The thesis
Bollinger-reversion is a classic mean-reversion play: buy when price closes below the lower Bollinger band, sell when it pushes above the upper band. The idea is that extreme deviations from a moving average tend to snap back. The strategy trades a 24-name universe of large-cap US equities spanning tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), and staples/industrials (PG, KO, WMT, COST, CAT, HON). It is currently live on paper.
Backtest performance
On paper, the numbers look inviting. Over 451 trading days the strategy returned 17.55% (finishing at $11,755 on a $10,000 base), a 9.46% CAGR, with a 63.89% win rate across 76 trades. That high win rate is characteristic of mean-reversion — many small wins as prices revert.
The caveats are equally characteristic. The Sharpe ratio is a modest 0.66, and the strategy endured a 20.57% max drawdown — a meaningful drop for a return of this size. Turnover is a heavy 1,720%, meaning the book churns roughly seventeen times over; fees ($76 total here) and slippage would bite harder in a live account with realistic costs.
Where validation breaks down
This is the part that should give any allocator pause. Our walk-forward validation failed the strategy despite all four folds being positive. The problem is decay:
| Fold | Return | Sharpe |
|---|---|---|
| 1 (Aug'24–Jan'25) | 7.67% | 1.78 |
| 2 (Jan–Jul'25) | 1.42% | 0.25 |
| 3 (Jul–Dec'25) | 0.77% | 0.20 |
| 4 (Dec'25–May'26) | 0.40% | 0.14 |
The edge is almost entirely concentrated in the first fold and erodes monotonically thereafter. The most recent out-of-sample window returned just 0.40% at a Sharpe of 0.14 — statistically indistinguishable from noise. The Deflated Sharpe Ratio of 0.342 (adjusted for 6 trials) sits well below the confidence we require, even though the Probabilistic Sharpe Ratio of 0.814 looks superficially healthy. In plain terms: the headline backtest flatters a strategy whose live-forward edge has thinned to near zero.
Recent activity
Activity has been muted. Across the last six scheduled runs (Aug 3–10), the strategy executed only two trades and rejected six — most days produced zero fills, consistent with a book that is largely sitting on cash ($2,752) or fully allocated. The one clean signal, on Aug 7, was a textbook round-trip in DIS: bought 25 shares at $99.18 in June, sold at $105.15 — plus a single-share CAT buy at $843.72. Total equity has hovered near flat around $9,950–$9,980.
Verdict
Bollinger-reversion is a well-behaved, high-win-rate strategy with an intuitive thesis — but the walk-forward evidence says its edge is fading, not compounding. The failed validation gate is doing its job. We would not scale this beyond paper until fold-over-fold performance stabilizes; right now it looks like a strategy living on the memory of one good half-year.