Thesis
The donchian-breakout strategy runs one of the oldest ideas in systematic trading: buy the 20-day high, sell the 20-day low. It trades a 24-name universe of US large caps — the megacap tech complex (AAPL, MSFT, GOOGL, NVDA), payments (V, MA), healthcare, and consumer staples. The logic is deliberately simple: ride established trends and cut positions when momentum reverses. As with all breakout systems, the design accepts many small losses in exchange for a few large winners.
Recent activity
Activity has been thin. Across the six scheduled runs from 2026-09-16 to 2026-09-23, only one — 2026-09-22 — actually filled orders, executing four trades (buying WMT and AAPL, selling V and MA). Every other session logged zero executions, with two or three rejected orders each. The portfolio currently sits at a total value of $9,906.24 against $3,513.36 in cash, meaning roughly a third of capital is uninvested. That idle-cash posture, combined with the run of rejections, suggests few clean 20-day breakouts are firing in the current tape — appropriate behaviour for a trend follower in a range, but a drag on returns while it waits.
Backtest and validation
Over a 451-day backtest the strategy returned 6.95% (final equity $10,695.11), a 3.83% CAGR, with a Sharpe of just 0.34 and a maximum drawdown of 21.73%. The 38.46% win rate across 108 trades is textbook trend-following — profitability leans on winner size, not hit rate. Turnover was extreme at 2,083%, though total fees came to only $108 with no FX cost.
The walk-forward validation is where the picture gets nuanced. Three of four folds were positive, and the out-of-sample numbers look strong: 11.35% OOS return at a 1.86 Sharpe. Fold 3 (2025-07 to 2025-12) was the standout at +14.09% and a 2.72 Sharpe. But fold 2 (2025-01 to 2025-07) lost 7.46% at a -1.07 Sharpe — a reminder of how badly breakout systems can behave in choppy, mean-reverting conditions.
Crucially, validation did not pass. With a PSR of 0.674 but a deflated Sharpe ratio (DSR) of only 0.198 across six trials, the framework judges the edge insufficiently robust once multiple-testing is accounted for. The headline OOS Sharpe flatters a strategy whose full-sample Sharpe is a mere 0.34.
Verdict and risks
Donchian-breakout is a legitimate, well-understood strategy currently earning its keep in the more trending folds and giving it back in the choppy ones. The strengths are transparency and a genuine OOS lift; the risks are a low DSR, a 21.7% drawdown, and heavy turnover that would bite harder under real slippage. It runs live, but the failed validation gate is the honest signal here: treat its recent gains as unconfirmed until the deflated metrics improve.