The Thesis
Channel-pullback is a classic dip-buyer with discipline. It waits for a confirmed uptrend, then buys pullbacks to the lower regression channel or volume support, aiming to exit at the upper channel or a resistance level. The logic is sound: in a trending name, temporary weakness is an entry, not a warning. The strategy runs a 24-name universe spanning mega-cap tech (AAPL, MSFT, NVDA), financials (JPM, BAC, V, MA), healthcare, and staples — a deliberately diversified, liquid book.
Recent Activity
Live trading has been calm. Over the last week, scheduled runs executed at most one trade per day, often zero. Recent fills include buys in BAC (31 sh @ $62.98), CAT (2 sh @ $887.13), and AAPL (6 sh @ $302.62), alongside a completed PG round-trip — bought at $145.99, sold at $150.90 for a clean win. A CAT position bought earlier was closed at $834.28, and KO and UNH were also rotated out. Portfolio value has hovered tightly around $10,000 ($9,947 total on Aug 6), so the live book is essentially flat since inception — neither confirming nor refuting the backtest yet.
Backtest Performance
Over 451 days the strategy returned 7.62% (4.19% CAGR), ending at $10,761 on a $10K base. That headline masks a mediocre risk profile: a 0.40 Sharpe, a 14.83% max drawdown, and a 39.39% win rate across 137 trades. A sub-40% hit rate that still nets positive means the winners are meaningfully larger than the losers — consistent with a channel strategy that lets moves run to the upper band. Turnover is a heavy 2,311%, though modest fees ($137 total) kept frictions in check.
Validation: The Red Flag
Here the story sours. Walk-forward validation failed. Across four folds, three were positive — but the spread is alarming:
- Fold 1: +6.53% (Sharpe 1.25)
- Fold 2: −11.42% (Sharpe −1.70, 16% drawdown)
- Fold 3: +20.68% (Sharpe 3.86)
- Fold 4: +3.63% (Sharpe 0.74)
The out-of-sample return averages just 3.63% with a 0.74 Sharpe. More damning, the Deflated Sharpe Ratio is 0.196 (across 7 trials) versus a Probabilistic Sharpe of 0.702 — the deflation for multiple testing all but erases confidence that the edge is real rather than lucky.
Verdict
Channel-pullback has an intuitive thesis and one genuinely strong regime (Fold 3), but its performance is regime-dependent and fragile. Fold 2 shows what happens when trends break: the pullbacks keep coming and never bounce. The near-flat live results and failed validation argue for keeping this one on a short leash — small size, close monitoring, and no promotion until it survives a hostile out-of-sample stretch.