Developing a Profitable Long Term Trading System

by: Charles Tanti B.Sc.(Eng).

Prelude: If it was possible to create a profitable trading system using conventional indicators, like the MACD, SMA, DEMA, RSI etc., then everybody cab be rich. Coventional mathematics do not work either. In order to create the ultimate trading system you have to go beyond the obvious. You have to be able to create your own unique custom indicators.

Smart Traders Think and Trade Long Term
(All profit and test results are hypothetical)

What is a trading system? A trading system is not the "Holy Grail". Most traders expect too much from a trading system. If a system makes a few losing trades they usually give up just when the market is about to turn.

A good trading system is a tool to turn the odds in your favor, by employing a proven trading methodology to help keep your emotions from interfering with your trading.

To be successful, a trader must overcome hope, fear and greed and learn to be patient. The market is always right. A trader needs to be aware of the direction of the market. He must be able to analyze and interpret the information in the charts correctly and objectively without any emotion. Risk is a part of trading, and it cannot be avoided or eliminated. The successful trader knows this and accepts risk as part of the business of trading. However, he also has the knowledge and confidence that risk can be controlled.

To trade a system profitably, one must have the discipline to execute the signals generated. Discipline comes from confidence, and confidence comes from knowledge. Therefore, it is important to understand how your system works. Confidence and discipline also come from experience. A trader must have the patience to observe a trading system in real time to see how and where signals are generated.

Every trader knows that in trading TIMING IS EVERYTHING. If you can time your trades perfectly, you will never have to worry about a drawdown. But, as everyone knows, it is not as simple as this. No matter how good you are at technical analysis, most of the time the market is confusing:

Is it going: Up, Down, or Sideways.

Most traders, especially day traders, look at the market too closely. If you look at the market on a day-to-day basis, you will lose sight of the big picture. In other words, you will get confused about the long-term trend.

We all know the saying, "THE TREND IS YOUR FRIEND". There is also another saying, "A FRIEND IN NEED IS A FRIEND INDEED". As a trader, your only "FRIEND" is the "TREND". As long as you trade with the major trend, you should not be too concerned about timing. The "TREND" will always come to your rescue. As long as you stay with it, and as long as you trade in the direction of the major trend.

Trading is like a business. To succeed you need to:

  1. Be objective.
  2. Be well capitalized.
  3. Control your emotions.
  4. Have a long-term plan.
  5. Overcome hope, fear and greed.
  6. Turn the odds in your favor.

You need a Trading System.


The key components of a Profitable Trading System are:

  1. Choosing a trending market.
  2. Getting the right price data to reduce noise.
  3. Using a reliable long-term indicator to find the major trend.
  4. Applying simple filters to stay with the trend.
  5. Using more filtering algorithms to improve the performance.
  6. Incorporating genetic algorithms to sell near the top and buy near the bottom.
  7. Using functions to control risk and reduce the number of losing trades.

1. Choosing a trending market.

A long-term trading system needs a market that has a tendency to trend for long periods of time. The trend must be evident on daily and weekly charts. Examples are the T-Bonds and T-Notes markets, as well as Currencies. These markets are driven by interest rates, which in turn are driven by the state of the economy. These forces do not turn on a dime. They build momentum and move in the same direction for months or years at a time.

2. Getting the right price data to reduce noise.

The data you feed into the system is very important. Daily price data has a lot of noise, which in turn can produce false reversals that can generate false trading signals, which are very hard to filter out. On the other hand, weekly data is smoother and makes the major trend more evident. Weekly data will have less noise in it and is much easier to analyse.

3. Using a reliable Long-Term indicator to find the major trend.

Indicators are a mathematical way to measure the state and strength of the market. I have been developing and creating indicators for my systems for the last 30 years. During this time, I have found out that simple works best.

Stochastics work similar to moving averages, with one big advantage. A stochastic will show the trend of the market and also indicate when a market has become over bought or over sold. The elements of the stochastic can be applied to generate better BUY and SELL signals, It is possible that by using a stochastic together with other supporting indicators to SELL at the TOP and BUY at the BOTTOM on a consistent basis.

