Thesis
Channel-pullback operates on a classic mean-reversion premise: in a confirmed uptrend, price rarely moves in a straight line. The strategy waits for pullbacks to the lower linear regression channel or a volume-supported level before entering, then targets the upper channel or a resistance zone on exit. Applied to a 24-stock universe of large-cap U.S. equities — tech (AAPL, NVDA, MSFT, GOOGL), financials (JPM, BAC, V, MA), healthcare (JNJ, UNH, PFE, ABBV), and consumer/industrial names — the setup favors patience over breadth and quality over frequency.
Backtest Snapshot
Over 451 days and 137 trades, the strategy returned 7.62% (CAGR 4.19%) with a Sharpe of 0.40 and a max drawdown of 14.83%. The win rate of 39.39% is characteristic of a payoff-ratio approach: the strategy tolerates frequent small losses in exchange for proportionately larger wins. Turnover came in at 2,311% — elevated on paper, though the universe consists of highly liquid large-caps where execution slippage beyond the modeled $1/trade commission is unlikely to be severe.
Walk-Forward Validation
Four-fold cross-validation reveals wide performance dispersion:
| 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 (OOS) | Dec 2025 – May 2026 | +3.63% | 0.74 | 8.44% |
Three of four folds were profitable, and the out-of-sample period — the cleanest test of real predictive value — delivered 3.63% with a 0.74 Sharpe. That is the most honest data point in the set. Fold 3's 20.68% return and 3.86 Sharpe are eye-catching, but outlier periods like this tend to inflate aggregate statistics and should be treated with proportionate skepticism.
Fold 2 is the unavoidable counterweight. The –11.42% return and –1.70 Sharpe in the January–July 2025 window suggest the strategy is regime-sensitive: in environments where apparent support levels keep breaking lower, a pullback-buyer can catch a falling knife.
The headline flag: validation has not passed. With a PSR of 0.702 and a Deflated Sharpe Ratio of only 0.196 across 7 parameter trials, the statistical case for a genuine edge — net of in-sample fitting — sits below the acceptance threshold. The DSR in particular is difficult to argue away.
Recent Activity
The past week has been deliberately quiet. Four of the last six scheduled runs produced zero executions, consistent with a strategy that waits for conditions rather than forcing entries. Two trades stand out:
- June 18 — Closed DIS: sold 1 share at $103.99 after entering June 4 at $99.29, a tidy gain on a two-week hold.
- June 25 — Opened AAPL: bought 7 shares at $279.39, deploying roughly half of idle cash.
Earlier in the month, a two-day CAT round-trip (buy $857.78 → sell $912.06) illustrated the strategy's ability to capture sharp intra-trend bounces. MSFT told the other side of the story: entered at $427.54 on June 4, exited at $400.31 on June 10 — a loss that exemplifies the regime-sensitivity risk. The portfolio sits at approximately $9,995 as of the June 25 run.
Strengths and Risks
Strengths: The OOS fold is a genuine positive data point. A Sharpe of 0.74 on unseen data suggests the core entry logic carries real information. The strategy's selectivity — often going days without a signal — is a feature, not a bug.
Risks: The sub-40% win rate means consecutive losing trades are routine, demanding strict discipline. Fold 2's profile raises the question of how the strategy behaves in risk-off or trending-lower regimes. Most importantly, the DSR of 0.196 across 7 trials means the backtest edge cannot yet be distinguished cleanly from parameter-fitting.
Verdict: Channel-pullback has a coherent thesis and a credible OOS result. It stays in paper-trading mode until cross-fold consistency improves and the deflated Sharpe clears a meaningful bar.