Strategy Thesis
Bollinger Reversion operates on a classic mean-reversion premise: buy when price closes below the lower Bollinger Band, sell when it crosses above the upper band. The universe spans 24 large-cap U.S. equities across technology, financials, healthcare, consumer staples, energy, and industrials — a deliberately diversified pool designed to surface episodic oversold conditions in otherwise fundamentally sound names.
Recent Activity: A Quiet Stretch
The past week has been markedly dormant. Every daily run from June 24 through July 1 executed zero trades, with between two and three signals rejected each session. Portfolio value has drifted from $10,060.97 on June 24 down to $9,706.53 as of July 1 — a paper loss of roughly 3.5% over the period, likely reflecting mark-to-market moves in existing positions rather than new activity.
The most recent executed trades date to early June: a buy in DIS at $99.18 (June 10), a PG round-trip — buy at $140.03 and sell at $148.40 nine days later — alongside entries in GOOGL, WMT, and COST at the end of May. The PG trade is a clean illustration of the thesis working as intended: a roughly 6% gain in under two weeks on a defensive large-cap.
Backtest Performance
Over 451 days, the strategy generated a 17.55% total return (9.46% CAGR) on 76 trades, with a 63.89% win rate. Those are credible headline numbers. The Sharpe of 0.66 is modest but positive, and fees were a flat $1 per trade — negligible at this scale.
The drawdown picture is less comfortable. Maximum drawdown reached 20.57%, which is steep for a strategy targeting large, liquid names. Turnover of 1,720% annualized is also notable — reflecting the strategy's willingness to rotate aggressively as band-touch signals fire.
Validation: The Critical Concern
The walk-forward validation did not pass, and the fold-by-fold breakdown explains why. Performance has deteriorated in a near-monotonic pattern:
| Fold | Period | Return | Sharpe | Max DD |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +7.67% | 1.78 | 7.06% |
| 2 | Jan 2025 – Jul 2025 | +1.42% | 0.25 | 20.56% |
| 3 | Jul 2025 – Dec 2025 | +0.77% | 0.20 | 6.66% |
| 4 | Dec 2025 – May 2026 | +0.40% | 0.14 | 11.19% |
Out-of-sample return is just 0.40% with a Sharpe of 0.14 — barely above flat. The Probabilistic Sharpe Ratio (PSR) of 0.814 suggests the true Sharpe is likely positive, but the Deflated Sharpe Ratio (DSR) of 0.342 — which adjusts for the six parameter trials run — is a meaningful red flag. It implies that when multiple configurations were tested, the probability that this is a genuinely superior strategy (rather than a best-of-several fits) is below 35%.
The most plausible interpretation: the strategy extracted a real edge in the volatile, mean-reverting environment of late 2024, but that regime has faded. In trending or momentum-driven markets, Bollinger reversion trades can fire as false entries on stocks breaking down, not bouncing.
Strengths and Risks
Strengths: Diversified large-cap universe limits idiosyncratic blowups. Win rate above 63% is consistent. The thesis is economically intuitive and has documented academic support in range-bound regimes.
Risks: Performance decay across folds suggests regime sensitivity rather than a durable edge. A 20%+ max drawdown is high for the return profile. Extended signal droughts (as seen this week) can cause opportunity cost to accumulate.
Outlook
Bollinger Reversion is best treated as a conditional strategy: effective when the market oscillates, vulnerable when it trends. The current quiet period — with signals being consistently rejected — may indicate the strategy is appropriately sitting on its hands, or it may reflect a market environment that structurally disfavors this approach. Monitoring fold 4 performance through Q3 2026 will be the clearest signal of whether the edge is cyclically suppressed or permanently impaired.