The ATS-3200 was built in 4 stages:

  1. Applying an 11 week stochastic "KC" to weekly data.
  2. Adding a simple "F" filter to filter out minor reversals.
  3. Adding the "V" filtering algorithms to to stay with the trend.
  4. Applying the "X" and "L" genetic algorithms to the ATS-3200.

You can see these on the detail report, week by week.

Stage 1: Applying an 11 week stochastic "KC" to weekly data.


          Trades Summary:
    No of winning trades: 133  
     No of losing trades: 128 - almost 50/50 
      Total No of trades: 281

          Average Profit: $2682  
            Average Loss: $2145 
           Average Trade: $ 314 / trade.
           Yearly Profit: $2564 / year.
           Total Profits: $82,187

        Biggest drawdown: $11,781 - 1999

Adding a simple adaptive filter.

No matter how good the indicator is, it will have false signals generated by minor trend reversals. A good system must be able to filter out these minor reversals, otherwise most of the profits will be dissipated. Markets are dynamic. They can become more volatile or move faster. A good filter should be adaptive to changing market conditions.

Stage 2: Adding a simple "F" filter to filter out minor reversals.


          Trades Summary:
    No of winning trades: 123  
     No of losing trades:  76 
      Total No of trades: 199 

          Average Profit: $2846  
            Average Loss: $2787
           Average Trade: $ 701 / trade.
           Yearly Profit: $4355 / year.
           Total Profits: $139,587 

        Biggest drawdown: $14,813 - 2010.

Using more filtering algorithms to improve the performance.

The performance of a trading system can be vastly improved by applying filtering algorithms to the long term indicator, in this case a long term stochastic. For example, the stochastic may go from "-" to "+", i.e., from negative(-) to positive(+) while the system is in a short trade. In this case, it will signal that the trend has changed. However, the market may still be moving lower. The close will be lower or the low will be lower. Therefore, even though there was a sign change in the stochastic from negative(-) to positive(+), since the market actually moved lower the filtering algorithm will filter this false signal out and keep the system short in the direction of the trend.

A number of these simple yet powerful filtering algorithms are incorporated into the ATS-3200 to improve its performance. In fact, both the ATS-3200 and the ATS-6400 each have nine such filtering algorithms called the "V-Filters".

Stage 3. Adding the "V" filtering algorithms to to stay with the trend.

        
          Trades Summary:
    No of winning trades: 119
     No of losing trades:  65
      Total No of trades: 184

          Average Profit: $3245  
            Average Loss: $2855
           Average Trade: $1089 / trade.
           Yearly Profit: $6257 / year.
           Total Profits: $200,546

        Biggest drawdown: $11,281 - 2012.

Incorporating genetic algorithms to SELL near the top and BUY near the bottom.

A long-term trading system usually gets the signal for the trend from either a moving average or a stochastic. These indicators always lag the market, and unless the market keeps the trend for very long periods of time and there is a substantial price difference between the top and the bottom of the market, it is very difficult for a system to produce good profits per trade. This is because a slow moving average or a long term stochastic normally misses up to $3000 from the top and the bottom, i.e. $6000 in total profits lost.

For a system to be truly profitable, it must be able to signal when a market is just going to turn around. Therefore, the system must be able to go SHORT at or near a market top, and also be able to go LONG at or near a market bottom.

This is possible by designing and creating genetic algorithms that track several indicators at the same time inside the system to signal an over bought or an over sold market. Then take the signal to go SHORT or LONG from several of these indicators at the same time to improve the odds of winning.

Since commodity markets cannot go up or down forever, they inevitably become over bought or over sold. A good algorithm will track the main indicator, but will only generate the BUY or SELL signal after a signal has been verified by several other built in indicators to produce more reliable and accurate results. Both the "ATS" systems have 15 of these genetic algorithms. These are called the "X" algorithms and the "L" algorithms. There are six "X" algorithms, these generate the SELL signals, and there are nine "L" algorithms, these generate the BUY signals.

