The thesis
bollinger-reversion runs one of the oldest ideas in technical trading: buy when price closes below the lower Bollinger band, sell when it pushes above the upper band. It operates on a 24-name large-cap universe spanning tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, V, MA), staples (PG, KO, WMT, COST) and healthcare (JNJ, UNH). The premise is that extreme moves in liquid blue chips tend to snap back — a clean, interpretable edge if it holds.
Headline backtest
On paper the numbers look inviting. Over 451 days the strategy returned 17.55% (a 9.46% CAGR), turned $10,000 into $11,755, and won 63.89% of its 76 trades. Those are the strengths, and they are real: a high hit rate and a positive full-sample Sharpe of 0.66.
The caveats sit right next to them. Maximum drawdown reached 20.57% — a fifth of capital — for a Sharpe below 0.7, which is a modest reward for that pain. Turnover of 1,720% means the book churns aggressively; here fees were only $76, but at scale that activity is a headwind, not a rounding error.
Why validation says no
This is where the story turns. The walk-forward validation failed. All four folds were positive, which sounds reassuring until you read them in order:
| Fold | Return | Sharpe |
|---|---|---|
| 1 | 7.67% | 1.78 |
| 2 | 1.42% | 0.25 |
| 3 | 0.77% | 0.20 |
| 4 | 0.40% | 0.14 |
The edge decays monotonically. Nearly all of the headline return was earned in the first fold; by the most recent out-of-sample window the strategy returned 0.40% at a Sharpe of 0.14 — statistically indistinguishable from flat. The Deflated Sharpe Ratio of 0.342, computed across 6 trials, confirms the concern: after accounting for selection, the observed performance is not convincing. The PSR of 0.814 is the one bright spot, but it can't outweigh a DSR that low and an OOS return that has collapsed.
Recent live activity
The live paper account echoes the caution. The last executed trades were on 2026-08-07 (buying 1 CAT, selling 25 DIS). Every scheduled run since — 10th through 17th September — reads the same way: 0 executed, 2–3 rejected. Cash has been frozen at $2,752.28 while total equity drifted from ~$9,764 down to $9,647, leaving the account modestly underwater against its $10k start.
Six weeks of rejected-only runs suggests the signal simply isn't firing, or that risk/position checks are blocking entries. Either way, the strategy is currently a spectator.
Verdict
bollinger-reversion has an honest thesis and a flattering full-sample record, but the evidence points to an edge concentrated in one early window and fading since. The failed validation, decaying folds, and stalled live book all say the same thing: treat this as a candidate under observation, not a deployable strategy. It earns its place in the lab — not yet in the allocation.