BEAT THE COUCH · REQUIRED READING
THE DATA, CONTINUOUSLY UPDATED · FIRST FILED JULY 2026
Probably not, and for the first time there is a public scoreboard. BEAT THE COUCH is a free market timing game played on real S&P 500 history. Every player gets $10,000 and two secret years drawn from 1928–2025, dividends included, and one button: buy or sell. The opponent is a couch. It buys on day one and never sells. Whoever ends with more money wins.
Live figures from the global ledger. A win only counts as SKILLED if it beats 90% of 1,000 simulated monkeys replaying the same two years with the same time in the market and the same number of trades, on random days. Anything less is stamped LUCKY.
Read that middle number again. Roughly two out of three people, armed with hindsight-free charts and complete freedom to trade, lose to furniture. Of the third who win, most fail the monkey audit: their timing was indistinguishable from chance. Certified skill shows up in about one game in eight.
Timing the market means being right twice. You must get out before the fall and back in before the rise, and the second decision is the one that ruins people. The market's best days cluster inside its worst stretches, precisely when being in cash feels safest. Miss a handful of them and the damage compounds for the rest of the window. The couch makes zero decisions, so it makes zero timing errors, collects every dividend, and sits through every recovery it never saw coming.
S&P's SPIVA scorecard has tracked fund managers for over two decades. Over any 15-year window, roughly 90% of actively managed US equity funds underperform the S&P 500. These are trained professionals with research teams and very good suits.
Source: SPIVA scorecards, S&P Dow Jones Indices
In 2008 Warren Buffett bet that a plain S&P 500 index fund would beat a hand-picked portfolio of hedge funds over ten years, net of fees. The index fund gained 125.8%. The hedge funds averaged roughly 36%. The losers conceded early.
Sources: Long Bets #362 · Buffett's 2017 shareholder letter
That is the actual title of a famous Journal of Finance study. Barber and Odean examined 66,465 households trading through a discount broker from 1991 to 1996. The average household underperformed the market by about 1.5% a year. The most active traders underperformed by about 6.5% a year. The more people traded, the worse they did.
Source: Barber & Odean (2000), Journal of Finance
Javier Estrada studied about 160,000 daily returns across 15 markets. Missing the 10 best days cut final wealth by roughly half, and those days are statistical outliers that cluster in the middle of panics. In fairness, the worst days hide in the same neighborhoods, which is exactly why timing them apart is so hard.
Sources: Estrada, "Black Swans and Market Timing" · Faber, "Where the Black Swans Hide"
Burton Malkiel wrote in 1973 that a blindfolded monkey throwing darts at the stock listings could pick a portfolio as well as the experts. Decades of dart-throwing contests have largely proven him right. Our audit makes the idea rigorous: a permutation test against 1,000 random schedules decides whether a win was skill or a banana.
Sources: Malkiel, "A Random Walk Down Wall Street" · Permutation tests
The evidence says approximately no. Nine out of ten professional fund managers fail to beat the index over 15 years, and on our ledger only about 12% of games end in a win the statistics accept as skill. Individual streaks happen; repeating them on demand is the part nobody has demonstrated.
A game where you try to beat buy-and-hold by choosing when to be in or out of the market. Ours deals you two secret years of real S&P 500 daily history, so you cannot look up the answer, and settles the result against a couch that bought on day one and never moved.
Across 15 markets and a century of data, final wealth roughly halves. The best days cluster inside crashes and panics, which is when market timers are most likely to be sitting in cash.
In this dataset, for most people, most of the time: yes. The couch wins about 62% of games outright, and most human wins fail the skill audit. This page is a scoreboard and a reading list, not financial advice. The couch is not a licensed advisor. The couch is not licensed for anything.
The ledger looks different depending on which two years you are handed. In windows starting in 1929 players beat the couch 72.4% of the time. In windows starting in 1995 they managed 3.3%. The archive scores ten famous windows one at a time: the 1929 crash, Black Monday, the dot-com bubble, 2008 and six more. The full ledger, live and with the charts and caveats, is here.
Try to beat it yourselffree, no signup, takes about two minutes. You will probably lose to furniture.BEAT THE COUCH · the game · the archive · the data · Privacy · Terms