Golden cross and moving average crossovers: the most famous strategy, tested

Moving average crossover is a trend-following strategy in one sentence: buy whenThe moving average the fast one crosses above the slow one, and sell when it crosses below. The famous version, the "Golden Cross," uses a 50-day average versus 200; the "Death Cross" is the opposite crossover. Fast versions (20 versus 50) react earlier and reverse more often.

In our tournament, the fast 20/50 crossover achieved +110% versus +95% for the index, and the Golden Cross +70%on just 28 trades. Both had a maximum drawdown similar to that of the index. What this means and what it doesn't, below.

The rules we ran

• Signal: The fast average closes above the slow one after being below it (a crossover, not a state). Buy at the open the next day. Sell on the reverse crossover. No stop, no target, as the method is taught.
• Portfolio: 10% of equity per position, up to 10 positions; when all are filled, additional crossovers are skipped. The choice between them is by order of arrival, not by quality. This is a limitation of the measurement, not of the method.
• Regime: Entries only when S&P 500 is above 200 average. The measurement: 184 American stocks and ETFs, daily bars, May 2023 to September 2026 (840 trading days), execution at the open following the signal, cost of 0.05% per side, long only and without leverage. The universe is today's and not 2023's — stocks that collapsed and disappeared are not included, which biases every result upward. A rising market for most of the period.

The results

How to read this

• The Golden Cross on the index itself lost to simple buying. It exited with a Death Cross in spring 2025, missed the recovery, and returned late. This is the built-in lag: the 200 average reacts to a change only after it has already happened. In a market that falls and recovers quickly, the crossover is always late in both directions.
• 28 trades is not a sample. The Golden Cross on individual stocks is rare, and the cap of 10 positions left a large portion of equity in cash. Its row in the table is statistically weak, for better or worse.
• The fast crossover looks good, and pay attention to why. 129 trades, profit factor 2.5, one year near zero. This is a trend-following system with all the characteristics of the turtlesit loses most of the time, and wins big on the few. The "order of arrival" selection between crossovers on the same day is a source of randomness that can move the number in either direction.
• There is no stop. In the pure method, you exit only on the reverse crossover, and that can be dozens of percent below the entry. Anyone trading this with real money needs stop beneath the structure, even if the classic measurement doesn't include it.