The Thesis
The mean-reversion strategy is one of the oldest ideas in systematic trading: prices that have stretched too far in one direction tend to snap back. The Headmars implementation keeps it simple — buy when RSI drops below 30 (oversold), sell when RSI climbs above 70 (overbought). The strategy runs across a 24-stock universe of large-cap US equities spanning tech, financials, healthcare, consumer staples, and energy.
Simplicity is a feature, not a limitation. A one-rule system leaves little room for curve-fitting, and the universe is diversified enough to give the signal regular opportunities without concentrating in a single sector.
Backtest Highlights
Over a 451-day backtest period, the strategy executed 38 trades and compounded to a 14.73% total return (7.98% CAGR), carrying a Sharpe ratio of 0.58 and a maximum drawdown of 15.64%. The win rate of 70.6% is notably high for a rules-based system — more than two in three closed trades were profitable.
Turnover came in at 879%, which reflects the system's willingness to rotate frequently as RSI signals trigger across the universe. Total fees equalled $38 (flat $1/trade), a negligible drag on the overall return.
Cross-Validation: Where It Gets Interesting
The four-fold walk-forward validation tells a more nuanced story:
| Fold | Period | Return | Sharpe | Max DD | Trades |
|---|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +2.06% | 0.57 | 4.24% | 8 |
| 2 | Jan 2025 – Jul 2025 | +11.10% | 1.32 | 10.53% | 10 |
| 3 | Jul 2025 – Dec 2025 | +2.21% | 0.46 | 8.80% | 6 |
| 4 | Dec 2025 – May 2026 | −2.84% | −0.33 | 14.96% | 10 |
Three of four folds were profitable, but the most recent — and most relevant — fold is the exception. Fold 4 produced a −2.84% out-of-sample return with a −0.33 Sharpe and the largest drawdown of any period at 14.96%. The Deflated Sharpe Ratio (DSR) of 0.304, combined with six prior trials, indicates the strategy does not yet clear the bar for statistical robustness. The validation status is formally failed.
Recent Activity
The strategy has been quiet in live paper trading. Its last executed trade was a 21-share buy of WMT at $115.75 on May 31, and the six daily scheduled runs since (June 2–9) all returned zero executions. As of June 9, the portfolio holds $7,569 in cash against a $10,062 total value, implying roughly $2,493 in open positions. The silence likely reflects a market where large-cap RSI readings have stayed out of the extreme bands the strategy requires.
Strengths and Risks
Strengths: The in-sample win rate and fold 2 Sharpe (1.32) suggest the signal has genuine edge during mean-reverting regimes. The universe is liquid and diversified, reducing execution risk. The rule is transparent and easy to audit.
Risks: RSI-based mean reversion struggles in trending markets — if large-caps enter a sustained directional move, oversold readings can keep getting more oversold. The fold 4 deterioration coincides with increased drawdown and more trades, suggesting the signal fired but markets didn't cooperate. The low DSR (0.30) is the clearest caution flag: with six trials in the search, the measured Sharpe needs meaningful skeptical discounting.
Bottom Line
Mean-reversion is a disciplined, readable strategy with a respectable backtest record. But validation correctly flags that recent out-of-sample performance has not matched the in-sample story. The strategy remains live in paper-trading mode — the right place to continue accumulating evidence before any capital commitment. Watch whether fold 5 (June 2026 onward) recovers, or whether the regime shift in fold 4 proves persistent.