make you prefer Urn B for black, yet almost no one does. Daniel Ellsberg used this in 1961 to show people aren't just averse to risk, they're averse to ambiguity itself, unknown odds, even when the known and unknown bets are mathematically equivalent, a finding that broke a core assumption of classical decision theory.
— Daniel Ellsberg, Risk, Ambiguity, and the Savage Axioms — Quarterly Journal of Economics, 1961