The Thesis
Channel-pullback is a classic buy-the-dip mean-reversion strategy with a trend filter. It waits for a confirmed uptrend, then buys pullbacks to the lower regression channel or a volume-support level, exiting at the upper channel or resistance. The universe is 24 large-cap US names spanning tech, financials, healthcare, staples, and energy — a defensive spread that should smooth out single-sector shocks. The idea is intuitive: in a rising market, temporary weakness is an entry, not a warning.
Recent Activity
The strategy is trading and its equity is hovering near breakeven. As of the 2026-08-04 scheduled run, total account value stood at $10,018.57 against $2,835.29 in cash — the low cash balance signals it is currently near fully invested. Recent runs have been quiet, executing zero to three orders each. The latest fills include a 2-share CAT buy at $887.13 and a 6-share AAPL buy at $302.62. Notably, the strategy round-tripped Procter & Gamble profitably in late July — buying 13 shares at $145.99 and $144.98 and selling at $150.90 — and exited Coca-Cola and Caterpillar positions on 2026-07-28. This is textbook behavior: accumulate on weakness in staples, harvest into strength.
Backtest & Validation
Over 451 days the backtest returned 7.62% (4.19% CAGR) with final equity of $10,761.52. Two numbers demand attention. First, the win rate is only 39.39% across 137 trades — the strategy is wrong more often than right, and relies on winners outrunning losers. Second, turnover of 2,311% is extremely high; the strategy churns its book more than twenty times over, so slippage and fees (which totaled $137 here) matter far more in live trading than in a frictionless test.
The walk-forward validation is where the picture gets honest. Three of four folds were positive, and fold 3 (2025-07 to 2025-12) was superb: +20.68% at a 3.86 Sharpe with a tiny 3.26% drawdown. But fold 2 (early-to-mid 2025) lost -11.42% at a -1.70 Sharpe with a 16% drawdown — the strategy's uptrend filter clearly failed to protect it when the regime turned. Aggregate Sharpe is a modest 0.4, and out-of-sample return (3.63%) is roughly half the full-sample figure.
Crucially, validation did not pass. The probabilistic Sharpe ratio (0.702) is reasonable, but the deflated Sharpe ratio — which penalizes for the 7 trials run — collapses to 0.196, well short of confidence. In plain terms: after accounting for how many variants were tested, we cannot rule out that the edge is luck.
Verdict
Channel-pullback is a coherent, disciplined strategy with genuinely strong performance in trending markets. But its low win rate, heavy turnover, and ugly fold-2 drawdown make it fragile precisely when investors most need protection. Until the deflated Sharpe clears the bar, treat it as a promising candidate under observation — not a validated edge.