The thesis
channel-pullback runs a classic mean-reversion play inside a trend filter: buy pullbacks to the lower regression channel or volume support in confirmed uptrends, then exit at the upper channel or resistance. It's a disciplined idea — you're not fading a downtrend, you're buying dips in names that are already working. The universe is 24 large, liquid US names spanning tech (AAPL, MSFT, NVDA), financials (JPM, V, MA), staples (PG, KO, WMT) and energy (XOM, CVX), which keeps single-name blow-up risk contained.
Recent activity
The strategy is live and trading on a scheduled daily cadence. Over the last week it has executed a steady 1–3 orders per run with almost no rejections. Recent tape includes trimming NVDA (selling 9 shares near $212–214 across three sessions), rotating into XOM (11 shares at $163.54), BAC (32 at $59.86) and HD (6 at $312.34), plus a quick V round-trip — bought 5 at $366.62 on the 10th, sold 5 at $371.31 on the 11th. That's the channel logic in miniature: enter on weakness, exit into strength. Reported account totals have hovered near $9.3k–$9.4k through mid-September, with the latest run marking $11,166.84.
Backtest and validation
On paper the record is respectable-but-modest: 7.62% total return over 451 days (a 4.19% CAGR), a 0.40 Sharpe, and a 14.83% max drawdown. The win rate is a low 39.39% across 137 trades, so the edge depends entirely on winners being larger than losers — typical for pullback systems, but it means psychological and execution discipline matter. Turnover is very high at ~2,311%, though fees stayed trivial ($137 total).
The honest headline is that formal validation did not pass. Three of four walk-forward folds were positive, and out-of-sample return was a live-looking +3.63% (OOS Sharpe 0.74) — encouraging. But the fold-to-fold spread is alarming: fold 3 printed +20.68% at a 3.86 Sharpe, while fold 2 lost 11.42% at a −1.70 Sharpe with a 16% drawdown. That is not a stable equity curve; it's regime-dependent. The statistics agree: the Probabilistic Sharpe Ratio is a decent 0.702, but the Deflated Sharpe Ratio is just 0.196 after adjusting for 7 trials — below the threshold you'd want before trusting the edge as real rather than lucky.
Verdict
channel-pullback has a coherent, defensible thesis and is trading it cleanly in live paper. The strength is the logic and the risk-controlled universe; the weakness is that the numbers don't yet clear the overfitting bar. I'd treat it as a promising candidate under observation, not a validated allocator — watch whether live results track the +3.63% OOS fold or the −11% fold, and whether the strategy can raise its win rate or cut the fold variance before it earns more capital.