Five simple strategies, one hidden year of real crypto prices. Study 120 days of chart, pick the strategy you think will finish the year with the most money, and watch the real candles decide.
Price as an index, 100 = the close before the test year. Left of the yellow line: the 120 days you can study. Right of it: the hidden year.
Each round picks one of seven Bybit spot markets and a random 365-day stretch that starts in May 2022 or later. You see the 120 days before it, drawn as an index with the coin and the dates hidden. Pick the strategy you expect to finish the year with the most money. The year then plays out, all five strategies trade the same real candles, and the table ranks them by final balance. Your pick earns 100 points for first place, 60 for second, 30 for third and 10 for fourth. Five years make one game.
Every strategy starts with $10,000, trades long only, meaning it either owns the coin or holds cash, and pays the same costs: a 0.1% fee on every buy and sell plus 0.05% slippage. Signals are read at the daily close and filled at the next day's open, so no strategy can trade on a price it could not have known yet.
The RSI uses Wilder's smoothing over 14 days. If moving averages or RSI are new to you, start with moving averages explained and what is the RSI indicator.
The maximum drawdown calculator and the Sharpe ratio calculator explain the last two measures step by step.
We ran all five strategies with the default settings over every 365-day window that starts on a multiple of 7 days after May 1, 2022, in all seven markets. That makes 1,246 overlapping years of real Bybit data, with the same fees and slippage as the game.
| Strategy | Won the year | Median return | Average max drawdown | Median trades |
|---|---|---|---|---|
| Buy and hold | 37.4% | +18.4% | 53.2% | 1 |
| RSI dip buyer | 23.4% | +7.7% | 27.3% | 2 |
| Stay in cash | 19.5% | 0.0% | 0.0% | 0 |
| Breakout 20/10 | 11.8% | -1.0% | 39.9% | 5 |
| Trend: SMA 20/50 | 7.9% | +0.3% | 40.8% | 4 |
No strategy won most years. Buy and hold won most often and had the best median result, but it also carried the deepest falls, an average maximum drawdown of 53.2%, and it lost money in 41.6% of the windows. Holding cash won almost one year in five, which says a lot about the harsh stretches in this sample. The RSI dip buyer beat buy and hold in 50.3% of the windows with about half the average drawdown. The trend and breakout rules had median returns close to zero, +0.3% and -1.0%, while still suffering average drawdowns near 40%. Because the windows overlap, read these numbers as a description of this period, not as odds for the future.
Trend and breakout rules buy after a rise has already started and sell after a fall has already started. In a strong, lasting trend that lag costs little. In a market that swings back and forth, it means buying high and selling low several times, and every round trip pays fees and slippage. The RSI rule does the opposite: it buys after sharp drops and sells into the next bounce, which works while prices move sideways and hurts when a fall keeps going. Buy and hold never pays for switching, but it rides every crash in full. Each rule is a bet on a kind of market, and the game shows you which kind each hidden year turned out to be.
After each round, the lab under the results lets you change the moving average lengths, the RSI buy and sell levels and the breakout lookbacks, then rerun them on the same year. It is a quick way to see how much a result depends on settings. It is also a lesson in overfitting: try enough combinations on one year and something will look brilliant, yet the same settings can fail on the next year. The next round gives you a fresh, unseen test.
For the wider debate behind this game, read buy and hold vs active trading. To trade real history yourself instead of picking a rule, try the chart replay.
Educational, not financial advice. The candles are real past prices from Bybit, but past results of a strategy do not predict future results. The game uses play money only.
Yes. Every candle is a real daily candle from Bybit's spot market for BTC, ETH, SOL, XRP, DOGE, ADA or AVAX against USDT, from January 1, 2022 onward. The coin and the dates are revealed after each round.
A signal based on the daily close is only known after that close. Filling at the next open avoids look-ahead bias, which is trading on a price the strategy could not have known yet.
No. All five are long only: they either own the coin or hold cash, like a spot account without borrowing.
None won every year. Over 1,246 real 365-day windows, buy and hold won 37.4% of them, RSI dip buying 23.4%, cash 19.5%, breakouts 11.8% and the moving average trend 7.9%. The ranking changed from year to year, which is the point of the game.