GOAL OF THIS ARTICLEIn the name of full disclosure we are a forex software company and we are in the business of making money. However, and this is a big, however, you can use all of our tools for free and you can get free advice from our programming staff at no charge whatsoever. You can also develop a strategy for free and test it using a demo account with a real broker, all for free and for an unlimited time.
When you have a strategy developed and are ready to trade real money you can opt in to use our software for a small fee or you can apply your strategy elsewhere. If you decide to automate your strategy you can use our forex robot development tools for free, or you can hire our team to create it for you, usually within a few days. As of this writing the fee to use our software for 12 months and have a robot developed for you is only $299 (there is also a monthly option for much less). If you search for similar services you will find that software subscriptions alone are usually 10 times this amount and programming fees are equally greater. Furthermore, our personal support is what keeps us in business. At the end of this article we will provide you with a special bonus to help make your strategy even more profitable.
How to Create Your Own Forex StrategyThe first part of this section is about finding ideas for a strategy, which in my opinion is the fun part. In brief, we will discover ideas by reverse engineering charts and price behavior to form simple strategies that can then be analyzed and revised until they prove to be profitable. In the event that initial attempts prove to be futile we can move on to the virtually endless strategies waiting for us to discover.
Reverse Engineering Charts
What exactly does it mean to reverse engineer a chart? Get ready because this is the method that you will use over and over again to discover ideas for strategy development. Once you learn this practice it will eventually occupy your mind with an obsession like never before. Why, because you will inevitably discover money making strategies on your own. This feeling will provide you with the power of independence in that these are new creations made by you and that there are virtually endless opportunities waiting for you to discover.
Equally important, you did not have to purchase something from a more than likely unscrupulous individual just looking to sell you some garbage for that magical $197. Plus, if you traded that $197 forex robot you probably lost even more.
Reverse engineering charts involves a simple process. Select a currency pair that you would likely trade and then add an indicator that you are either familiar with or have a desire to learn. Observe price patterns as they relate to this indicator and formulate a working strategy. This will make more sense as we apply the process using a real world example.
Moving Average Crossover - Reverse Engineered
Important: Although adequate, the purpose of this example is not to teach you how to trade the moving average crossover or how to write code to create a forex robot. Instead, as is the goal of this entire article, it is to teach you how to dissect charts in a way that you can reverse engineer them into a trading strategy.
This example in reverse engineering uses a popular charting indicator – the Simple Moving Average (SMA). The SMA is a main panel indicator plotted over the price bars and represents an average price for a specified period. For example, an SMA of 7 plotted on a 1 hour chart would represent the average price over 7 hours.
Multiple SMA’s with varying time periods may be applied to a chart to assist in detecting a change in trend. This is known as a moving average crossover. The following image represents a price chart with a 7-hour SMA crossing over a 14-hour SMA from above.
The SMA with the smaller time period is known as the fast SMA in that it reacts faster to price changes. In the example above the pink line is the fast SMA and the light blue line is the slow SMA.
When the fast SMA crosses the slow SMA from above it represents a change in trend to the downward side, as shown above. This is a clue to place a sell order. The opposite is true for buy orders.
Most traders know this simple trading strategy. As an important reminder the purpose of this example is not to teach the SMA trading strategy, but instead to demonstrate the application of reverse engineering.
Imagine for a moment that you never heard of the moving average crossover strategy and that you only have a working knowledge of simple moving averages. We can see that when the pink line crosses the light blue line from above, that price tends to move downward. The opposite is true when the pink line crosses the light blue line from below, price tends to move up. The next step is to look at additional crossovers to see how often this phenomenon holds true and if we can profit from trading this strategy.
Looking at additional charts does show some potential, however, it is by no means conclusive. Sometimes it works, sometimes it does not. Inspect the following five crossovers.
- The first crossover (up) works nicely. As soon as the fast SMA crosses the slow SMA from below the rate increases as expected.
- The second crossover (down) was not as successful, if any at all. If you entered a sell position at the close of the candle when the crossover (down) occurred you may have been able to capture a few pips on the next small red candle, but would you have exited so early? More on this later.
- Crossover (up) number three appears to be profitable; however, there was a turn to the downside just after the entry signal. Profitability would have been dictated by exit parameter rules which we have yet to define.
- The fourth crossover (down) works as expected providing some potential for profit but again, profits would have been dictated by exact entry and exit rules.
- Crossover (up) number five also appears to work as expected providing plenty of room for profit.
