The Thesis
Mean-reversion runs one of the oldest ideas in technical trading: buy weakness, sell strength. Concretely, it accumulates names where the 14-period RSI drops below 30 (oversold) and exits when RSI pushes above 70 (overbought). It operates on a 24-name large-cap universe spanning tech, financials, healthcare, staples and energy — AAPL, NVDA, JPM, JNJ, XOM and friends. These are liquid, well-covered stocks where RSI extremes tend to be transient rather than terminal, which is exactly the regime this logic needs to work.
Backtest Performance
On paper the headline numbers flatter the strategy. Over 451 days it returned 14.73% (final equity $11,473 on a $10k base), a 7.98% CAGR, with a 70.59% win rate across 38 trades. Drawdown topped out at 15.64% and the Sharpe landed at a modest 0.58. Costs were negligible — $38 in fees, zero FX. The one flag worth noting is turnover of 879%: this is an active strategy that churns its book roughly nine times over, so the high win rate is doing a lot of lifting against frequent, small decisions.
What Validation Says
This is where the story turns. The walk-forward check fails. Across four folds, three were positive (+2.06%, +11.10%, +2.21%), but the most recent fold — Dec 2025 to May 2026 — posted −2.84% with a −0.33 Sharpe and a 14.96% drawdown. The out-of-sample Sharpe is negative, and while the Probabilistic Sharpe Ratio (0.785) looks respectable, the Deflated Sharpe Ratio collapses to 0.304 once you account for the six trials run. In plain terms: much of the full-sample return is concentrated in a single strong fold (fold 2), and the edge does not generalise forward cleanly.
Live Activity
The live book reinforces the caution. The last executed trades were buys in late September — MCD, HD, BAC — on top of a May WMT position. Since then, every scheduled run (Sep 30 through Oct 7) reports 0 executed, with a couple of rejections, leaving cash pinned at $329.71 and total equity drifting between roughly $9,357 and $9,464. The strategy is effectively fully invested and waiting: no fresh oversold signals are clearing, and the book is treading water below its $10k start.
The Balance
Strengths: a genuinely high hit rate, positive results in three of four folds, low trading costs, and a disciplined, rules-based entry/exit that is easy to audit. Risks: the validation gate is red for good reason — negative out-of-sample returns, a weak deflated Sharpe, and return that leans heavily on one favourable window. The near-fully-invested, signal-starved live state means it currently offers little optionality to buy a genuine dislocation.
Takeaway
Mean-reversion is a textbook strategy producing textbook ambiguity: attractive in-sample, unconvincing out-of-sample. The high win rate is real but should not be mistaken for a durable edge. Until a fresh fold confirms the signal survives forward in time — and the live book shows it can actually deploy capital into oversold names again — this one stays firmly in the 'watch, don't scale' column.