Turtle Trading Strategy: Rules, Statistics, and Backtests – Does It Still Work?

Last Updated on July 31, 2022 by Oddmund Groette

Turtle trading strategies are a trading approach named after Richard Dennis’ popular trading experiment in 1983. The experiment was to determine whether trading was an inborn skill or could be learned. The turtle trading experiment was a huge success, and the strategy employed by those traders was very successful all through the 1980s. But does it still work today? And what exactly are the turtle trading rules?

Yes, the turtle trading strategy still works today. It is a trend-following strategy, so it works in markets with clear trends. While the original strategy, which is based on identifying breakouts, still works reasonably well, traders have modified the turtle trading rules by using technical indicators for trend identification. The technique may not be as profitable as in the 1980s, but traders can still use it to earn good returns, as indicated by the Barclay CTA index and successful hedge funds, like Swedish Lynx, for example.

What is Turtle Trading – The Turtle experiment

What is turtle trading? The turtle trading experiment started as a bet:

Richard Dennis and his trading partner, William Eckhardt, conducted the turtle experiment to determine whether trading was an inborn skill or could be learned, following an argument they had. Dennis was of the opinion that anyone could be taught to trade the futures markets, but Eckhardt thought otherwise; he strongly believed that successful traders are born with a special talent that helps them to make massive gains from their trades.

To settle the debate between the duo, Dennis decided to conduct an experiment by recruiting and training some novice traders to ascertain the possibility of turning them into successful traders. This was the turtle experiment. They placed an advertisement and received applicants from aspirants to be trained.

Fourteen successful applicants were trained for two weeks. During the training, Dennis referred to his students as Turtles as a reminder of the turtle farms he had visited in Singapore and decided that he could grow traders as quickly and efficiently as farm-grown turtles.

After the training, Dennis believed strongly that they would become successful traders if only they adhered strictly to instructions. He funded their trading accounts with amounts ranging from $500,000 to $ 2,000,000. In the end, many of those traders were successful, and Dennis was proven right that trading skills can be learned. Following the experiment, the strategy Dennis taught his students became very popular and has been known as the Turtle Trading Strategy.

Who were the original Turtles?

The original turtles were random fellows Dennis selected for his experiment. They were normal fellows with no special trading skills or prior training.

For his experiment, Dennis placed an advert in the Wall Street Journal to get the original turtles, and thousands of novice traders applied for the training. He then carried out some selection processes to screen the applicants. Although no one knows the exact criteria that Dennis used to select the original turtles, it is believed that several true and false questions were used to sort through the thousands of applicants. Eventually, 14 of the applicants were accepted as the original turtles.

These original turtles were specially taught how to use hard-and-fast turtle trading rules. Dennis stressed the importance of adhering to instructions, assuring them they would become successful traders if they stuck to the instructions. According to Dennis, it’s one thing to understand a strategy and a different to implement the instructions.

Original Turtle trading rules

The original turtle trading strategy is essentially a trend-following strategy. The rules of the turtle trading system focused on these three key elements (original turtle trading rules):

  • Volatility-based position sizing methods
  • Pyramiding
  • Turtle trading exits and stops

Volatility-based position sizing methods

One important turtle trading rule of the turtle trading strategy was to vary their position sizes based on market volatility. Dennis taught them to use the average true range (ATR) indicator to calculate volatility and use this to vary their position size. They were to take larger positions in less volatile markets and lower their exposure in highly volatile markets.

The goal was to maintain equivalent risk per dollar because higher volatility indicates higher risk. The turtles could diversify their portfolio by investing in similar risk levels. Dennis taught the turtles how to calculate risk using a set of calculations to limit how much danger they may take.

Pyramiding – Turtles added to winners

The turtle trading rules also involved using the pyramiding technique rule to maximize profits on winning trades. That is, they added more positions when they are in a winning trade, using their increasing profits to carry the extra positions. They did this systematically so that the profits from earlier positions can more than cover the new positions. If the market remains favorable, they bag a lot of profits with a much bigger position.

The key to pyramiding success is to keep your risk-to-reward ratio low, which means you should never risk more than half of what you stand to gain or your reward. A winning trade can compound your profit if done correctly. However, recognizing which trades are ideal for pyramiding needs a lot of skill and insight.

Like other speculative activities, pyramiding employs leverage to increase the size of a stake. It is dangerous and can result in amplified gains or losses. While some hedge funds and private investors use this strategy, many do not have the resources. Furthermore, most hedge funds avoid taking such a high risk in a single trade. If you try to use pyramiding, you must be correct, or else the strength of leveraging will work against you.

