The thesis
Bollinger-reversion is about as classical as systematic trading gets: buy when price closes below the lower Bollinger band, sell when it pushes above the upper band. It's a pure mean-reversion bet, applied across a 24-name universe of large-cap US stocks spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), staples (PG, KO, COST) and energy (XOM, CVX). The strategy is currently flagged live.
The headline numbers look fine
On a 451-day backtest, the strategy returned 17.55% (9.46% CAGR), ending at $11,755 on a $10k base. Win rate is a healthy 63.89% across 76 trades, and fees are negligible at $76 total. Taken alone, that's a strategy you'd happily fund.
The catch is the risk profile behind it. A Sharpe of 0.66 is mediocre for a 63% win rate, which tells you the losers are meaningfully larger than the winners — exactly the fat-left-tail signature mean-reversion strategies are prone to. The 20.57% max drawdown confirms it: you'd have needed a strong stomach to hold through it.
Validation says: don't
This is where the story turns. The walk-forward validation failed. All four folds were positive, which sounds reassuring, but look at the decay:
| Fold | Return | Sharpe |
|---|---|---|
| 1 (Aug'24–Jan'25) | 7.67% | 1.78 |
| 2 | 1.42% | 0.25 |
| 3 | 0.77% | 0.20 |
| 4 (Dec'25–May'26) | 0.40% | 0.14 |
Nearly all the edge lives in the first fold. The most recent out-of-sample window returned just 0.4% at a Sharpe of 0.14 — statistically indistinguishable from noise. The Deflated Sharpe Ratio of 0.342 (against a PSR of 0.814 over 6 trials) reinforces the verdict: once you adjust for how many configurations were tried, the apparent skill mostly evaporates.
Live activity confirms the fear
The recent trade log is telling. The last executed trades were back in early June (buys in DIS, GOOGL, PG, WMT, COST). Since late July, every scheduled run reads the same: 0 executed, 1–2 rejected. Cash has sat frozen at $1,001.98 while total portfolio value drifts between roughly $9,350 and $9,870 — below the $10k line. The bands simply aren't triggering clean entries, and whatever the risk checks are, they're vetoing the rest.
Verdict
Bollinger-reversion is a clean, low-cost, high-win-rate idea whose backtest flatters it. The strengths are real: cheap turnover-adjusted execution and a majority of winning trades. But the risks dominate — front-loaded historical edge, a punishing drawdown, a failed walk-forward, and a live account that's now idling below par. This is a strategy to keep on a short leash and study, not one to scale.