The thesis
Bollinger-reversion is a textbook mean-reversion play: buy when price pushes below the lower Bollinger band, sell when it stretches above the upper band. It runs against 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 that liquid mega-caps snap back to their moving average after short-term dislocations — a well-worn edge that has grown harder to harvest as markets have gotten more efficient.
Backtest performance
Over 451 days the strategy returned 17.55%, ending at $11,755 from a $10,000 base — a 9.46% CAGR. The win rate is a healthy 63.89% across 76 trades, and fees were negligible ($76 total, no FX cost). Those are the strengths.
The caveats sit right next to them. The Sharpe ratio is a modest 0.66, and the maximum drawdown reached 20.57% — a fifth of capital surrendered at the worst point. Turnover of 1,720% signals a high-churn book that leans heavily on execution quality; in a live account, slippage the backtest never modelled could quietly erode that edge.
Validation: the record doesn't hold up
This is where the analyst hat comes off diplomatically. The walk-forward validation failed. On paper all four folds were positive, which sounds reassuring — until you read the trend. Returns decay monotonically: fold 1 delivered 7.67% at a 1.78 Sharpe, then 1.42%, 0.77%, and finally 0.40% at a 0.14 Sharpe in the most recent window. The out-of-sample Sharpe of 0.14 is barely distinguishable from noise.
The statistics confirm the worry. The Probabilistic Sharpe Ratio is a comfortable 0.814, but the Deflated Sharpe Ratio — which penalises for the 6 trials run — is just 0.342. Once you account for the number of configurations tested, the evidence that this edge is real, not selection, is weak. A strategy whose earliest fold carries most of its lifetime alpha is the classic signature of a decaying or overfit signal.
Recent activity: a strategy that has gone quiet
Live behaviour reinforces the caution. The last executed trades were on 7 August (a CAT buy, a DIS sell); before that, activity dates to May and June. Every scheduled run from 1–8 September logged the same line: 0 executed, 2 rejected. Cash has been pinned at $2,752.28 while total equity drifts between roughly $9,638 and $9,740 — below the $10,000 start. The bands simply aren't triggering qualifying entries, and the two candidates each day keep getting rejected.
Verdict
Bollinger-reversion has an intuitive thesis and a good-looking headline backtest, but the honest read is cautious. The out-of-sample collapse, sub-0.5 Deflated Sharpe, 20%+ drawdown and a live book that has stopped acting all point the same direction: this is a candidate for review and possible parameter re-fitting, not a strategy to scale up. Watch whether new signals fire or whether the silence persists.