Donchian Channel — the Turtle breakout system step by step

In 1983 two Chicago traders made a bet about whether trading could be taught to anyone. Richard Dennis believed it could; his partner William Eckhardt thought it was an inborn talent. To settle it, Dennis handed a group of beginners a simple, mechanical system built around a channel that Richard Donchian had invented decades earlier, and let them trade his own money. That experiment became the legend of the Turtles, and it is why the Donchian Channel still draws attention today. Here I take the indicator apart and show how its breakouts are traded.

What the Donchian Channel is

The Donchian Channel is one of the simplest indicators in technical analysis, and that simplicity is its strength. It is made of three lines drawn from the last N candles: the upper line is the highest high over that period, the lower line is the lowest low, and the middle line is their average. With the common setting of twenty candles, the upper edge of a daily chart marks the highest level of the past twenty days and the lower edge the lowest of the same span.

The indicator was devised in the 1950s by Richard Donchian, widely regarded as the father of systematic trend-following. The logic is intuitive. If price climbs above its own twenty-day high, something is happening that did not happen anywhere in that window: buyers have taken control. The channel does not forecast anything; it merely records that the market has stepped outside its recent range.

How the Turtles turned the channel into a system

„A good trading system does not beat the market by predicting anything; it simply lets profits run and cuts losses short." — Curtis Faith, Way of the Turtle, McGraw-Hill, 2007.

The Turtle experiment turned the Donchian Channel into a complete set of rules, taught in two variants. The first opened a position on a twenty-day breakout and closed it on a ten-day break in the opposite direction. The second was slower: entry on a fifty-five-day breakout, exit on the opposite twenty-day breakout. In both, the shorter channel served as a moving exit rather than an entry signal.

What mattered most was what the system deliberately did not do. It did not guess tops or bottoms and never tried to be right more often than it was wrong; the Turtles knew that most trades would end in a small loss and that the whole result would come from a handful of strong trends. This is trend-following in its purest form.

How to trade the channel breakout

The entry mechanics are almost embarrassingly simple; the difficulty lies in the discipline. In the basic variant you open a long position when a candle closes above the upper line of the twenty-day channel, above the highest high of the past twenty days, and a short when price closes below the lower line. Waiting for the candle to close, rather than reacting while it is still forming, filters out some of the false breaks in which price barely grazed the edge and turned back.

A channel breakout is simply one way of trading range breakouts, so the same laws govern it: it works beautifully in a clear trend and fails painfully in consolidation. That brings us to the method's greatest weakness.

The greatest weakness — whipsaw in a sideways market

The Donchian Channel has one painful flaw: in a trendless market it produces a string of false signals known as whipsaw. Picture a currency pair circling inside a narrow range for weeks. Price touches the upper line, so the system tells you to buy, and a moment later it turns and falls to the lower edge, turning the entry into a loss. There a sell signal fires and the same thing happens in reverse. Each jolt trims the account by a small amount, and in a consolidation there can be a dozen of them in a row.

This flaw cannot be removed completely, because it is written into the nature of breakout trading, but it can be reduced. A higher timeframe helps, since a daily chart produces fewer false breaks than a five-minute one, and so does a simple trend filter that only trades breakouts agreeing with a long moving average. The most honest approach, though, is to accept that whipsaw is the entry fee you pay to catch the next big trend. When the market is plainly going nowhere, it is wiser to reach for a mean-reversion strategy; the broader trading-strategies guide on ForexMechanics sets the two approaches side by side.

A shorter channel as the exit, and sizing your position

Two elements separate a grown-up system from naive clicking on breakouts. The first is an exit based on a shorter channel. Instead of a fixed price target, you close a long position only when price breaks the lower line of the ten-day channel, below the lowest low of the past ten days. Such a moving level shifts only with the trend and never retreats, letting profit grow as long as the trend lasts. It is, in fact, one variety of a trailing stop.

The second element is sizing the position according to volatility, which the Turtles measured with the ATR, the average true range. You decide what fraction of your capital you are willing to lose on a single trade, say 1 percent, and you measure the stop-loss distance as a multiple of the ATR, the way the Turtles used twice the ATR. You then choose the position size so that price travelling through that distance means losing exactly that 1 percent — larger positions on calm markets, smaller ones on jittery markets, with the cash risk identical every time.

Take a purely hypothetical, illustrative setup. On a 10,000-dollar account you risk 1 percent, so 100 dollars; price breaks the twenty-day high, and two ATR is the distance at which a full position would lose those 100 dollars if the market turned against us. You enter at that size, set a stop two ATR away, and leave the position alone until price breaks the ten-day low or reaches the stop.

What to do tomorrow

1. Load the Price Channels or Donchian Channels indicator with a twenty-candle setting on the daily chart of a pair you follow, scroll back several months and mark every place where a candle closed beyond the edge of the channel.
2. Count by hand how many of those breakouts developed into a clear trend and how many were false breaks that ended with a return to the middle, because this simple tally shows you with your own eyes how low the hit rate is and why only the big moves matter.
3. Add a second, shorter channel with a ten-candle period and trace how its opposite edge would have behaved as an exit level, checking how often it would have kept you in the position longer than a rigid, predefined price target.
4. Before you risk real capital, practise the full system on a demo account for at least a month, and always tie the position size to volatility measured by the ATR so that the loss on any single trade never exceeds 1 percent of the account balance.