Only employ pyramiding when the market is trending strongly, and make sure you have an exit strategy before you start trading. Resist the need to grow greedy and stick to your risk-mitigation strategy, always keeping the optimum risk-to-reward ratio in mind.

Turtle trading rules about exits and stops

The third important turtle trading rule is the use of stop-loss orders. The turtles were taught to determine ahead of time when to cut losses and move on and were mandated to always exit once the market reaches its predetermined stop price.

For a winning trade, the turtles were taught to exit when a breakout occurs in the opposite direction of their position. They understand that holding a few lost positions for an extended time eventually wipes out the profits from other trades, resulting in disaster. So, they have specific criteria for exiting a winning position.

Pyramiding, stops, and exits have easy to understand formula. However, they complicate backtesting slightly, especially if you are backtesting by using Excel.

Which markets did they trade?

The Turtles began by trading over two dozen instruments, including US bonds of various maturities, cotton, sugar, gold, coffee, crude oil, heating oil, gasoline, S& P 500 futures, silver, and a few currencies such as the Swiss Franc, French Franc, Deutschmark, British Pound, Eurodollar, and Japanese Yen. Some of these have been replaced by the Euro, such as the French Franc and the Deutschmark, so a modern basket would look a little different. The main issue with this setup is the number of funds required to trade many units.

What is turtle theory? What is the logic behind Turtle Trading?

The turtle theory is based on the fact that the market can stay in a trend for a prolonged period. In the market, it is believed that the trend, once established, remains until it is proven to have reversed. So, the logic behind the strategy is to identify the emergence of a new trend early, trade in the direction of the new trend, add more positions as the trend progresses, and finally exit from the trade once a new trend seems to emerge in the opposite direction.

Because no one knows when a trend starts, all turtle trading rules are based on purely mechanical rules. However, a big trend up or down will inevitably happen – it’s just a question of time – but in the meantime, you might get whipsawed.

To identify the emergence of a new uptrend, Dennis uses the breakout strategy. Specifically, the turtles used, for example, the breakout of a 20-day high to identify the beginning of a new uptrend and then enter a long position. Subsequent breakouts, such as the breakout of the 55-day high, were seen as opportunities to add more positions (pyramiding).

The trend is believed to have reversed if a breakdown of the 20-day low occurs. When that happens, they exit their positions or even look to go short and profit from the emerging downtrend. They aim to stay in the short position until another breakout of the 20-day high occurs.

So, the strategy is all about following the trend and using the breakout of the 20-day high or low to identify emerging trends. Other aspects of the strategy were just managing position sizing and risks.

What happened to the original Turtles?

Those turtles who followed the rules were the ones who survived the experiment. Unfortunately, not all of the turtles survived. After struggling to follow the rules Dennis had taught his turtles, some were asked to leave the experiment.

The most difficult aspect of following the rules for most turtles was the exit strategy, which required them to wait for a new low. This entailed watching as 20%, 50%, or even 100% of profits vanished. One turtle was released before the end of the first year because he did not follow the rules for the exit strategy.

Those who followed the rules and stayed in the experiment made large profits by basing their trades on the turtle trader rules. Many of them became very successful and went on to establish their own trading firms.

However, not every turtle was successful. One of the turtles, Curtis Faith, established his money management firm. It failed spectacularly, but it’s unclear how well Faith followed the turtle trader rules. On the other hand, Chesapeake Capital is still managed by Jerry Parker.

What happened to Richard Dennis?

What happened to Richard Dennis is an intriguing side note to the story of the Turtle Trading experiment. Dennis made his first million dollars before the age of 25. At the height of his trading success, he was dubbed the “Prince of the Pit.” Dennis made $80 million in 1986 alone. During this time, Dennis’s name was added to those of other industry titans such as George Soros and Michael Milken. But his success did not last.

Dennis’ strategy was always fraught with risk. On some days, Dennis could be millions of dollars in debt, but he believed the wins outweighed the losses. For a long time, they did. But there came a time when this was no longer the case. Dennis lost more than half of his assets between 1987 and 1988 when his turtles completed their five-year experiment. It’s debatable whether Dennis strictly adhered to his Turtle Trading system when he lost this money. Dennis stopped trading as a result of this loss. His name is now remembered far more for his Turtle Trading experiment than for his successful trading career.

What happened to William Eckhardt?

William Eckhardt became famous with the turtle trading experiment. He was Dennis’ trading partner, and they both worked on the experiment. While the experiment’s details have become legendary, its worth can be seen in the many highly successful trading firms it led to, as evidenced by our Top Traders of 2010.

