The Thesis
Channel-pullback runs a straightforward mean-reversion play inside a trend filter: buy pullbacks to the lower regression channel or volume support in a confirmed uptrend, then exit at the upper channel or resistance. It trades a 24-name large-cap universe spanning tech, financials, healthcare, staples and energy — AAPL, MSFT, NVDA, JPM, UNH, XOM, CAT and peers. The design intent is sound: only fade weakness when the primary trend is up, so pullbacks resolve into continuations rather than reversals.
Recent Activity
The strategy is live and trading daily. Recent scheduled runs (Aug 12–21) executed cleanly with zero rejections — mostly single-name entries and exits. Notable round-trips include a BAC position opened at 64.07, sold at 64.52, then re-entered at 64.24, plus a CAT exit at 837.16 (bought at 863.68 — a loss taken) and paired UNH trades around the ~385–390 level. Portfolio total equity has drifted down over the window, from roughly $9,747 on Aug 14 to $9,446 on Aug 21 against a $10,000 paper base. That is a mild, orderly bleed rather than a blow-up, but it is a bleed.
Backtest & Validation
Over 451 days the backtest returned +7.62% (final equity $10,761, ~4.19% CAGR) with a Sharpe of 0.40 and a max drawdown of 14.83%. Win rate is a low 39.39% across 137 trades — meaning the edge, such as it is, comes from winners outrunning losers, not from being right often. Turnover is heavy at 2,311%, so execution and fee assumptions matter.
The honest headline: validation did not pass. Walk-forward analysis across four folds went 3-for-4 positive, but the spread is wide — Fold 3 posted a stellar +20.68% (Sharpe 3.86) while Fold 2 lost -11.42% (Sharpe -1.70). Out-of-sample return averaged +3.63% at 0.74 Sharpe, respectable but thin. The statistical flags are where it stumbles: the Probabilistic Sharpe Ratio is a decent 0.70, but the Deflated Sharpe Ratio is just 0.196 across 7 trials — once you penalise for selection across variants, the evidence that the edge is real rather than lucky becomes weak.
Balance Sheet: Strengths vs Risks
Strengths. The trend-gated entry is disciplined, drawdowns in three of four folds stayed contained, and the OOS numbers are positive rather than degenerate. The Fold 3 performance shows the concept can pay when conditions cooperate.
Risks. The low win rate leaves it dependent on a handful of large winners — a fragile profile. Fold 2's double-digit loss shows the trend filter does not save it in choppy or downtrending regimes. Most importantly, the sub-0.2 DSR says we should treat the backtest edge as unproven, and the live paper equity slipping below par is consistent with that caution.
Verdict
A plausible idea that has not yet earned conviction. Watch whether live results claw back toward par and whether it survives a Fold-2-style regime before sizing up.