Stage 4: Applying the "X" and "L" genetic algorithms to the ATS-3200.


          Trades Summary:
    No of winning trades: 115
     No of losing trades:  48
      Total No of trades: 163

          Average Profit: $5907  
            Average Loss: $2373 
           Average Trade: $3469 / trade. 

           Yearly Profit: $17,643 / year.
           Total Profits: $565,468
        Secondary Trades: $475,589 - from 159 trades.
    Average trade profit: $2991 - secondary trades.


        Biggest drawdown: $11,938 - 2014.

Note: Notice also the number of secondary trades and profit generated by the algorithms.
Secondary trades can only be generated by the algorithms.

The strength of the ATS-3200 comes from the genetic algorithms built into it. There are 2 set of algorithms. The "X" algorithms scan the indicators for over bought market conditions to go SHORT. The "L" algorithms scan the indicators for over sold market conditions to go LONG. Each algorithm checks several indicators for more accurate BUY or SELL signals.

These algorithms have been going SHORT at the TOP and LONG at the BOTTOM every year since the system was first developed in 1993. The prove of this is, that it is consistently in Futures Truth Top-Ten Tables year after year.

Performance Summary

ATS-3200 Stage #1 Stage #2 Stage #3 Stage #4 Stage:4/Stage:3
Total No of trades: 281 199 184 163 N/A
No of winning trades: 133 123 119 115 N/A
No of losing trades: 128 76 65 48 N/A
Average Trade: $314 $701 $1089 $3469 +219%
Average Profit: $2682 $2846 $3245 $5907 +82%
Average Loss: $2145 $2787 $2855 $2373 -17%
Biggest drawdown: $11,781 $14,813 $11,281 $11,938 +.06%
Total Profits: $82,187 $139,587 $200,546 $565,468 +182%
Yearly Profit: $2564 $4355 $6257 $17,643 +182%
Secondary Trades: - - - $475,589 N/A
System Ratios Stage #1 Stage #2 Stage #3 Stage #4 Stage:4/Stage:3
Winners/Losers: 1.04 1.62 1.83 2.40 +31%
Profit/Loss: 1.25 1.02 1.14 2.49 +118%
Profit Factor: 1.30 1.65 2.09 5.98 +186%
Notice how with every stage the ratios are improved.

For a system to be profitable, it must:

  1. Either produce twice the number of winning trades to the number of losing trades, or
  2. The average profit is equal to twice the average loss for an overall profit factor of 2.

In the final stage after the genetic algorithms are applied to the system
the Profit Factor of the ATS-3200 is almost 6.00.

The ATS-6400 works the same way as the ATS-3200.

Functions to control risk, and reduce the number of losing trades.

A good system must have functions to manage a trade after it is started.
Both the "ATS" systems have the following built in functions:

  • The "Stop Loss" function. This function calculates and sets the stop when a new trade is started. It is an adaptive function and is controlled by market volatility.
  • The "Moving Stop" function. This function moves the stop, once a trade is making sufficient profit, and prevents many profitable trades from turning into losing trades, therefore reduces the number of losing trades and protects profits.
  • The "Drawdown Parameter" function. This function takes the system out of the market when a trade starts to lose money, and at the same time the market has turned against the trade. This function too reduces the number of losing trades.
  • The "Fail Safe" function. This function prevents the system from entering into a trade if the chances of making a profit are not very good. For example if the system gets a signal from the stochastic to go long, but at the same time the market is already over bought, this function will prevent the system from taking the trade. it will also keep the system out of market until there is a change of direction.

I have an online seminar that you can do to understand how these principles work.
Please go to my web site to download the FREE TRIAL and register for the seminar.

If you need help, just send me an email or I can call you.
It is important that you understand how to use the system correctly.