By 1993, Eckhardt was ready to declare the experiment over and himself incorrect. Eckhardt admitted in an interview for Jack Schwager’s compelling book The New Market Wizards: “I assumed a trader added something that couldn’t be encapsulated in a mechanical program. I was proven wrong. The turtle trader program was a huge success,” he said.

Meanwhile, Eckhardt founded his commodity trading advisor (CTA) in 1991, and it has produced a compound annual return of 17.35 percent over the last 20 years, with a profit of 21.09 percent in 2010. In addition to developing trading systems, Eckhardt has created a trading science and written academic papers on the philosophy of science.

Turtle trading books

Many books and articles have been written about the turtle trading experiment and system. But one of the most popular ones is “The Complete TurtleTrader: How 23 Novice Investors Became Overnight Millionaires”. It was written by Michael W Covel and featured the true story behind Wall Street legend Richard Dennis, his disciples, the Turtles, and the trading techniques that made them millionaires.

Another great book is Curtis Faith’s Way of the Turtle. Despite failing spectacularly as a money manager and spending time in jail, we believe his book is a great read and is all you need to learn mechanical trading. As a matter of fact, we regard Curtis Faith’s book as a “must have” if you want to learn mechanical trading.

Turtle trading strategies

What are the turtle trading strategies? Curtis Faith gave us some clues in his brilliant book called The Way Of The Turtle. Mr. Faith mentioned six specific turtle trading system backtests in his book:

Curtis Faith backtested all the strategies in his book. The backtest period was from 1996 until the end of 2006. The six strategies performed like this:

  • ATR CBO: 49.5% CAGR
  • Bollinger CBO: 51.8% CAGR
  • Donchian Trend: 29.4% CAGR
  • Donchian Time: 57.2% CAGR
  • Dual MA: 57.8% CAGR
  • Triple MA: 48.1% CAGR

These are fantastic results, and further down in the post we test the same strategies from 2007 and onwards (we do our own turtle trading backtest).

MLM – Mt. Lucas Management Index

A famous index is the Mt. Lucas Management Index. It’s a simplistic, yet very effective trend-following index. The idea behind the index is to measure trend-following, thus indirectly turtle trading strategies, and the index is enormously simple:

It’s a 200-day moving average strategy! It tracks 25 futures contracts. When the price of the contract is above the 200-day moving average at the end of the month, a long position is held until the end of next month. If the price is below the average, a short position is held for one month. It can’t get any simpler, and that’s the beauty of it.

We have discussed the MLM index in more detail in our post about does trend following trading strategies work?

Systematic Turtle Trading – low correlation to stocks

There are several indices that could be labeled trend-following or systematic trading. One of them is Barclay CTA Index which has a track record back to 1980. The index is a mix of different strategies, but the majority are typical turtle trading strategies.

In the table below you can see the performance of Barclay CTA Index together with the hedge fund Lynx. Lynch is a Swedish systematic trend-following hedge fund that was set up in 1999.

Barclay CTA Index S&P 500 (dividends reinvested) Lynx
1980 63.69 31.74
1981 23.9 -4.7
1982 16.68 20.42
1983 23.75 22.34
1984 8.74 6.15
1985 25.5 31.24
1986 3.82 18.49
1987 57.27 5.81
1988 21.76 16.54
1989 1.8 31.48
1990 21.02 -3.06
1991 3.73 30.23
1992 -0.91 7.49
1993 10.37 9.97
1994 -0.65 1.33
1995 13.64 37.2
1996 9.12 22.68
1997 10.89 33.1
1998 7.01 28.34
1999 -1.19 20.89
2000 7.86 -9.03 15.33
2001 0.84 -11.85 18.68
2002 12.36 -21.97 31.58
2003 8.69 28.36 12.61
2004 3.3 10.74 8.31
2005 1.71 4.83 9
2006 3.54 15.61 15.3
2007 7.64 5.48 38.24
2008 14.09 -36.55 -9.43
2009 -0.1 25.94 19.18
2010 7.05 14.82 -2.24
2011 -3.09 2.1 -6.73
2012 -1.7 15.89 11.11
2013 -1.42 32.15 27.03
2014 7.61 13.52 -8.73
2015 -1.5 1.38 -3.29
2016 -1.23 11.77 -4.05
2017 0.7 21.61 0.35
2018 -3.17 -4.23 18.36
2019 5.17 31.21 7.78
2020 5.43 18.02 1.29
2021 5.04 28.47 28.78
2022 8.1 -23.41 40.81
Correlation to S&P 500 0.05938063   -0.0716767063

 

Correlation in trading is important because you want to own assets that are uncorrelated to each other. And, as you can see in the table above, both Barclay CTA Index and Lynch do a great job in achieving that. Lynx even has a slightly negative correlation to stocks. We have written in previous articles why correlation is such an important factor in investing and trading:

Turtle trading in bitcoin and crypto

Up until 2022, many of these turtle trading systems performed well in bitcoin and other cryptos.