Entry and Exit Analysis
In our example above we had the potential to profit from 2 or 3 out 5 moving average crossovers. The term “potential” is used because we have not yet defined exact entry and exit rules. We only know that the basic strategy of entering on crossovers shows some opportunities for profit.
Knowing the basic theory behind the crossover strategy we could just guess at entry and exit rules. For example, we could enter on the close of the candle after the crossover occurred. Exits could be defined as reversal signals on the next crossover in the opposite direction. Let’s see how that would have resulted by examining each position.
The first crossover would have provided us with a buy order which moved a significant amount to the upside. Entering on the close of the candle after the crossover was timely and exiting on the next reversal signal resulted in profits, however, look at how much money was left on the table with wasted profit potential. Maybe there is a better time to exit? Do you see it? If not, it will be pointed out later in this document.
At the time of the second crossover (down) the rate has already moved significantly lower, getting us in the position late. The sell position ends up being a loss when it closes on the next reversal signal. Try to determine a solution to avoid this late entry situation. Do you see one?
The third crossover (up) is almost identical to the first one. The entry was timely; however, the exit was not. It left too much money on the table. Let’s move on to the fourth crossover.
Notice the surprise? Our first glance at this position seemed to be profitable simply based on the significant move lower after the crossover (down). However, when we apply entry rules to the candle’s close at the time of the crossover (down) and exit at the candle’s close at the time of the reverse crossover (up), we immediately discover that this is a large losing position. Could we change our exit parameters to turn this into a profitable position? Do you see a repeating pattern? Let’s examine the fifth crossover.
The fifth crossover (up) returns a small profit, however, also leaves some money on the table. In addition, this crossover has us entering late in the game. Maybe we could add a filter here to avoid this type of late entry?
Entry and Exit Optimization
We have two overall goals to accomplish – entry and exit optimization. Specifically, we want to try to avoid entering and exiting positions too late. Entering too late increases risk in that the position has already reversed or that it has moved too far in the desired direction. Exiting too late, as we have seen, frequently leaves too much money on the table. This will make more sense as we once again dissect each crossover.
Compare the two images above. The image on the left displays our original approach, entering and exiting at moving average crossovers. Take a look at the image on the right. See what we did here? Entry was the same; however, the exit rule was changed to close the position when the current rate touches the fast (pink) simple moving average. Notice how many more pips we accomplished with this approach. How did we figure this out? Simple, just look at the chart and it will tell you what to do.
This solves the problem of exiting too late and leaving too much money on the table. Will this work in the majority of situations? There is only one way to know, program or have the strategy programmed and test it against a reasonable amount of historical data. Let’s look at the second crossover.
Applying entry and exit on crossovers proved to be unprofitable in our first attempt (left image). The problem arose in that the entry point was a significant distance from the cross over point, providing a hint that the move may be close to ending. A possible filter for entry could be:
Do not enter if the current rate at the time of the entry signal (crossover) is x pips away from the crossover point.
Will this work in the majority of situations? There is only one way to know, program or have the strategy programmed and test it against a reasonable amount of historical data. Let’s look at the third crossover.
Similar to the first crossover, this one begged for a better exit point as it left too much money on the table with such a late exit. Before we get too involved with the exit take a look at the entry point (left image). Notice that the signal is for a buy, however, the rate is currently well below the crossover. This may be inconsequential, however, should be noted as a possible filter for entry if we determine that that there may be a repeating pattern. Since this particular situation worked in our favor, we will continue as planned and enter the position.
Exiting is another story. The first exit follows the rule of closing the order when the current rate touches the fast moving average. The second exit also touches the fast moving average, however, would provide that we eliminate the first exit by adding an additional filter. Perhaps something such as:
For buy orders, close the position when the current rate touches the fast moving average from above. The opposite could be true for sell orders.
The third possible exit follows the original rule of closing the position on a reverse crossover. All positions returned a profit; however, the second one was the greatest, just as it was in the first crossover. The only way to know which will work best over time is to program each option and test it over historical data. Let’s proceed to the fourth crossover.
The entry rate is a bit of a distance away from the crossover point, therefore, if we applied the entry rate filter as described in the second crossover this position may have been avoided (left image). However, since we have the benefit of hindsight here let’s proceed. By changing the rule to exit the position when the rate touches the fast moving average this position results in a profitable move.
The question to ask yourself is…. Which is more important? Filtering out late entries, or letting them ride and exiting earlier? As you may have guessed by now the answer can only be determined programmatically through trading scripts and backtesting. Otherwise, you are playing a guessing game. Let’s check out the last crossover in our sample data set.
