FAMOUS STRATEGIES

Turtle Trading

Richard Dennis & William Eckhardt · 1980s. A rules-based breakout experiment: enter on new highs or lows over a fixed lookback window, size positions from market volatility, and exit on an opposing breakout.

← All famous strategies

Classification: published rulebook. Rulebook: the Turtle entry, exit, sizing and stop rules were later published in detail, so the system can be described precisely and tested. The entry, exit and stop-loss sections below describe documented rules.

ORIGIN

Where it comes from

In the early 1980s, commodity trader Richard Dennis and his partner William Eckhardt disagreed about whether great traders are born or can be taught. Dennis recruited and trained a group of novices, nicknamed the Turtles, in 1983–84, gave them a written rule set and let them trade real capital. The rules were later made public by former participants, most notably Curtis Faith.

Core principles

Follow a written system and remove discretion from entries and exits.

Trade breakouts of recent highs and lows, and let trends come to you.

Size every position from market volatility so risk is comparable across markets.

Accept many small losses in exchange for a few large trend winners.

Diversify across many markets so that some trends are always in play.

AT A GLANCE

RULEBOOK

Richard Dennis & William Eckhardt

1980s

Well documented and publishable as a rule set. You can write it down, test it and follow it step by step.

Mechanical rules over discretion

HOW IT WORKS

The method in plain terms.

A documented rule set, summarised for education.

The Turtles ran two breakout systems side by side. System 1 used a shorter 20-day channel and System 2 a longer 55-day channel. Both are close cousins of a Donchian channel: the highest high and lowest low over a lookback window.

Position size came from "N", a measure of average daily volatility (a 20-day average of true range). One "unit" was sized so that a one-N move equalled roughly 1% of account equity, which made a unit in soybeans comparable in risk to a unit in gold or a currency.

Winners could be added to in steps (pyramiding) as the trend extended, inside strict limits on units per market, per correlated group and per direction.

ENTRY

Entry rules

System 1: go long on a break above the highest high of the previous 20 days (short on a break below the 20-day low).

System 1 skip rule: ignore the signal if the previous System 1 breakout in that market would have been a winner. System 2's 55-day breakout then acts as the fail-safe entry.

System 2: enter on any 55-day breakout in either direction, with no skip filter.

Add one more unit each time price moves a further half N in your favour, up to four units in one market.

Trade only liquid markets, spread across several sectors.

EXIT

Exit rules

System 1 exit: close a long position when price falls below the lowest low of the previous 10 days (for a short, when it rises above the 10-day high).

System 2 exit: the same idea using a 20-day low (long) or 20-day high (short).

Exit the whole position when the exit signal fires. Do not hold on hoping for a bounce.

PROTECTION

Stop-loss rules

Place the initial protective stop 2N from the entry price, so one unit risks about 2% of equity at most.

When units are added, the stop for the whole position moves so that it stays 2N from the most recent entry.

A stop is honoured every time. The exit channel can end a trade earlier than the stop, but never later.

RISK

Risk management

Risk is built into unit size: one unit is sized from N, and a 2N stop therefore risks roughly 2% of equity per unit.

Original exposure limits: four units in one market, six in closely correlated markets, ten in loosely correlated markets and twelve in one direction.

Expect long losing streaks and deep drawdowns. The edge only shows over many trades, so position size must let you survive the bad stretches.

The original rules scaled down the notional account size used for sizing after a drawdown, to shrink risk while the system was losing.

WORKED EXAMPLES

One that worked, one that failed.

Both charts are schematic drawings of the idea. They are illustrative scenarios, not recorded trades or real market data.

SUCCESSFUL EXAMPLE

Illustrative winner: a sustained trend

A liquid futures market has ranged for weeks, then breaks above its 55-day high and trends for months with only shallow pullbacks.

Entry
Buy the 55-day breakout with one unit sized from N, then add units every half N as the trend extends (maximum four).

Exit
Exit when price finally falls below the 20-day low, giving back part of the open profit.

Stop-loss
The stop trails 2N below the latest entry and is never touched.

Lesson
One large trend can pay for a long string of small losses, but only if you take every signal.

FAILED EXAMPLE

Illustrative loser: a false breakout

In a choppy, range-bound market price pokes through the 20-day high, then reverses almost immediately.

Entry
Buy the 20-day breakout with a single unit.

Exit
Stopped out when price falls 2N below entry, before the exit channel is reached.

Stop-loss
The 2N stop keeps the loss to roughly 2% of equity for that unit.

Lesson
Small, planned losses are the cost of catching trends; the mistake is skipping the stop, not taking the loss.

WHEN IT FITS

Suitable market conditions

Markets that trend for weeks or months, such as commodities, currencies, interest rates and stock indices.

Liquid instruments where breakouts can be traded without heavy slippage.

A portfolio of many markets, so that at least a few trends appear each year.

WHERE IT STRUGGLES

Limitations

Sideways markets produce whipsaws and long strings of small losses.

A meaningful share of open profit is usually given back before the exit signal fires.

The rules are public and widely known, so the original results should not be assumed to repeat.

Volatility-based sizing across many markets needs adequate capital and market access.

AVOID

Common mistakes

Skipping signals after a few losses, when the next trade may be the one that makes the year.

Sizing by gut feel instead of from N.

Removing or widening the stop once a trade goes against you.

Ignoring correlation limits and stacking several bets on the same theme.

Judging the system on a handful of trades.

PRACTICE

Practice checklist

☐ Write the exact entry, exit and stop rules on one page before opening a chart.

☐ Calculate N and the unit size for one market on paper, and confirm that a 2N stop matches your risk limit.

☐ Log every breakout signal, including the ones the skip rule would filter out.

☐ Record correlation and total open risk each week.

☐ Review at least 30 signals before drawing any conclusion.

Frequently asked questions

Are the Turtle rules public?
Yes. The rule set was released publicly in the early 2000s and is described in books by former participants. Because it is widely known, treat it as a teaching example rather than a guaranteed edge.

Can the system be applied to stocks or forex?
The original system was built for futures. Its ideas (breakouts, volatility sizing, fixed stops) are used elsewhere, but any adaptation needs its own testing.

What is N?
A volatility measure: the average true range over about 20 days. It is used to size positions and to place stops.

Does TCT teach or sell this system?
No. TCT lists it as historical, educational reference only and does not recommend trading it.

SOURCES

Where to read more.

Plain-text references for further reading. Check specifics against the original material before relying on them.

Curtis Faith — Way of the Turtle (2007)

Curtis Faith — The Original Turtle Trading Rules (rule set published online by a former Turtle, early 2000s)

Michael Covel — The Complete TurtleTrader (2007)

Educational reference only. TCT does not teach, sell or recommend this strategy. Descriptions are summaries of publicly known approaches, the examples are illustrative, and past performance of any trader or system does not indicate future results. Trading involves risk of loss.