Turtle Trading: The Original Breakout System Built on Rules, Not Prediction

Overview

Turtle Trading is one of the clearest examples of a fully systematic trading method. It was built around a simple idea: big money comes from catching a small number of large trends.

Not forecasting. Not valuation. Not narratives.

The original system taught by Richard Dennis and William Eckhardt was a complete trend-following framework covering entries, exits, position sizing, and risk. It was explicitly designed to remove discretion.

That is what made it powerful. The edge was not any one indicator by itself. The edge was the combination of breakout entries, volatility-based sizing, pyramiding into strength, and strict risk control.

The heart of Turtle Trading is the channel breakout. Price is tracked against its rolling 20-day high and 20-day low. When price exceeds the upper channel, the system enters long. When it drops below the lower channel, the system exits. Most breakouts fail, but a small number turn into very large trends — and those few trends dominate overall returns.

The 1989 Wall Street Journal article profiled 14 Turtle-related commodity trading advisers and reported an average annual compound return of roughly 80 percent over the prior 4.5 years. This is the strongest credibility visual for the Turtle system: contemporaneous reporting showing extraordinary results from a rules-based, teachable process.

Turtle-style systems have a distinctive payoff shape: many small losing trades and a few very large winners. The win rate is often below 50 percent, but the average winner is far larger than the average loser. That positive skew is the entire game.

The Turtle experiment started in 1983 as a test of whether trading could be taught. Richard Dennis believed it could; William Eckhardt was more skeptical.

Ads were placed in The Wall Street Journal, Barron's, and The New York Times. According to a 1989 Wall Street Journal article, about 1,000 applicants answered each ad, 80 were interviewed, and Dennis chose 13 for the 1984 program and 10 for the 1985 program. By 1989, three had been dropped, leaving 20, of whom 16 were known to be trading public funds.

The same article says the group of 14 commodity-trading advisers it profiled had earned an average annual compound return of 80 percent over the prior 4.5 years.

That statistic is the reason Turtle Trading became legendary. The story was not just that the system worked. It was that a rules-based process could be taught to relative novices and still produce extraordinary results.

At a high level, Turtle Trading says: enter when price proves strength or weakness, hold until the trend is clearly broken, and size positions by volatility, not by conviction.

That is classic trend following. The system does not try to predict turning points. It waits for price to move beyond a defined channel and then rides the move if it persists.

The original rules document describes two entry systems: System 1, based on a 20-day breakout, and System 2, based on a 55-day breakout.

The breakout is the heart of the method because it is a way of operationalizing trend emergence.

A market that trades above its highest level of the last 20 or 55 days is doing something observable and objective. This is why Turtle Trading is best understood as a channel breakout system, often associated today with Donchian channels.

The insight is simple: most breakouts fail, but a small number turn into very large moves, and those few moves can dominate overall returns.

A lot of people reduce Turtle Trading to buy 20-day highs, but that misses the most important quantitative part: position sizing by volatility.

The Turtles used N, a volatility measure equivalent in spirit to ATR. The rules define true range and then compute N from a 20-day exponential process. Conceptually, N is the average daily range.

The point of this was to normalize risk across markets. A highly volatile market got a smaller position; a quieter market got a larger one.

This is one of the deepest insights in the Turtle rules.

The Turtles capped risk tightly. The rules state that no trade could incur more than 2 percent risk, and since 1N of movement represented 1 percent of account equity, stops were placed at 2N below entry for longs and above entry for shorts.

That is a very elegant construction. It makes risk explicit and comparable across instruments.

Another major reason the system could produce huge returns was that it added to winning positions.

The rules say additional units were added every half-N in the direction of the trade, up to a maximum number of units. So once a breakout started working, the system increased exposure into strength rather than taking quick profits.

This is one of the strongest parts of the method. Turtle Trading was not about being right often. It was about being big when right.

The exits were deliberately uncomfortable. For System 2, the rules specify an exit on a 20-day low for long positions and a 20-day high for short positions. That means the system often gave back a meaningful chunk of open profit before finally exiting.

That is psychologically difficult, but it is structurally necessary for long-trend capture. The rules themselves note that traders would sometimes watch 20, 40, even 100 percent of significant profits evaporate while waiting for the exit signal, but that discipline was required to stay in the biggest moves.

The best historical performance snapshot comes from the 1989 Wall Street Journal table titled A Turtle Race Worth Watching. It lists 14 Turtle-related advisers and shows an average annual compound return of 80 percent, versus 25.1 percent for the Barclay CTA Index and 19.2 percent for the S&P 500 total return over the comparison period.

Individual annual compound returns in the table ranged from 38.9 to 124.1 percent.

Those numbers are striking, but the more important insight is the structure behind them. Turtle-style systems generally have a low win rate combined with fat right tail winners and deep discipline.

So the system can look frustrating for long periods and still be very profitable over full cycles.

The real lesson of Turtle Trading is not that 20-day or 55-day breakouts are magic. It is that a profitable trading process can be decomposed into clear, teachable rules.

That was the genius of the Turtle system. It converted trend following from something vague and instinctive into something teachable and repeatable.

A better summary than buy breakouts is this: cut losses fast, size by volatility, add to winners, and hold trends longer than feels comfortable.

That is the true Turtle philosophy. It is a complete framework, not a single indicator.

Turtle Trading is not about channel breakouts being inherently magical. It is about showing that a complete, rules-based trading system — covering entries, exits, sizing, and risk — can be taught to novices and still produce extraordinary results.

The edge was never any single indicator. It was the combination: breakout entries to identify trends, volatility-based sizing to normalize risk, pyramiding to amplify winners, strict stops to contain losses, and trailing exits to ride big moves.

That decomposition of a trading edge into explicit, teachable rules is the most valuable lesson the Turtle experiment ever produced.