Similar to the second and fourth crossover samples this one also provides us with an entry rate far away from the crossover (left image). It could be that if we filter out all of these late entry positions we could lose out on profitable opportunities. Perhaps rather than filtering out late entries we could try to find a way to enter earlier? More on this later, let’s analyze the exits according to the current entry rule.
There are two possible exits according to our current approach and one possible new exit rule. The first exit describes this new rule:
Exit a buy position on the close of the candle following a lower high. For sell positions, exit on the close of the candle following a higher low.
The second exit follows the rule of closing the position when the current rate touches the fast moving average. The third exit is our original rule where the position closes on a reverse crossover. Since all three prove to be profitable the only way to determine which will provide the largest amount of profits over time will be by programming this strategy and backtesting.
Always Ask Yourself…
Reverse engineering charts involves the activity of always asking yourself – “What could I change on this chart to make this position profitable?” By performing this simple action you will inevitably discover profitable opportunities. For example, you could apply the following tests to our sample above:
- Filter out false entries using an additional indicator, i.e. a momentum indicator such as the RSI (Relative Strength Index), or Stochastic Oscillator.
- Enter on the exact time of the crossover, instead of on the close of the following candle.
- Exit using a trailing stop.
- Confirm exit using momentum indicators.
Selecting a Data Set
When selecting a small sample as we have done above it is wise to include a set that consists of several signals. The data need not be consecutive, however, should have periods where signals occurred during range bound activity and also during strong trends. You may find out that different strategies are needed when price is trending and when it is not.
Don’t give up too early. Reverse engineer your approach and then test it with various currency pairs and with various compressions. This can all be accomplished automatically in the backtester after a script has been created. It is very common to discover that your strategy works with certain pairs and on certain compressions while on others it does not. This can only be detected by backtesting a script. To make this task even easier it can be automated using the genetic optimization feature in the backtesting application.
Endless Strategy Ideas
There are virtually endless methods to reverse engineer a chart and with them come endless opportunities for strategy development and profits. Start with simple indicators, chart patterns and price activity. With time you will find that certain tools help you confirm entry and exit, thus, eliminating noise and capturing profits. Here are some ideas to get you started:
- Plot the RSI (Relative Strength Index) on a chart. Pay attention to price behavior as the rate reaches RSI levels of 30 and 70.
- Add Daily Pivot Points to a 15 minute or 1 hour chart. Notice how price reacts when it crosses the primary pivot point (PP) or touches other pivots.
- Analyze price action around the Parabolic SAR indicator.
- Plot the Bollinger Band on your favorite chart. Notice how price behaves when it crosses the middle band or touches one of the outer bands.
- Without using a single indicator just look at price behavior. Do you see certain reactive numbers where price tends to react? For example, 50, 100, 80, 20. Combine this price behavior with certain times of the day or week.
- Candlestick patterns make great robot signals. Did you know that our software has its own candlestick pattern recognition module? You may use this to your advantage as you reverse engineer charts.
If you are even remotely familiar with trading the above ideas may not be much of a surprise to you. However, most just take the indicator for face value and trade it as it is widely publicized and hope for the best. By applying the reverse engineering techniques you have learned in this document you are on your way to creating undiscovered profitable trading strategies.
With your annual subscription to FxSpyder you will receive the following:
- Unlimited Access to the FxSpyder-Oanda Trading Platform.
- Unlimited Access to the FxSpyder-FXCM Trading Platform.
- Unlimited Access to the Robot Inventory.
- Unlimited Access to the BackTesting Application.
BONUS #1: We will double your custom programming credit. This means that you will receive up to $600 worth of custom programming with your annual subscription of only $299. This is more than enough to create several robots or to revise a robot several times as you continue to reverse engineer your charts.
Most robots can be completed within a few days depending on our current schedule. Start reverse engineering charts today and you could have your first profitable strategy within a week.
BONUS #2: For a limited time our programmers and traders will review your script and backtest it for you. Our team will offer suggestions based on our experience on what may make your script more profitable and more efficient. This information is invaluable and is included as ongoing support for your script development.
BONUS #3: Scripts that we develop are eligible for entry into a 30 day live trading performance contest. Fifty participants for each contest will compete for up to $1,000 in prize money.
Where to go from here…
You already have the knowledge to begin the reverse engineering process. Once you have a working strategy in mind you have the following options:
- Consult our detailed help section for robot programming. A good place to start is with our brief tutorial – How to Create a Robot.
- Submit your idea using an informal description to our Help Desk > Robot Programming department. Your strategy can usually be converted into a working trading and backtesting robot within a few days.