The thesis
Channel-pullback is a buy-the-dip-in-an-uptrend strategy. 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. It runs live over a 24-name large-cap universe spanning tech (AAPL, MSFT, GOOGL, NVDA), financials (JPM, BAC, V, MA), healthcare, staples, and energy. In principle this is a disciplined way to enter strong names at a discount rather than chasing breakouts.
Recent activity
The last week has been quiet on the execution side. Scheduled runs on 2026-09-17 and 09-18 each executed 0 trades and rejected 2 — the model is finding candidate setups but declining to act on them, likely on entry or cash constraints (free cash sat at $2,033.05). Earlier runs were busier: three fills on 09-11, two each on 09-14 through 09-16.
The trade tape shows active NVDA rotation — selling 9 shares near $214 and $212 while also buying 9 shares at $212.39 — alongside fresh buys in XOM (11 @ $163.54), BAC (32 @ $59.86), and HD (6 @ $312.34), and sells trimming V, AAPL, and UNH. Total equity moved up sharply from roughly $9,341 on 09-15 to $11,166 on 09-16 and has held near $11.1k since.
Backtest and validation
Over 451 days the backtest returned 7.62% (4.19% CAGR) to a final equity of $10,761.52, with a 0.4 Sharpe and a 14.83% max drawdown. Win rate is only 39.39% across 137 trades — so the edge comes from winners outrunning losers, not from being right often. Turnover is very high at 2,311%, though modelled fees were a benign $137 total.
Crucially, formal validation did not pass. Three of four walk-forward folds were positive, and out-of-sample return was a respectable +3.63% (OOS Sharpe 0.74). But the folds are wildly inconsistent: fold 3 posted +20.68% (Sharpe 3.86) while fold 2 lost -11.42% (Sharpe -1.7, 16% drawdown). After deflating for 7 trials, the Deflated Sharpe Ratio collapses to 0.196 — well short of confidence — even as the raw PSR reads 0.702.
Strengths and risks
Strengths: a coherent, well-diversified thesis; positive OOS returns; and a favourable asymmetry that carries the low hit rate.
Risks: the low DSR says the headline numbers may be luck across 7 trials, the -11.42% fold shows the approach breaks down in unfavourable regimes, and heavy turnover leaves it exposed to real-world slippage the flat fee model understates. The rising rejection count suggests fewer clean setups today. Verdict: promising mechanics, but not yet validated — size it as an experiment, not a conviction position.