Turtle Trading Rules

Chapter 18 Turtle Trading System (1)
Simpler.A simple timed exit method done well will always beat out those complex, fancy fantasy methods.

This chapter introduces some turtle trading systems, commonly known as long-term trend-following systems.They are:

ATR Channel Breakout System: A volatility channel system that uses ATR as a volatility indicator.

Bollinger breakout: A volatility channel system whose volatility indicator is the standard deviation.

Donchian Trend System: A breakout system with a trend filter.

Donchian trend with time exit: A breakout system with a trend filter that uses a timed exit strategy.

Dual moving average system (dual moving average): A system that buys or sells when the short-term moving average crosses the longer-term moving average.Unlike other systems, this one stays with the market, whether long or short.

Triple moving average system (triple moving average): This system also buys or sells when the short-term moving average crosses the longer-term moving average, but only if the crossing direction is in line with the general trend (judged by one of the longest-term moving averages).

To examine the difference between these systems, I ran historical simulations of each to see how much money each system made over the past 10 years.This chapter compares the relative performance of these systems using some of the criteria discussed in this chapter.

history test
Historical testing is also known as post-hoc testing.There are many traders, including quite a few successful ones, who do not believe in the test of history.They believe that it is meaningless to use historical data as a test, because the past does not represent the future.For those readers who are not familiar with this debate, I will spend a little time explaining to you something that you may not think needs to be explained.For those of you who don't believe in historical testing, here are a few questions: Do you have another method?How do you develop a strategy without knowing the past?How do you decide when to buy or sell?Are you just guessing?
The only information you can have is how the market has performed so far.Even if you are a random trader, not using any rules or systems, you use past experience as a guide.You're relying on an interpretation of the past; in fact, you're relying on historical data.

Smart self-directed traders may develop their own systems after years of practice.They observe recurring patterns that harbor profitable opportunities and devise trading strategies to exploit these opportunities.Before entering the market, novices often spend a few months studying the past trend charts to understand what the past market was like.Because they know that the best guide to future market trends is contained in history.

It is not hard to show that a computer can test an idea more effectively with the same historical data.With computer simulation testing, traders can conduct a more rigorous examination of a specific strategy before taking action in the real world.They often find that, due to unforeseen factors, seemingly promising concepts are not valid.Obviously, it is much better to use a computer to discover this early than to learn a lot of wisdom in actual combat.

Some traders do not believe in historical testing because such post-hoc testing is often misrepresented.The problem can easily arise of borrowing the power of a computer to discover a method that seems to work, only to fail miserably in the real market.These problems can be overcome, as long as you can avoid the most common hidden danger: excessive optimization (overoptimization).will discuss this issue.

Proper historical testing requires certain experiences and skills that novices do not possess.However, while a sharp knife cannot be handed over to a child, it can come in handy in the kitchen.You just have to be careful with those sharp tools.

A history test cannot predict the future, but it can help you determine whether an approach will be profitable in the future.It's not optimal, a crystal ball or a time machine are better than it, but it's one of the best tools available.

deified expert

The misguided expert advice of "don't optimize" stems from a phenomenon in the trading world that my friends and I like to call the apotheosis of the experts.Unfortunately, in most fields, true experts are few and far between.Every real expert is drowned in a mass of fake experts.These fake experts occupy a place in the industry, cobble together a lot of knowledge, and in the eyes of laymen, they are indistinguishable from real experts.They have a role too, but they don't really understand the field they're in.

True experts don't have hard-and-fast rules; they know what they're doing, so they don't need rigid rules.

But fake experts don't know how to do it, so they like to observe experts and imitate experts.They know how to do it, but they don't know why.So none of the strict laws they created in the guise of the experts was what the experts intended.

Counterfeit experts have one telltale sign: their writing is obscure and incomprehensible.I can't write clearly because I can't think clearly.A real expert can explain complex concepts clearly and easily.

Another common trait among counterfeiters is that they know how to apply complex procedures and techniques and are well trained, but they are unaware of the limitations of those techniques.

