Myth-busting backtest
Does the Golden Cross Beat Buy-and-Hold? We Tested It on QQQ (2019–2026)
The 50/200-day golden cross is one of Wall Street's most famous signals. We backtested it mechanically on QQQ over 2019–2026: +238.5% total return vs +289.1% buy-and-hold. It made money — but lost to doing nothing. Full trade log, methodology, and an honest read.
The myth
Buy on the golden cross, sell on the death cross. It's one of Wall Street's most-quoted rules: when the 50-day moving average crosses above the 200-day moving average, get in; when it crosses back below, get out. The implication is that the cross lets you sidestep bear markets and capture the real moves — beating a lazy buy-and-hold investor without riding the drawdowns.
That's a testable claim. So we tested it, mechanically, on QQQ.
The test
Strategy: 50-day / 200-day SMA golden cross on QQQ, long-only, fully invested-or-cash. Golden cross → buy at the close. Death cross → sell at the close. Strategy window: 2019-10-16 → 2026-09-18 (the first date where both the current and previous 50/200-day SMAs are fully formed).
No look-ahead bias: the signal is observed at bar i's close, the trade prints at bar i's close, and the new position starts earning from bar i+1 — the signal bar's own move is not credited. Exits retain the exit bar's return, since the position was held that day. Cash earns 0%. No commissions, slippage, or taxes modeled — real trading does worse by roughly those costs.
The numbers
Strategy total return: +238.5%. Buy-and-hold over the same window: +289.1%. Strategy CAGR: 19.3% vs 21.7%. Max drawdown: −22.8% vs −35.1%. Time in market: 74.0% vs 100%. Trades: 3, all winners — a 100% win rate, average winning trade +50.51%.
The verdict: the golden cross made money but lost to doing nothing. It earned its keep via a smoother ride (lower drawdown, about a quarter of the time in cash), not outperformance.