Strategy Thesis
The donchian-breakout strategy applies a Richard Donchian channel rule to a curated 24-stock universe of large-cap U.S. equities spanning technology, financials, healthcare, consumer staples and discretionary, energy, and industrials. The logic is deliberately simple: enter when a security closes at a 20-session high; exit when it closes at a 20-session low. No fundamental overlay, no macro filter — pure price-structure trend following.
Backtest Performance
Over 451 trading days, the strategy posted a total return of +6.95% (CAGR ~3.83%), growing a $10,000 starting account to $10,695. The headline return is modest, and the risk profile makes that more pointed.
A win rate of 38.46% across 108 trades is structurally expected for a breakout system — trend followers win less often but target larger gains per winner. The concern is elsewhere: a Sharpe ratio of 0.34 barely clears the "better than cash" threshold, and a maximum drawdown of 21.73% produces an uncomfortable 3.1× drawdown-to-return ratio. Annual turnover came in at roughly 2,083%, meaning real-world slippage and commissions could meaningfully erode an already thin edge.
Walk-Forward Validation
The four-fold walk-forward analysis reveals a strategy with pronounced regime sensitivity:
| Fold | Period | Return | Sharpe | Max DD |
|---|---|---|---|---|
| 1 | Aug 2024 – Jan 2025 | +0.73% | 0.21 | 6.71% |
| 2 | Jan 2025 – Jul 2025 | −7.46% | −1.07 | 15.58% |
| 3 | Jul 2025 – Dec 2025 | +14.09% | 2.72 | 3.14% |
| 4 (OOS) | Dec 2025 – May 2026 | +11.35% | 1.86 | 6.01% |
Three of four folds were profitable, and the most recent out-of-sample period delivered a strong +11.35% with a Sharpe of 1.86 — genuinely encouraging. Fold 2, however, was a sharp detractor, a period where trend signals in this universe failed to sustain and the strategy gave back capital steadily across 38 trades.
The formal validation gate did not pass. The Probabilistic Sharpe Ratio (PSR) of 0.674 falls well below the standard 0.95 confidence threshold, and the Deflated Sharpe Ratio (DSR) of 0.198 — adjusted for six parameter trials — indicates roughly a 20% probability that the observed edge is genuine rather than an artefact of in-sample fitting. The recent folds are promising, but more live history is needed before the signal can be trusted at scale.
Live Activity: A Quiet Stretch
The strategy has entered a low-activity phase. Every scheduled run from June 26 through July 3 reported zero executions, with 1–3 signals rejected per session — likely hitting position-sizing or cash constraints with no clean breakout setups clearing all filters simultaneously.
The last executed trades were in early-to-mid June: entries in KO, UNH, CAT, ABBV, AAPL, and MSFT on breakout signals, with CAT and MSFT subsequently sold on 20-day low violations. Despite the execution pause, live portfolio value recovered from $9,586 on June 26 to $10,005 by July 3, crossing the $10k mark as existing holdings appreciated.
Risks to Watch
- Regime sensitivity: Fold 2 demonstrated the cost of choppy, range-bound markets — repeated false breakouts accumulating losses across 38 trades.
- Validation gap: A DSR near 0.20 is a statistical caution flag; the live track record is too short to override it.
- Turnover friction: At 2,000%+ annual turnover, real-world execution costs could materially erode the modest backtest edge.
donchian-breakout earns a watchlist position based on its recent out-of-sample strength, but broader allocation should wait for validation to clear or live performance to extend the positive trend across another regime shift.