However, cryptocurrencies have become more and more risk on/off assets and thus trends have weakened (at least in our backtests). That is not to say turtle trading doesn’t work on crypto, but we assume it gets a bit more difficult as the market “matures”.

Backtests of turtle trading strategies

We’ll end the article by doing some backtests on a basket of continuous futures contracts. The backtest period is from 2007 until today and we include the following contracts that we downloaded from the database in Tradestation:

  • Australian dollar
  • British pound
  • Corn
  • Cocoa
  • Canadian dollar
  • Crude oil
  • Cotton
  • Euro
  • Eurodollar
  • Feeder cattle
  • Gold
  • Copper
  • Heating oil
  • Unleaded gas
  • Japanese yen
  • Coffee
  • Cattle
  • Hogs
  • Mexican peso
  • Natural gas
  • Soybeans
  • Sugar
  • Swiss franc
  • Silver
  • Treasury notes
  • Treasury bonds
  • Wheat

The data comes from Tradestation, but we performed the backtests in Amibroker. The backtests are done with no leverage, no position sizing, no stop losses, and no profit targets. We also excluded “false signals” where during market ranges the price might trigger a breakout by a single tick (which can hardly be labeled a breakout).

Also, we have ignored short signals. We only take long signals. Short selling is difficult, and thus we ignore it in this backtest.

Turtle Trading strategy backtest 1: Close higher than 6 months ago (momentum)

Trading rules (monthly bars):

  • If the close is higher than 6 months ago, buy and hold the position for one month.
  • If the close is lower than the close 6 months ago, sell and stay out for the coming month.
  • Rinse and repeat monthly.

The equity curve looks like this:

The strategy performance metrics are these:

Turtle Trading strategy backtest 2: Dual moving average

The Dual Moving Average strategy is a simple strategy that can be backtested in just a few minutes. The strategy, in plain English, works like this:

  • Buy when the 100-day moving average crosses above the 350-day moving average.
  • Sell when the 100-day moving average crosses below the 350-day moving average.

Two indicators turned into one indicator. This is all there is to it. The equity curve looks like this:

The strategy performance metrics are these:

  • CAGR: 3%
  • Max drawdown: 41%
  • Risk-adjusted return: 6.5%

Turtle Trading strategy backtest 3: ATR Channel breakout

The ATR channel breakout uses the following rules in plain English:

  • Buy when the price breaks above the 350-day moving average plus a seven-day ATR.
  • Sell when the price breaks below the 350-day moving average deducted a seven-day ATR.

This is the equity curve:

The strategy performance metrics are these:

  • CAGR: 5.3%
  • Max drawdown: 30%
  • Risk-adjusted return: 11.25%

Turtle Trading strategy backtest 4: Bollinger Band breakout

The Bollinger Band Breakout strategy uses the following rules in plain English (we have a separate article about Bollinger Bands):

  • Buy when the price breaks above the 350-day moving average plus a standard deviation of 2.5.
  • Sell when the price breaks below the 350-day moving average deducted a seven-day ATR.

The equity curve is rather erratic but has taken off recently:

The strategy performance metrics are these:

  • CAGR: 2.1%
  • Max drawdown: 17%
  • Risk-adjusted return: 7.5%

Turtle trading strategies backtest – summary

The performance has been lower from 2007 than it was in prior periods. This could be due to a number of reasons:

  • Quantitative easing after 2008. In times of monetary expansion trend following and turtle trading strategies have performed poorly.
  • The strategies are too simplistic.
  • We have avoided very important risk parameters like position sizing, stop-losses, and profit targets.
  • And surely some other aspects we have not thought of.

Turtle Trading strategies – summary and conclusions

The turtle trading experiment has turned out to be immensely popular and well known within the trading community, and perhaps for good reason. It indicates that trading can be learned and most likely is not a skill you are born with. Many hedge funds still employ the turtle trading rules one way or the other with great success.

The equity curves we provided show pretty clearly that turtle trading rules only work for the long term, but there are indications that the systems are not as effective as they once were (?). What is typical for trend-following and turtle trading strategies is that there are many whipsaws and thus a low win rate and the potential returns happen only in a relatively short period of time. Periods of drawdown are often long and this can be detrimental to your belief in turtle trading. As usual, what seems so easy on paper and in backtests, is not so easy when you employ real money!

Similar Posts