When it comes to the trading industry, history check is a good example.Someone who can do complex statistical analysis does a simulated test, generates 1000 trades, and then thinks he can draw conclusions from these transactions, but he does not realize that such conclusions may only be drawn from short-term data of two weeks of.This type of person can use math, but doesn't understand a simple truth: if next week's situation is very different from the last two weeks, math is useless.

Don't confuse experience with expert skill, or knowledge with wisdom.

Test parameters
The tests described in this chapter use a common market combination and a common money management law in order to clarify differences in the test results from differences caused by changes in the rules.The following are the parameters used in the tests.

market mix
The market combinations we tested included the Australian dollar, British pound, grains, cocoa, Canadian dollar, crude oil, cotton, euro, Eurodollar, feeder cattle, gold, copper, heating oil, unleaded gasoline, Japanese yen, Coffee, cattle, hogs, Mexican pesos, natural gas, soybeans, sugar, Swiss francs, silver, treasuries, treasuries, and wheat markets.

These markets were selected from highly liquid (highly traded) markets in the United States.Several highly liquid markets were also excluded due to their high correlation with other liquid markets.We limited our testing to the U.S. market because many historical data providers provide only sporadic information about foreign markets - for this reason, many new traders only start out in the U.S. market, and we want to It is convenient for traders to apply our test results to their own trading.

money management rules
The money management rules used here are the same as the Turtles' method, but a little more conservative in one indicator.The Turtles made 1 ATR equal to 1% of trading capital, while our standard is 0.5%.That is, when calculating the number of contracts for a given market, we divide 0.5% of the account equity by the ATR value of this market at the time of the transaction.

test period

This test uses data from January 1996 to June 1, which is true for all systems.

six systems
Before revealing the test results, let's take a closer look at the various systems.

ATR Channel Breakout System
The ATR channel breakout system is a volatility channel system that uses the average true volatility (ie ATR) as a volatility indicator. The 350-day moving average closing price plus 7 ATRs is the top of the channel, and minus 3 ATRs is the bottom of the channel.Go long on today's open if the previous day's close crossed the top of the channel, or short on the open if the previous day's close fell below the bottom of the channel.When the closing price crosses the moving average in the opposite direction, traders exit.

Trader Mark Johnson popularized a system called PGO (Pretty Good Oscillator) on Chuck LeBeau's System Traders Club Forum (www.traderclub.com) and here is A variant of the ATR channel breakout system.It is also a variation of the Bollinger Breakout System described below.Figure 10-1 is an amplitude channel of the ATR channel breakout system.

The middle curve is the 350-day moving average, and the top curve is the top of the volatility channel obtained by adding 350ATR to the 7-day moving average.

Bollinger Breakout System

This system was developed by Chuck LeBeau and David Lucas in their 1992 book "The Technical Trader's Guide to Computer Analysis of the Futures Markets" (Criteria for Calculating Moving Average Times and Calculating Swing Widths difference has changed).Bollinger bands are a type of volatility channel invented by John Bollinger.The system's Bollinger Bands are derived from the 350-day moving average closing price plus or minus 2.5 standard deviations.Go long on the open if the previous day's close crossed the top of the channel, or short on the open if the previous day's close fell below the bottom of the channel.Figure 10-2 shows the Volatility Channel of the Bollinger Breakout System.

Donchian Trend System

We have been talking about the Donchian trend system, which is a simplified version of the Turtle system back then.It uses a 20-day breakout entry strategy, a 10-day breakout exit strategy, and a 350-day/25-day exponential moving average trend filter.Traders strictly abide by the direction indicated by the short-term moving average: if the 25-day moving average is above the 350-day moving average, they can only go long; if the 25-day moving average is below the 350-day moving average, they can only go short.This system also stipulates a stop loss exit point of 2ATR, which is the same as the original turtle system.Figure 10-3 shows the breakout levels and moving averages of the Donchian trend system.

(End of this chapter)