The thesis
channel-pullback is a disciplined mean-reversion strategy dressed in trend-following clothing. It only shops for dips inside confirmed uptrends: when a name in its 24-stock large-cap universe (AAPL, MSFT, NVDA, JPM, UNH, PG, KO, and peers) pulls back to the lower regression channel or a volume-support shelf, the strategy buys, then exits at the upper channel or resistance. The intuition is sound — you are trying to harvest noise around a rising trend rather than fight it.
Recent activity
The strategy is live and trading a ~$10K paper book. As of the 2026-07-30 run it held roughly $6,461 in cash against a $10,016 total, so it is currently under-deployed and patient rather than pressing. Recent scheduled runs have been quiet — 0 to 3 executions per day, zero rejections — which is exactly what you want from a pullback system in the absence of qualifying setups.
The trade tape tells the real story: this book lives on defensive consumer staples. It round-tripped PG twice (buying near $145–146, selling near $150.6–150.9) and cycled UNH (in around $419–420, out near $433) and KO. These are textbook channel round-trips — small, repeatable edges on low-beta names. That also explains the eye-watering 2,311% turnover: the strategy trades constantly, so fees and slippage matter more here than for a buy-and-hold peer.
Performance and the validation verdict
Headline backtest numbers are respectable-but-modest: +7.62% total return over 451 days (a 4.19% CAGR), a 0.40 Sharpe, and a 14.83% max drawdown. Note the 39.4% win rate — the strategy loses more often than it wins, so its profitability depends entirely on winners being larger than losers. That asymmetry is fine in theory but fragile in practice.
Our walk-forward validation did not pass, and the fold-by-fold breakdown shows why. Three of four folds were positive, but they were wildly inconsistent:
| Fold | Return | Sharpe | Max DD |
|---|---|---|---|
| 1 | +6.53% | 1.25 | 6.2% |
| 2 | −11.42% | −1.70 | 16.1% |
| 3 | +20.68% | 3.86 | 3.3% |
| 4 | +3.63% | 0.74 | 8.4% |
Fold 2 (early-to-mid 2025) was ugly: a double-digit loss with a −1.70 Sharpe. Pullback strategies get punished exactly when an "uptrend" quietly rolls over into a downtrend, and that fold looks like one of those regimes. Out-of-sample the strategy returned a gentler +3.63% at a 0.74 Sharpe, healthier than the full-sample Sharpe but well below the fireworks of fold 3.
The robustness statistics reinforce caution. Across 7 trials the PSR is 0.70 — decent — but the deflated Sharpe (DSR) is just 0.196, which is the number that fails the gate. Once you penalize for multiple testing, the confidence that this edge is real, not lucky, is thin.
Verdict
channel-pullback is a plausible, capital-efficient idea that is currently profitable but unproven. Strengths: patient deployment, sensible universe, and a clean edge on staples. Risks: a punishing failure mode in trend reversals (fold 2), a sub-40% win rate that leans on tail winners, and a DSR that says the sample is too thin to trust. Keep it live and paper-only until the deflated Sharpe and out-of-sample consistency improve.