The Thesis
Bollinger-reversion is a textbook mean-reversion play: buy when price closes below the lower Bollinger band, sell when it pushes above the upper band. It trades a broad, liquid universe of 24 large-cap US names spanning tech, financials, healthcare, staples, and energy. The premise is that short-term price extremes revert, and the band width adapts to volatility — a sensible, well-understood edge that has historically worked in range-bound markets and struggled in strong trends.
Backtest Performance
On paper, the numbers are inviting. Over 451 days the strategy returned 17.55% (9.46% CAGR), finishing at $11,755 from a $10,000 base, with a 63.89% win rate across 76 trades. That hit rate is the headline strength — nearly two in three trades closed green.
But the risk profile is less flattering. The Sharpe of 0.66 is modest for a strategy with a 20.57% maximum drawdown, meaning returns came with meaningful volatility. Turnover of 1,721% is also high; the strategy churns its book aggressively, and fees ($76 across 76 trades) nibble at the edge.
Validation: The Warning Sign
This is where the story turns cautionary. Walk-forward validation failed. All four folds were positive — good — but the trajectory is a steady decay: fold 1 returned 7.67% at a healthy 1.78 Sharpe, then the edge collapsed to 1.42%, 0.77%, and finally 0.40% (Sharpe 0.14) in the most recent out-of-sample window. The strongest fold sits furthest in the past.
The deflated metrics confirm the concern. The PSR of 0.814 looks reasonable in isolation, but the Deflated Sharpe Ratio of 0.342 — which penalises for the 6 trials run — sits well below the confidence threshold. In plain terms: once you account for how many variants were tested, we cannot be confident the true Sharpe is positive. The full-sample return is likely flattered by an early, favourable regime rather than a durable edge.
Live Activity: Stalled
The live book echoes the validation picture. The last executed trades were on 7 August (a CAT buy, a DIS sell). Since then, every scheduled run from 19–26 August logged zero executions and one-to-two rejected orders — the strategy wants to trade but its orders are being turned away, likely on cash or sizing constraints, with $2,752 idle. Meanwhile total equity has drifted from $9,973 down to $9,739, a slow bleed while the book sits largely inert.
Verdict
Bollinger-reversion has a coherent thesis and an encouraging win rate, but the evidence points to a fading, possibly overfit edge: the out-of-sample return is a rounding error, the DSR fails, and the live book is both stalling and slipping. It earns a place as a monitored experiment, not a conviction allocation — worth watching for whether the recent order rejections are a fixable plumbing issue or a symptom of an edge that has simply run out.