The El Farol Bar Problem was proposed by the economist W. Brian Arthur in 1994, named after the El Farol bar in Santa Fe, New Mexico. One hundred people must decide independently, each Thursday evening, whether to go to a bar that is enjoyable when fewer than sixty attend and unpleasant when sixty or more show up. There is no communication, no coordination mechanism, and no shared signal. Each person must predict attendance based only on past patterns — and each knows that every other person is trying to do the same thing. The problem is diabolical because any publicly known prediction method defeats itself: if everyone uses the same model and it predicts low attendance, everyone goes, and the prediction fails. Arthur showed that heterogeneous populations of agents, each using different heuristic prediction methods, spontaneously converge to an average attendance near sixty — the bar's comfort threshold — through a process of ecological competition among strategies. Damien Challet and Yi-Cheng Zhang reformulated the problem as the Minority Game (1997), in which players try to be in the minority, and showed that it exhibits rich dynamics including phase transitions and memory effects. No optimal strategy exists. The problem has become a foundational model in complexity economics, agent-based modelling, and bounded rationality. The Academy hosts the El Farol Bar Problem in the World School because it is the game of coordination without coordination — a problem every city-dweller plays, unconsciously, every weekend.