The Thesis
Bollinger-reversion is a textbook mean-reversion play: buy when price closes below the lower Bollinger band, sell when it pushes above the upper band. It runs across a 24-name universe of large-cap US equities spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), staples (PG, KO, WMT, COST), and healthcare (JNJ, UNH, PFE). The premise is that statistically stretched prices revert toward their moving average — a strategy that historically rewards patience and punishes trending markets.
Backtest Performance
On paper, the numbers are attractive. Over 451 days the strategy returned 17.55% (9.46% CAGR), turning a $10,000 book into $11,755. The 63.89% win rate across 76 trades is genuinely high, and total fees were a negligible $76. The catch is risk-adjusted quality: a Sharpe of 0.66 is mediocre, and the 20.57% max drawdown means investors endured a fifth of their capital underwater to capture that return. Turnover of 1,720% also signals a strategy that trades hard for its gains.
Where Validation Breaks Down
This is where the story turns cautionary. Walk-forward validation did not pass. All four folds were positive — a good sign on the surface — but the edge decayed monotonically:
- Fold 1 (Aug 2024–Jan 2025): +7.67%, Sharpe 1.78
- Fold 2: +1.42%, Sharpe 0.25
- Fold 3: +0.77%, Sharpe 0.20
- Fold 4 (Dec 2025–May 2026): +0.40%, Sharpe 0.14
The most recent out-of-sample window returned just 0.40% at a Sharpe of 0.14 — effectively flat. The deflated Sharpe ratio (DSR) of 0.342, accounting for 6 trials, confirms the concern: once you adjust for multiple testing, the strategy's statistical significance is thin. The PSR of 0.814 is more encouraging, but the two metrics together paint a strategy whose early strength looks increasingly like a favorable-regime artifact rather than a durable edge.
Recent Live Activity
Live paper trading reinforces the caution. The last executed trades were in late May and early June — buys in DIS, GOOGL, PG, WMT, and COST, plus one PG sell. Since then, scheduled runs from July 10–17 have executed zero trades, with several signals rejected. The account sits at roughly $9,787 total against $1,001 in cash — currently below its $10,000 starting line. In short, the bands simply aren't triggering actionable, accepted entries in the current market, and the deployed book is drifting.
Verdict
Bollinger-reversion has a coherent thesis and an eye-catching backtest, but the evidence for a persistent edge is weak. The fold-by-fold decay, sub-0.35 DSR, and flat recent out-of-sample results argue against trusting the headline 17.55%. Combined with a stalled live book and a 20% drawdown profile, this strategy belongs in the watch-and-refine bucket — not the deploy-with-conviction one. The next question worth answering: is the signal genuinely broken, or just waiting on volatility that hasn't shown up?