RSI(14) mean reversion on QQQ — Published Backtest & Verdict

We tested it in public: RSI(14) mean reversion on QQQ (2019-01-23 to 2026-09-25) returned +90.9% vs +381.4% buy-and-hold across 7 trades. Exact rules, full stats, and our honest verdict.

• Total return: +90.9% (trailed buy-and-hold at +381.4%)
• CAGR: +8.8% vs +22.8% buy-and-hold
• Max drawdown: -29.6% vs -35.1% buy-and-hold
• Win rate: 85.7% over 7 trades
• Period: 2019-01-23 to 2026-09-25 (QQQ), published 2026-09-26

The claim on trial

Buy when RSI(14) goes oversold, sell when it goes overbought — mean reversion beats buy-and-hold.

The exact rules

• RSI(14) with Wilder smoothing, seeded from the first 14 one-bar changes; first signal at bar 14 (2019-01-23).
• Signal observed at bar i close; trade prints at bar i close.
• Entry on bar i starts earning on bar i+1 (NO look-ahead: the signal bar's own move is not credited to the new position).
• Exit on bar i retains bar i's return (position was held during the bar).
• RSI < 30 → buy at close. RSI > 70 → sell at close.
• Long only, fully invested when long, cash earns 0%. No transaction costs, no taxes modeled.

Our honest verdict

Six of seven trades won (85.7%), yet the strategy returned less than a quarter of buy-and-hold (+90.9% vs +381.4%): it sat in cash 72% of the time during a historic bull run, and its one loser bought near the top of the 2022 bear market and held 13 months underwater. High win rate, tiny exposure — win rate is not the score, total return is.

Full trade log on Substack · Original X thread · StratIQ track record · All published backtests

Charted on TradingView (affiliate link — we may earn a commission at no extra cost to you): tradingview.com

Not financial advice — research and education only; do your own due diligence.