Thesis
Bollinger-reversion is a textbook mean-reversion play: buy when price closes below the lower Bollinger band, sell when it closes above the upper band. The assumption is that large short-term deviations from a rolling mean are statistically likely to correct. The strategy runs across a 24-name large-cap universe spanning tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), consumer staples and discretionary (PG, KO, WMT, COST, MCD, NKE, HD), energy (XOM, CVX), and industrials (CAT, HON, DIS).
Backtest Performance
Over 451 days, the full backtest produced a 17.55% total return (9.46% CAGR), a 63.89% win rate across 76 trades, and a Sharpe of 0.66. Those headline numbers are respectable for a rules-based strategy with no parameter tuning beyond the band itself.
The max drawdown of 20.57% is the most important caveat. For a strategy that targets stable large-caps, a drawdown of that magnitude implies it was exposed during at least one sustained trending move — the exact environment where mean-reversion logic breaks down. Turnover of 1,721% over the period also signals frequent in-and-out activity, which in a live setting would magnify friction beyond the flat $1-per-trade fee modeled here.
Walk-Forward Validation: The Red Flag
The four-fold walk-forward tells a much soberer story:
| Fold | Period | Return | Sharpe | Max DD |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +7.67% | 1.78 | 7.06% |
| 2 | Jan – Jul 2025 | +1.42% | 0.25 | 20.56% |
| 3 | Jul – Dec 2025 | +0.77% | 0.20 | 6.66% |
| 4 | Dec 2025 – May 2026 | +0.40% | 0.14 | 11.19% |
All four folds are positive — a technical pass on that criterion — but the trend is unambiguous: returns and Sharpe have collapsed monotonically across every fold. Fold 1 looks like a genuinely good strategy; fold 4 looks like noise. The out-of-sample Sharpe of 0.14 and OOS return of 0.40% effectively mean the strategy is flat on a risk-adjusted basis in its most recent period.
The Probabilistic Sharpe Ratio (PSR) of 0.814 and Deflated Sharpe Ratio (DSR) of 0.342 — with 6 trials — confirm the concern. A DSR below 0.5 means that after adjusting for multiple testing, there is no statistically reliable evidence the backtest Sharpe reflects a real edge. Validation: failed.
Live Activity (Week of June 2026)
Deployed on May 31 with $10,000, the strategy opened two positions immediately: WMT (21 shares at $115.75) and COST (2 shares at $956.32). On June 1 it added GOOGL (6 shares at $377.90) and PG (17 shares at $140.03). Since then — June 2, 3, and 4 — each daily run generated one rejected signal and zero executions, with the portfolio oscillating between $9,870 and $10,043 against a $1,008.70 cash reserve.
The rejected signals are worth watching: they suggest the strategy is finding setups but hitting position-sizing or risk limits, not that the market is offering no opportunities.
Strengths and Risks
Strengths: Positive across all backtest folds; transparent, auditable logic; diversified across six sectors; decent win rate.
Risks: The edge appears to be decaying in real time. Mean-reversion strategies are acutely sensitive to regime changes — trending or volatile markets punish them disproportionately, as fold 2's 20.56% drawdown illustrates. The DSR failing threshold is a meaningful statistical signal, not just a conservative formality.
This strategy deserves a longer live observation window before any capital scaling. If fold-5 performance (the current live period) tracks fold 4, a parameter review or universe refresh will be warranted.