Thesis
channel-pullback is a mean-reversion strategy dressed in trend-following clothing. It buys pullbacks to the lower regression channel or volume support — but only within confirmed uptrends — and exits at the upper channel or overhead resistance. The universe is a conservative slice of US large caps: mega-cap tech (AAPL, MSFT, NVDA), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV) and staples (PG, KO, WMT, COST). The design intent is sensible: fade short-term weakness rather than chase strength, and let the broader uptrend provide a tailwind.
Recent activity
Activity has been thin. Across the scheduled runs from 2026-09-01 to 2026-09-08, most days executed nothing (0 executed, 0 rejected), with single trades on 2026-09-02 and 2026-09-03. Recent fills include a BAC round trip around $62–63, several small AAPL scalps in the $308–$315 range, and additions in UNH and JPM. Notably, the paper account's total value has hovered between roughly $9,467 and $9,618 — below the $10,000 starting mark — with cash drawn down to $34.52. So while the backtest is green, the live paper book is presently underwater.
Backtest and validation
The headline backtest covers 451 days: a 7.62% total return, 4.19% CAGR, and a final equity of $10,761.52. That comes with a Sharpe of just 0.40, a 14.83% max drawdown, and a 39.39% win rate across 137 trades — meaning the strategy wins less than two times in five, relying on winners outsizing losers. Turnover is a heady 2,311%, so this is a high-churn approach; fees ($137) are modest here but would matter more at scale or with slippage.
The more important number is that validation failed. Walk-forward testing across four folds tells a volatile story:
- Fold 1 (Aug 2024–Jan 2025): +6.53%, Sharpe 1.25
- Fold 2 (Jan–Jul 2025): −11.42%, Sharpe −1.70, 16.09% drawdown
- Fold 3 (Jul–Dec 2025): +20.68%, Sharpe 3.86
- Fold 4 (Dec 2025–May 2026): +3.63%, Sharpe 0.74
Three of four folds were positive, and the out-of-sample return (3.63%, Sharpe 0.74) is real but slim. The deflated Sharpe ratio (DSR) of 0.196 is the red flag: after adjusting for the 7 trials run, the evidence of genuine edge is weak, even though the probabilistic Sharpe (PSR) of 0.702 looks superficially healthy. Fold 3's Sharpe of 3.86 is almost certainly regime luck, not repeatable skill.
Verdict
The strengths are a coherent thesis, a defensible universe, and an asymmetric payoff profile that can survive a sub-40% win rate. The risks are more serious: a low headline Sharpe, one badly negative fold, extreme turnover, and — most tellingly — a failed validation with a DSR that suggests the backtest is partly a product of trial selection. Running it live in paper is the right call, but the current below-water equity and quiet execution reinforce caution. This is a candidate for refinement, not for scaling capital toward.