BreakoutDonchian Channel Breakout
Buy fresh N-period highs as price breaks above the Donchian channel and ride the trend — the original Turtle-style breakout, exiting on a shorter opposite channel.
Swing tradingIntermediate1h - daily
The idea
The Donchian channel breakout is the original trend-following entry, devised by Richard Donchian and later made famous by the Turtle Traders of the 1980s. The channel simply plots the highest high and the lowest low of the last N bars — 20 is the classic setting — so its upper line is the best price buyers have paid recently and its lower line the best price sellers have accepted. When price pushes to a fresh N-bar high it closes above the upper channel, which by definition means the market has never been this strong over your chosen window, and that is the long signal. The logic is that a genuine trend must, at some point, make new highs, so buying every breakout guarantees you are aboard every meaningful up-move. The cost is that ranges produce false breaks that reverse, so like all breakout systems it trades frequent small losers for occasional large trend winners.
The setup
Plot a Donchian channel with a lookback that matches your horizon — 20 periods for multi-week swings is the canonical choice, with a shorter channel (often 10) used to time exits. The upper and lower bands mark the breakout triggers, and the middle line, the average of the two, acts as a trend pivot and trailing reference. A channel that is visibly sloping and widening tells you a trend is already underway, exactly the environment breakouts need; a flat, horizontal channel warns of a range where breaks tend to fail. Many traders overlay a longer-term filter, such as a 50- or 100-period moving average or a second, wider Donchian channel, and only take breakouts in its direction. The wider the channel lookback, the fewer and more reliable the signals, at the price of entering later in the move.
Entry
Go long the moment price prints a new N-period high and closes above the upper channel; mirror the rule for shorts on a new N-period low. Acting on the close avoids being faked out by an intrabar spike that snaps back, though the most aggressive traders buy a stop order the instant the old high is exceeded. A lower-risk refinement is to demand that the breakout close with conviction — a full-bodied bar, ideally on above-average volume or momentum — rather than a marginal new high by a tick. The Turtles used a clever filter: skip a breakout if the previous one would have been a winner, on the logic that trends rarely resume immediately after a failed break. Whichever entry you use, define it before the bar closes so you are reacting mechanically rather than talking yourself into a chase.
Exit and targets
The classic Donchian exit is a shorter opposite channel: close a long when price makes a new 10-period low, the Turtle System 1 rule, which lets a trend run while cutting it once momentum clearly turns. Because that exit gives back a chunk of open profit, some traders instead trail behind the middle channel line or a moving average, tightening as the move matures. There is generally no fixed profit target — the whole philosophy is to let winners run to their natural end rather than guess a top — although banking partial profit at a measured objective is a reasonable compromise. Whatever the rule, it must be chosen in advance, because the temptation to exit a lagging system early is what quietly destroys its edge. Trend systems make their money from a handful of outsized winners, so cutting them short to feel comfortable is self-defeating.
Risk management
Set the initial stop where the breakout is proven wrong — commonly the middle channel line, the opposite band, or a volatility multiple such as two ATR below entry — then size the position so that distance is a small fixed fraction of the account, often one percent. Because breakouts cluster losses during choppy, range-bound phases, expect several small losers in a row and make certain no single loss can dent the account. The Turtles famously sized every market by its volatility so that each position carried the same risk, which smooths the equity curve across instruments. Never widen a stop to avoid being taken out; the entire method depends on the losers staying small so the rare trend winners dominate. Cap your total open risk across correlated markets, since a basket of breakouts can all fail together when a trend fizzles.
Best timeframes and markets
Donchian breakouts were born in the futures pits and remain best suited to instruments that trend hard and persistently — commodity and index futures, trending large-cap stocks, and the more directional cryptocurrencies. On the daily chart a 20-period channel captures multi-week swings with tolerable noise, while the 1-hour chart speeds everything up for shorter swings at the cost of more false breaks. The higher the timeframe, the cleaner the breakout and the fewer the whipsaws. The strategy performs worst on rangebound, mean-reverting names that chop sideways and repeatedly fake breakouts in both directions. Deep liquidity matters too, because thin markets gap through channels and turn clean signals into slippage.
Common variations
The best-known variation is the two-channel Turtle system: a 20-period channel for entries paired with a 10-period channel for exits (System 1), or a slower 55-period entry with a 20-period exit (System 2) for longer holds. Some traders use only the middle line as a dynamic trend filter, or require price to break the channel on one timeframe while a higher timeframe already points the same way. Others blend the breakout with a momentum or volume confirmation to weed out the weakest signals. A popular modern tweak is to fade failed breakouts — when a new-high break collapses straight back inside the channel, price often runs to the opposite band. All of these keep the core idea intact: a move to a new N-bar extreme is the market declaring a trend.
A worked example
A crude-oil future has been grinding higher and its 20-day Donchian upper band sits at 78.50, the lower band at 74.00, and the midline near 76.25. Price closes at 78.80, a fresh 20-day high, so you go long the breakout at 78.80 with the stop at the midline, 76.25 — risking 2.55 per contract, sized so that equals one percent of the account. The trend extends over the next month toward 88 as you trail behind a rising 10-day low. Oil finally prints a new 10-day low at 85.40, triggering the Turtle exit and banking roughly 6.60 of profit, about 2.6 times the initial risk. That single winner comfortably pays for the two or three small false breaks that came before it.