About the developer

I first started trading the Bonds in May of 1985. I realized very early that if I was to succeed, I had to learn and study the market. My first introduction to technical analysis was to start plotting daily and weekly price charts manually. I also started keeping a journal of all my trades, and recorded all my observations about the market. By studying the charts I could see how certain patterns kept repeating themselves. It was as if the Bonds traded with some mathematical order. Many times I was able to project a turning point to within a few ticks.

After years of trading I formulated a set of complex rules for trading the Bonds. I also acquired an uncanny feel for the market from plotting manually my charts for many years. However even though I had all this, success in trading still eluded me. I lacked the objectivity needed to win consistently. I traded from 1985 to 1992 with limited success. The problem was too much emotion during trading, afraid to pull the trigger, forgetting important rules, or making the wrong interpretation of the market.

Finally in 1992 I started writing the ATS-3200 Trading System. I turned all my trading rules into indicators. By turning everything into numbers, I could then write computer programmes to find patterns in the indicators that consistently signaled either an over bought market condition or an over sold market condition. These rules then became the adaptive genetic algorithms of the System.

My first year of trading was very successful. After my first success with the System, I kept improving it, and during this time I too had to acquire the discipline to follow my own System.

All the System does is shift the odds in your favor. There is no holy grail. Only by following and trading the System can you hope to succeed.


Charles J. Tanti: I was born in Malta and studied Mechanical Engineering at the University of Malta.
My favorite subject was pure mathematics, because I liked solving very difficult problems.

When I lived in Malta, one of my favorite pastimes was to go to the Casino and play blackjack. I figured (no computers then) that as the cards were dealt from the deck the odds changed, and if you could keep track you had a better chance of winning.

When I discovered computers in 1983 the first thing I did was to write a professional blackjack program to teach players the card count and basic strategy. I proved to myself that if blackjack was played correctly, it was impossible to lose.

In 1980 I discovered commodity trading, through one of my customers who was a broker. I started by trading some gold and after losing $5000 in one week, I went back to my normal business. In 1985 I had more time and this time I went back to trading, but this time I took time to study the markets. I needed something that was volatile, so my broker suggested I trade T-Bonds. I decided then and there to specialize in just one market, and since 1985 I have only traded T-Bonds.

At first I was a day trader, trying to make $500 to $1000 per week. I soon found out that it was not easy. No matter how much studying I would do, the market always managed to spook me, and I would lose all the money I made, a few days later. I studied technical books and read the trade magazines, and one thing kept emerging: the professionals were all advising traders to think long term if they wanted to be successful traders.

Then one day I switched from daily charts to weekly charts and never looked back. As soon as I did this everything worked better, my indicators, the trend was more evident, the data had less noise, and the average profit per trade on my primitive system then, jumped from $250 per trade to over $1200 per trade. Trading much less, saving on commissions and bigger profits, what else can a trader ask for.

In 1992 I started writing the ATS-3200. I had the first version of the system ready to be traded in June of 1993. The system was long, but I decided to wait for the next "SELL" signal to make my first trade. It came on September 3rd, at 119-31, an all time high for the bonds. I did not have the guts to go short on that first signal, but when I saw the bonds go down $2000 in the next 2 weeks, I was amazed. The market rallied again and the system again went short at the very top at 121-30 on October 15th. Note: The chart is adjusted by the roll over amount to keep the data stationary.

On October 28th I made my first trade with the system and went short at 117.22. I made $1124.64 on a mini bond, and during my first six months of trading I made almost 90%. I knew then that finally my dream had come true. I had created a true trading system that actually had the ability to go short at the top and long at the bottom.

My dream was finally completely realized when in October of 2003, the ATS-3200 went to #2 in the FuturesTruth top-ten list with 216.3% annual performance, and was the #1 T-Bonds Trading System.

Growing up in Malta, I love the sea and boats. In 1972 I acquired a small yacht StarSong and enjoyed sailing around the island. We also had lots of dogs at home, German Shepards and they are my favorite pets.

Charles Tanti B.Sc.(Eng).
www.ats3200.com