Thesis
Channel-pullback targets mean-reversion within a trend structure: it buys when price retreats to the lower linear-regression channel or a volume-supported floor in a confirmed uptrend, then exits near the upper channel or a defined resistance level. The universe spans 24 liquid large-caps across technology, financials, healthcare, consumer staples, energy, and industrials — sectors with generally well-behaved trend-and-revert dynamics. The setup is intuitive and has a long practitioner history, but as the validation data shows, intuition alone is not sufficient.
Validation Status
The strategy is live but has not passed the platform's statistical gate. Two metrics warrant particular attention:
- Probabilistic Sharpe Ratio (PSR): 0.702 — above the 0.5 threshold, meaning there is roughly 70% confidence the true Sharpe exceeds a benchmark of zero. That is encouraging but not decisive.
- Deflated Sharpe Ratio (DSR): 0.196 — this is the more conservative figure, adjusted for the number of trials (7) and the length of the backtest. A DSR below 0.5 suggests the observed Sharpe is plausibly a product of selection bias rather than genuine edge. With 7 parameter or universe configurations tested, the bar for significance rises, and channel-pullback does not clear it.
The full-period Sharpe of 0.40 and a CAGR of 4.19% over 451 days are modest returns for a strategy that turns over its notional capital roughly 23× per year.
Cross-Fold Performance
Four walk-forward folds tell a mixed story:
| Fold | Period | Return | Sharpe | Max DD |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +6.53% | 1.25 | 6.16% |
| 2 | Jan 2025 – Jul 2025 | −11.42% | −1.70 | 16.09% |
| 3 | Jul 2025 – Dec 2025 | +20.68% | 3.86 | 3.26% |
| 4 | Dec 2025 – May 2026 | +3.63% | 0.74 | 8.44% |
Fold 3 stands out as exceptional — a Sharpe of 3.86 over five months suggests the strategy found near-ideal conditions (steady trending markets with clean pullbacks). Fold 2, by contrast, inflicted a 16% drawdown, most likely in a choppy, range-bound or fast-reversing tape where the regression channel offered false signals. Three of four folds are positive, and the most recent out-of-sample fold (Fold 4, which is also the OOS window used in validation) returned +3.63% with a 0.74 Sharpe — a reasonable result, and the basis for continued live operation despite the failed gate.
Recent Activity
Execution over the past two weeks has been selective. The strategy completed a clean roundtrip on CAT (bought at $857.78, sold at $912.06 six days later) and a smaller win on DIS (bought at $99.29, exited at $103.99 two weeks later). The MSFT position entered June 4 at $427.54 was closed June 10 at $400.31 — a losing trade, consistent with the strategy's 39% win rate. An NVDA entry at $204.82 on June 12 (9 shares) remains the most recent open position.
Since June 12, scheduled daily runs have found no new setups (June 15–17 all showed zero executions), with a single exit on June 18. Portfolio equity sits at $10,292, with roughly 37% in cash — the strategy is patient, not forcing trades.
Strengths and Risks
Strengths: The thesis is grounded in well-documented market microstructure (regression-to-trend, volume support). The OOS Sharpe (0.74) beats the in-sample full-period Sharpe (0.40), a mild positive sign against overfitting. Low fee drag ($1 flat per trade) keeps friction manageable even at high turnover.
Risks: The DSR failure is real and should not be dismissed. Fold 2's drawdown of 16% — nearly the full-period maximum — shows meaningful vulnerability in trend-hostile environments. The 39% win rate means the strategy depends on asymmetric payoffs; if average winners shrink (e.g., tighter channels, faster reversals), the edge erodes quickly. Continued monitoring through at least one more complete market cycle is warranted before any capital allocation beyond paper trading.