The thesis
channel-pullback is a disciplined dip-buyer. It waits for a confirmed uptrend, then buys weakness into the lower regression channel or a volume-support shelf, and sells into the upper channel or resistance. The universe is deliberately boring — 24 US large caps spanning mega-cap tech (AAPL, MSFT, NVDA), payments (V, MA), healthcare (JNJ, UNH), staples (PG, KO, WMT), and energy (XOM, CVX). This is a strategy that wants liquid names with clean channels, not lottery tickets.
Backtest performance
Over 451 days the strategy turned $10,000 into $10,761.52 — a 7.62% total return, or roughly 4.19% CAGR. The headline numbers are honest but unspectacular: a Sharpe of 0.40, a 14.83% max drawdown, and a win rate of just 39.39% across 137 trades.
That low win rate is not a red flag on its own. A pullback strategy that cuts losers and lets winners run to the upper channel is supposed to win less than half the time and make it back on payoff asymmetry. What is worth flagging is the turnover: 2,311% is a lot of churn for a 7.62% return. Fees were modest here ($137, effectively $1/trade, with zero FX cost), but in a higher-friction account that activity would eat meaningfully into the edge.
Why validation failed
The cross-validation tells the more useful story. Three of four folds were positive, and the out-of-sample slice returned +3.63% at a 0.74 Sharpe — not nothing. But the gate marked it failed, and the folds show why:
- Fold 1 (Aug 2024–Jan 2025): +6.53%, Sharpe 1.25
- Fold 2 (Jan–Jul 2025): -11.42%, Sharpe -1.70
- Fold 3 (Jul–Dec 2025): +20.68%, Sharpe 3.86
- Fold 4 (Dec 2025–May 2026): +3.63%, Sharpe 0.74
The dispersion is enormous. Fold 3's 3.86 Sharpe is almost certainly a favourable regime for buying dips, while fold 2 shows what happens when pullbacks keep pulling back — an 11% loss and a 16% drawdown. The statistics agree: a PSR of 0.702 is borderline, but the deflated Sharpe (DSR) of 0.196 — after accounting for 7 trials — says the edge may not survive the multiple-testing haircut. This is a strategy whose quality depends heavily on the market regime it meets.
Recent live activity
Live trading has been calm. Over the last week of scheduled runs, total equity hovered between $9,889 and $10,033 — currently $9,906.03, modestly below the starting line. Recent trades show the pattern working as designed: it bought MSFT at 522.07 and AAPL at 344.84 on 30 Oct and sold both into strength on 4 Nov (MSFT +6.9%, AAPL +2.2%), then re-entered and added energy exposure via XOM and CVX. The 6 Nov run executed nothing — appropriate when no name is sitting at channel support.
The verdict
channel-pullback is a coherent, well-behaved mean-reverter with a believable edge and a sensible universe. But the failed validation is earned, not noise: one bad regime erases two good ones, and the deflated Sharpe is weak. Treat it as a regime-dependent satellite, not a core allocation — and watch fold-2-style environments closely.