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 bet is that short-term dislocations in liquid large caps snap back toward their moving average. The universe is 24 blue-chip US names — AAPL, MSFT, NVDA, JPM, V, JNJ, WMT, XOM, CAT and peers — the kind of deep, well-behaved tickers where band-touch reversion has the best odds of holding.
What the backtest says
On paper the strategy looks solid. Over 451 days it returned 17.55% (CAGR 9.46%), finishing at $11,755 on a $10k base. The win rate is a healthy 63.89% across 76 trades, and fees were negligible ($76 total, no FX cost). The catch is the risk profile: a 20.57% max drawdown against a Sharpe of just 0.66. You are taking a full fifth of your capital in peak-to-trough pain for a return that a passive index arguably delivered with less turnover — and turnover here is a heavy 1,720%, meaning the book churns roughly 17 times over the test window.
Where validation breaks
This is where the story sours. The walk-forward suite fails. All four folds are positive, which sounds reassuring, but the returns decay monotonically as the test moves forward in time:
| Fold | Window | Return | Sharpe |
|---|---|---|---|
| 1 | Aug'24–Jan'25 | 7.67% | 1.78 |
| 2 | Jan'25–Jul'25 | 1.42% | 0.25 |
| 3 | Jul'25–Dec'25 | 0.77% | 0.20 |
| 4 | Dec'25–May'26 | 0.40% | 0.14 |
The most recent out-of-sample fold earns 0.4% at a Sharpe of 0.14 — effectively flat. The Deflated Sharpe Ratio (DSR 0.342) sits well below the confidence threshold once you account for the 6 trials run, even though the raw Probabilistic Sharpe Ratio (0.814) looks respectable in isolation. The single strong fold at the start is doing almost all the heavy lifting; the edge appears to have faded as the sample advanced.
Live behaviour confirms the caution
Recent activity reinforces the picture. The last executed trades were back in early August (a CAT buy, a DIS round-trip); every scheduled run since 28 August has executed nothing — six consecutive days of "0 executed, 2 rejected." Cash is parked at $2,752 and total equity has drifted between $9,638 and $9,740, i.e. the paper book is currently underwater versus its $10k start. Either the bands aren't triggering or risk/liquidity checks are rejecting the signals.
Verdict
Strengths: a coherent, well-understood thesis on a liquid universe, a genuinely high win rate, and trivial trading costs. Risks: a mediocre risk-adjusted return, a punishing drawdown, and — most importantly — a validation profile showing the edge concentrated in one early fold and decaying to noise by the most recent one. The backtest headline is flattering; the walk-forward math is the honest number. Until out-of-sample Sharpe recovers, this one stays a watch-list candidate, not a conviction allocation.