The thesis
Mean-reversion is one of the oldest ideas in systematic trading, and this strategy states it plainly: buy the oversold (RSI below 30), sell the overbought (RSI above 70). It operates over a 24-name universe of US large caps spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), healthcare (JNJ, UNH, PFE), and staples (PG, KO, WMT, COST). The premise is that short-term price extremes snap back toward a mean — a bet on noise, not trend.
Backtest performance
On paper, the numbers look inviting. Over 451 days the strategy returned 14.73% (7.98% CAGR), finished at $11,473 on a $10,000 base, and won 70.59% of its 38 trades. A high win rate is characteristic of mean-reversion, which harvests many small reversions.
Two caveats temper the enthusiasm. The Sharpe ratio is a modest 0.58, meaning the returns came with meaningful volatility, and the max drawdown reached 15.64% — larger than the annual return itself. Turnover of 879% signals heavy churn; the $38 in fees is trivial here, but at real-world costs and slippage, frequent trading erodes edge quickly.
Validation: where it breaks
This is the decisive section, and the verdict is unambiguous: validation failed. Across four walk-forward folds, three were positive, but the most recent fold (Dec 2025–May 2026) returned -2.84% with a -0.33 Sharpe and a 14.96% drawdown. That out-of-sample stretch is exactly the period a live deployment would have traded — and it lost money.
The risk-adjusted diagnostics reinforce the concern. The Probabilistic Sharpe Ratio sits at 0.785, but the Deflated Sharpe Ratio is just 0.304 after accounting for 6 trials. In plain terms: once you adjust for how many configurations were tried, the evidence that this strategy has genuine edge is weak. The strong full-sample return looks partly like a fit to history rather than a durable signal.
Recent activity
Live behaviour tells its own story. The last executed trade was a WMT buy (21 shares at $115.75) on May 31. Since then, six consecutive scheduled runs — August 4 through 11 — each logged 0 executed, 0 rejected. No RSI extremes have triggered, cash has been frozen at $7,569.25, and total equity has drifted between roughly $9,900 and $9,936. The paper account is sitting below its starting value and doing nothing.
That idleness is not necessarily a flaw; a disciplined mean-reversion system should stay flat when no extremes appear. But combined with the failed validation, it means the strategy is neither earning nor proving itself.
Verdict
Strengths: a clean, interpretable thesis; a high win rate; and folds 1–3 that show the logic can work in the right regime.
Risks: a failed out-of-sample gate, a deflated Sharpe near the floor, a drawdown exceeding annual return, and a recent live stretch that is flat-to-down. The honest read is that this is a promising idea that has not cleared the bar for capital. Until the RSI thresholds or universe are revisited and re-validated, mean-reversion belongs in the lab, not the live book.