Thesis
Channel-pullback bets on a simple structural idea: in confirmed uptrends, price frequently overextends and snaps back to its regression channel floor or volume support before continuing higher. The strategy buys those dips and exits near the upper channel or at resistance. It targets liquid large-caps — a 24-ticker universe spanning tech (AAPL, MSFT, NVDA, GOOGL), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), consumer staples and discretionary (PG, KO, WMT, COST, MCD, NKE, HD), energy (XOM, CVX), industrials (CAT, HON), and media (DIS). The universe concentration in quality names is intentional: regression channels are more statistically stable in stocks with consistent institutional participation.
Recent Activity
The past week has been unusually quiet. The June 15 run executed zero trades with one held open. June 12 saw the most action: a buy of 9 shares of NVDA at $204.82 and a sale of 2 shares of CAT at $912.06 — the latter closing a position entered just two days prior at $857.78, a clean channel-to-resistance exit. June 10 added 2 shares of CAT and closed a 4-share MSFT position at $400.31 (entered June 4 at $427.54 — a loss that underscores the strategy's directional risk on entries). A cluster of activity on June 4 rotated out of UNH and JPM while initiating DIS and MSFT longs.
Portfolio equity has ranged from $10,194 to $10,413 over the past week, with cash sitting at roughly $3,740 — suggesting the book is about 64% deployed. Rejection counts (1-2 per day) indicate the filter is doing its job, declining setups that don't meet channel or trend criteria.
Backtest & Validation
Over 451 days and 137 trades, the strategy returned 7.62% (CAGR: 4.19%) with a Sharpe of 0.40 and a maximum drawdown of 14.83%. Turnover is high at 2,311%, reflecting frequent short-duration swing trades. The win rate of 39.4% is below-average on a hit-rate basis, meaning the strategy depends on winners being meaningfully larger than losers — a classic asymmetry that needs active monitoring.
The four-fold walk-forward tells a more complex 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% |
Three of four folds are positive, but Fold 2's −11.42% loss with a 16% drawdown is a serious outlier. It coincides with the early-2025 volatility window, suggesting the strategy struggles when trend structure breaks down across the board. Fold 3's exceptional Sharpe of 3.86 may partly reflect a cooperative trending environment rather than edge.
The out-of-sample return is 3.63% (Fold 4 as OOS), and the probabilistic Sharpe ratio (PSR) of 0.702 falls short of the conventional 0.95 threshold — meaning the measured edge cannot yet be distinguished from noise with high confidence. The deflated Sharpe ratio (DSR) of 0.196, adjusted for the 7 parameter trials, is more sobering still. Validation: failed.
Strengths and Risks
Strengths: The thesis is theoretically grounded, the universe is liquid, the filter actively rejects marginal setups, and three of four folds produced positive returns. The OOS Sharpe of 0.74 is healthier than the full-period figure, which is mildly encouraging.
Risks: High turnover inflates transaction costs, the win rate requires a reliable payoff ratio to work, and Fold 2 shows the strategy can drawdown sharply in choppy or reversing markets. The PSR/DSR gap means the backtest edge is not yet statistically credible. Until the strategy accumulates more live paper-trade history — and a second OOS fold — treat this as a work-in-progress, not a validated system.