The Thesis
The mean-reversion strategy is built on one of the oldest signal pairs in systematic trading: RSI below 30 flags an oversold entry; RSI above 70 triggers the exit. The universe is 24 large-cap U.S. equities spanning tech, financials, healthcare, consumer staples, and energy — liquid names where temporary dislocations are more likely to snap back than to mark the start of a trend.
The appeal is its simplicity. There are no multi-factor overlays, no sentiment inputs, no macro regime filters. Either a stock has been beaten down far enough to trigger the rule, or it hasn't.
Backtest Performance
Over the 451-day backtest window, the strategy generated a 14.73% total return (7.98% annualised CAGR) on a starting equity base, finishing at $11,473. Across 38 trades, it won on 70.6% of them — a hit rate that would be the envy of many discretionary traders.
The Sharpe ratio of 0.58 is modest but positive. The maximum drawdown of 15.64% is the headline risk figure to watch; for a strategy holding blue-chip names and sitting in cash most of the time, a ~16% peak-to-trough loss is meaningful.
Turnover came in at 879%, which sounds alarming until you note that 38 trades over 15 months across a 24-stock universe means positions cycle through at a measured pace — the figure reflects position sizing methodology more than frantic churn.
Walk-Forward Validation: Where It Gets Uncomfortable
The four-fold walk-forward tells a more nuanced story:
| Fold | Period | Return | Sharpe | Max DD |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +2.06% | 0.57 | 4.24% |
| 2 | Jan 2025 – Jul 2025 | +11.10% | 1.32 | 10.53% |
| 3 | Jul 2025 – Dec 2025 | +2.21% | 0.46 | 8.80% |
| 4 | Dec 2025 – May 2026 | −2.84% | −0.33 | 14.96% |
Three of four folds are positive, and Fold 2 is genuinely strong at Sharpe 1.32. But the most recent fold — the one that matters most as a proxy for live performance — posted a −2.84% out-of-sample return with a negative Sharpe and a near-maximum drawdown of 14.96%.
The validation gate accordingly did not pass. The Deflated Sharpe Ratio (DSR) of 0.304 is the most damning number: adjusted for the 6 parameter trials run, the probability that the observed Sharpe is genuine rather than lucky is below the threshold the platform requires. The Probabilistic Sharpe Ratio (PSR) of 0.785 is more forgiving, but DSR is the stricter and more appropriate metric when a strategy has been iterated.
Recent Activity: Quiet on All Fronts
The strategy's most recent executed trade was a buy of 21 shares of WMT at $115.75 on 31 May 2026. Since then, six consecutive daily runs through early July have executed zero trades. Portfolio value has drifted between $9,848 and $9,975 against a cash position of $7,569 — meaning the single open WMT position accounts for the bulk of market exposure.
The inactivity is not a bug. It means none of the 24 universe names have crossed RSI 30 recently — a reasonable outcome in a market that has been trending rather than oscillating.
Strengths and Risks
Strengths: High win rate, clean logic, no curve-fitting risk from complex parameterisation, and a universe of names with genuine mean-reverting tendencies.
Risks: The most recent out-of-sample fold underperformed badly, and the DSR flags that the historical Sharpe may partly reflect the specific backtest window rather than a persistent edge. A trending or momentum-driven market regime — exactly the environment that characterises extended bull runs — will suppress RSI below 70 and keep the strategy idle or wrong-footed on entries.
This one warrants continued paper-trading observation before any capital commitment. The edge may be real; the recent fold suggests it needs a more volatile, mean-reverting market environment to express itself.