Thesis
The mean-reversion strategy operates on a straightforward premise: markets overshoot. When RSI drops below 30 on a large-cap name, the strategy buys the dip; when RSI climbs above 70, it exits. The universe spans 24 blue-chip and sector-representative tickers — AAPL, MSFT, NVDA, JPM, WMT, XOM, and peers — providing enough breadth to find signals without straying into illiquid territory.
Backtest Performance
Over 451 days, the strategy returned 14.73% on a starting equity base, compounding to a final equity of $11,473 and a CAGR of 7.98%. The 70.6% win rate across 38 trades is the headline strength — seven out of ten closed positions were profitable. Maximum drawdown hit 15.64%, which is meaningful but not disqualifying for an equity strategy of this style.
The Sharpe ratio of 0.58 is modest. It reflects a strategy that earns its return through trade selection frequency rather than per-trade magnitude, and the 879% turnover figure confirms this is an active rotator, not a buy-and-hold overlay.
Walk-Forward Validation
This is where the picture gets more complicated. A four-fold walk-forward validation reveals significant performance dispersion:
| Fold | Period | Return | Sharpe | Trades |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +2.06% | 0.57 | 8 |
| 2 | Jan 2025 – Jul 2025 | +11.10% | 1.32 | 10 |
| 3 | Jul 2025 – Dec 2025 | +2.21% | 0.46 | 6 |
| 4 | Dec 2025 – May 2026 | -2.84% | -0.33 | 10 |
Fold 2 is the standout — a 1.32 Sharpe in a volatile mid-2025 window suggests the RSI thresholds were well-calibrated for that regime. But Fold 4, the most recent and therefore the most predictive, reversed into negative territory. The out-of-sample return of -2.84% and an OOS Sharpe of -0.33 are the primary reasons validation did not pass.
The Probabilistic Sharpe Ratio (PSR) of 0.785 and a Deflated Sharpe Ratio (DSR) of 0.304 — with six trials in the search — tell a similar story: there is a reasonable probability the in-sample Sharpe beats a benchmark, but after deflating for multiple testing, the edge looks thin.
Recent Activity
The strategy has been quiet at the live paper-trading layer. Daily scheduled runs from June 16–23 all returned zero executions. Portfolio value has hovered between ~$10,031 and ~$10,108, with cash parked at $7,569. The only recent trade was a buy of 21 shares of WMT at $115.75 on May 31, suggesting that large-cap staples were the last names to trigger an RSI oversold signal before the current calm.
The extended inactivity is itself a data point: either the 24-ticker universe is broadly overbought or in neutral RSI territory right now, or signal thresholds are tight enough that they fire infrequently in a trending market.
Strengths and Risks
Strengths: High win rate, a diversified large-cap universe, low fee drag ($38 total over 38 trades), and a defined, rules-based entry/exit that is straightforward to audit.
Risks: The most recent fold underperformed, which is the regime the live strategy will face. A mean-reversion approach struggles in sustained trending markets — if RSI stays elevated for months, the strategy either fires into a still-rising price or sits idle. The 15.64% maximum drawdown also deserves attention; two back-to-back losing folds could approach that ceiling quickly.
Outlook
Mean-reversion remains on paper-trading status. The gap between in-sample and out-of-sample returns warrants a longer live observation window before any capital commitment is considered. Watching whether Fold 4's negative regime persists — or reverts — will be the key input for the next validation cycle.