BEAT THE COUCH · THE ARCHIVE · THE LONG GRIND
THE WINDOW: JAN 11, 1973 TO JAN 9, 1975 · 504 TRADING DAYS
January 1973 to January 1975: the couch fell to $6,515, down 44% at the worst. 62.4% of 1,664 players beat it, but only 7.5% passed the skill audit.
THE COUCH'S $10,000, DAY BY DAY, REAL DAILY DATA
The Nifty Fifty were supposed to be the stocks you could hold forever at any price. Then came the oil embargo, wage and price controls, a president resigning, and inflation that would not sit down. From January 1973 the market ground 44% lower over almost two years.
This is not a crash. It is a slope. The worst single day in the whole window is only 3.7%, which is smaller than a bad afternoon in 2008. It just never stopped.
The couch ends with $6,515. Cash, earning the high rates of the era, returned +15.4% and won comfortably.
Player figures cover every game whose window started in 1973, not only the exact hand above, and come from the run export of October 31, 2026. A win counts as SKILL only if it beats 90% of 1,000 monkeys replaying the same window with the same trade count and the same time in the market. Across the thirty windows in this archive players beat the couch 39.2% of the time, because these are the famous years. Across the whole ledger the couch wins 62% of 120,800 games.
62.4% of 1,664 players beat the couch here, but the skill rate is 7.5%, low for a crash. That combination is the tell. In a market that only falls, being out is automatically right, and a permutation test cannot tell a market read from a nap. Grinding bears make everybody look clever.
Play this exact windowfree, no signup, about two minutes. Same two years, same data, no hindsight.In this window the buy-and-hold couch loses -34.9%, ending with $6,515, with a worst-point drawdown of 44%. It took years to recover in real terms because inflation was running hot at the same time.
Better than usual, on the surface: 62.4% of players beat the couch in windows starting in 1973. But only 7.5% of them beat 90% of random monkeys with the same trade count, so most of that success was position, not prediction.
Method: real S&P 500 daily total returns, dividends included, calibrated to Damodaran's annual series; cash earns 3-month T-bills. The same data the game deals. See the evidence page for sources, and the full ledger for the whole ledger.
BEAT THE COUCH · the game · the archive · the daily hands · the evidence · the data · for teachers · the monkey audit · the fine print · embed the ledger · Privacy · Terms
Not financial advice. The couch is not licensed for anything.