Strategy Recipes
Donchian Channel Breakout Strategy: Backtested on BTC
The turtle-style Donchian breakout idea as no-code rules, plus an honest BTC/USDT backtest result and why breakout win rates are low by design.
Traseq··8 min read
The turtle-style Donchian breakout idea as no-code rules, plus an honest BTC/USDT backtest result and why breakout win rates are low by design.
Start with a no-code crypto spot strategy, lock the version, run the backtest, and keep the result traceable for comparison.
A Donchian channel breakout strategy buys when price closes above its recent high and exits when price closes below its recent low. It is the core idea behind classic turtle trading: stop trying to predict tops and bottoms, and instead follow price when it breaks out of its recent range. This post turns that idea into a no-code rule set, runs it on real BTC/USDT history, and shows you the honest result — including a window where it lost money.
Traseq is a research workspace, not a live trading or exchange-execution platform. It does not place orders, connect to exchange accounts, or guarantee performance. Everything below is historical research on past data, not a prediction or a recommendation to trade.
A Donchian channel is just the highest high and lowest low over a lookback window. With a 20-bar lookback, the upper band is the highest high of the prior 20 bars and the lower band is the lowest low. The breakout logic is simple:
The turtle traders of the 1980s built a famous trend-following system on exactly this structure: a longer channel to enter and a shorter channel to exit. The premise is that big trends start as breakouts, so you accept many small false signals in exchange for catching the occasional large move.
Here is the breakout rule set used in the interactive demo. None of this requires Pine Script, Python, or any code — you build it in Sentence mode, Canvas mode, or from the Donchian template.
The asymmetry is deliberate. A 20-bar entry channel keeps you out until a move is convincing, while a faster 10-bar exit channel gets you out before a full reversal develops. The 5% stop caps the damage on a failed breakout, and the 3% trailing stop locks in gains if a trend actually runs.
Conditions are evaluated on bar close, and signal-driven entries and exits fill at the next bar open. If you have never expressed rules this way, the no-code strategy rules guide walks through the pattern.
The Learn hub ships a no-signup interactive demo that runs three system templates on real BTC/USDT 1H candles from 2024-11-03 to 2024-12-31 (initial balance $10,000, zero fees, 100% position sizing). That window was a sideways-to-down chop after a rally — an honest, unflattering test. Here is how the Donchian breakout template did over the full sample:
It lost money, and it had the widest drawdown of the three demo templates. That is not a flaw in the report — it is exactly what a breakout system does in a choppy, non-trending market. There was no big trend to catch, so the strategy paid the cost of false breakouts without ever collecting the payoff that justifies them.
For context, the SMA(200) trend filter returned -6.89% and the RSI mean reversion template just broke even at +1.74% over the same window. Trend and breakout systems both struggled; only the mean-reversion idea held flat. That contrast is the whole point of trend following vs mean reversion: different rule families fit different market regimes, and this regime suited none of the directional ones.
A 34.5% win rate looks alarming, but for a breakout system it is normal — not a bug. Breakout strategies are built around a simple bargain:
So a low win rate is the price of admission. What matters is whether the average winner is large enough to cover the many losers — which is why profit factor matters more than win rate for breakouts. A profit factor of 0.66 means the strategy made $0.66 for every $1.00 it lost. The few winners in this window were not big enough, because no sustained trend ever showed up. If you want to dig into why a high win rate alone proves nothing, see why a high win rate can still lose money.
The Donchian template took 29 trades in two months — the most of any demo template. That activity is the breakout system's nature: it re-arms on every range expansion and chases every move past the channel. The demo runs at zero fees, so the 29 trades cost nothing extra. In a real research run with fees and slippage enabled, frequency turns directly into cost.
A rough way to think about it:
This is why breakout systems are sensitive to trade frequency. The more often they fire on false signals, the more the fee drag compounds — and the deeper the trend has to run to make the survivors worth it. When you move from the demo to a real Traseq run, set realistic fees and slippage early, and watch how the drawdown and risk picture changes. For a calmer, slower-firing comparison, the moving average crossover strategy trades far less often on the same kind of logic.
The fastest way to understand a breakout system is to watch its trades play out. Open the interactive demo on the backtesting basics page, select the Donchian template, and step through the 29 trades to see where the false breakouts cluster. Then change the sizing knob and notice how return and drawdown scale together — a Donchian breakout at 100% sizing carried a -15.54% drawdown for a reason.
When you are ready to test your own breakout idea on the pair and timeframe you care about, start in the crypto backtesting workspace. For the underlying execution model — bar-close conditions, next-bar-open fills, fees and slippage — read Core Concepts.
It is a trend-following strategy that enters when price closes above the highest high of a recent lookback window (a 20-bar high in this recipe) and exits when price closes below a recent low. It is the structure behind classic turtle trading and is designed to catch large trends while accepting many small false breakouts.
Low win rates are normal for breakout systems by design. Most breakouts fail and produce small losses, while a few become real trends that produce large gains. The 34.5% win rate in the demo reflects this bargain — what matters is whether the winners are big enough, which is captured by profit factor, not win rate.
The demo window (2024-11-03 to 2024-12-31 on BTC/USDT 1H) was a sideways-to-down chop with no sustained trend. Breakout systems only pay off when a big trend arrives to reward the trailing stop. Without one, the strategy paid the cost of false breakouts and returned -10.27% with a -15.54% drawdown.
Breakout systems trade often because they re-arm on every range expansion — 29 trades in two months in this demo. With real fees and slippage enabled, each round trip costs money, so high frequency adds a fee drag on top of any losses. The trend that finally arrives has to be large enough to cover both the false breakouts and the costs.
Yes. You can build the Donchian breakout rules in Sentence mode, Canvas mode, or from the template, finalize a version, and run a bar-based backtest on supported crypto spot pairs and timeframes. No Pine Script, Python, or custom code is required.
| Component | Rule |
|---|
| Entry | Enter long when close crosses above the prior 20-bar high |
| Exit (signal) | Exit when close crosses below the prior 10-bar low |
| Stop loss | 5% hard stop |
| Trailing stop | 3% trailing stop to protect open profit |
| Metric | Donchian Breakout |
|---|
| Return | -10.27% |
| Win rate | 34.5% |
| Max drawdown | -15.54% |
| Trades | 29 |
| Profit factor | 0.66 |
| Sharpe | -0.16 |
Mar 27, 2026