The Thesis
Mean-reversion is one of the simplest ideas in quantitative trading: prices that stretch too far from their recent range tend to snap back. This strategy operationalises that with the Relative Strength Index — buying names when RSI falls below 30 (oversold) and selling when it climbs above 70 (overbought). It runs on a 24-name universe of US large caps spanning tech (AAPL, MSFT, NVDA), financials (JPM, BAC, V), healthcare (JNJ, UNH), staples (PG, KO, WMT) and energy (XOM, CVX). It is currently live.
Backtest Performance
On paper, the headline numbers are appealing. Over 451 days the strategy returned 14.73%, compounding at a 7.98% CAGR to a final equity of $11,473 from a $10k base. The win rate is a high 70.59% across 38 trades, with a maximum drawdown of 15.64% and a Sharpe of 0.58. Trading costs were negligible — $38 in fees, no FX drag.
Two caveats temper the enthusiasm. First, turnover ran to 879%, so this is an active strategy that leans on execution quality. Second, a Sharpe below 0.6 means the return path was bumpy relative to its reward — the high win rate is doing a lot of the psychological heavy lifting.
Validation: The Red Flag
Here the picture darkens. The walk-forward validation did not pass. Of four out-of-sample folds, three were positive (+2.06%, +11.10%, +2.21%) but the most recent — spanning December 2025 to May 2026 — lost 2.84% with a negative Sharpe of −0.33. The aggregate out-of-sample Sharpe was also negative.
The deflated statistics reinforce the caution. The Probabilistic Sharpe Ratio sits at a respectable 0.785, but the Deflated Sharpe Ratio is just 0.304 once you account for the six trials run. That gap is the classic signature of a strategy that looks better in-sample than the evidence truly supports.
Recent Live Activity
Live behaviour has been quiet. Across scheduled runs from 9 to 16 September the strategy executed zero trades, sitting on roughly $7,569 in cash. On 17 September it finally acted, buying 42 shares of BAC at $58.21, drawing cash down to $5,119. Total account value has hovered around $9,800–9,870 — modestly below the $10k mark, meaning the live deployment is currently underwater.
Verdict
Mean-reversion is a coherent, low-cost, high-hit-rate strategy with a genuinely strong in-sample record. But the failed validation is not a footnote — the deteriorating final fold and the low Deflated Sharpe suggest the edge may be thinner than the backtest implies, and the live account's early softness is consistent with that. Watch the BAC position and the next few folds closely: this one has earned a live slot, but not yet our full confidence.