The thesis
Bollinger-reversion runs a classic mean-reversion playbook: buy when price closes below the lower Bollinger band, sell when it pushes above the upper band. It trades a 24-name large-cap universe spanning tech (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), staples (PG, KO, WMT, COST) and healthcare (JNJ, UNH, PFE, ABBV). The premise is simple and well-worn — fade short-term extremes and collect the snap-back.
Backtest performance
On paper, the numbers flatter the idea. Over 451 days the strategy returned 17.55% (9.46% CAGR), finishing at $11,755 from a $10,000 base, with a 63.89% win rate across 76 trades. That's a lot of small wins, which fits the mean-reversion signature.
The caveats sit right next to the highlights. The Sharpe of 0.66 is modest for a strategy carrying a 20.57% max drawdown — you're accepting a fifth of your capital at risk for a middling risk-adjusted payoff. Turnover is a striking 1,720%, meaning the book churns roughly 17x over the test window; in a live setting, slippage and financing would eat visibly into that headline return.
Validation: the strategy does not pass
This is where the story turns. Walk-forward validation returns passed: false. All four out-of-sample folds are positive, which sounds reassuring — but the trajectory is the tell:
- 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
Performance decays monotonically toward zero. The most recent out-of-sample return is just 0.4% at a Sharpe of 0.14 — statistically indistinguishable from noise. With a Deflated Sharpe Ratio of 0.342 across 6 trials, the framework is signalling that the in-sample edge likely reflects the test period rather than a durable effect. The PSR of 0.814 is the one bright spot, but it can't outweigh the OOS collapse.
Recent live activity
The live paper account echoes the caution. Scheduled runs from July 9–16 executed zero trades, logging several rejections and holding cash flat at $1,001.98. Total equity has drifted between $9,710 and $9,925 — below the $10,000 start and well under the backtest's final $11,755. The last executed fills were back in early June (buys in DIS, GOOGL, PG, WMT, COST; a PG sell), suggesting the current regime simply isn't triggering the band conditions.
Verdict
Bollinger-reversion is a clean, interpretable strategy with a genuine in-sample track record and a healthy win rate. But the evidence for a repeatable edge is thin: OOS returns fade to zero, validation fails, and live equity is underwater. High turnover compounds the concern once real costs apply. This one belongs in continued paper testing and parameter review — not in a capital allocation — until it can demonstrate an out-of-sample edge that survives the next fold.