Reversing with Equity

The basic concept behind this approach is to reverse the signals given from a system when the equity is below the moving average. This approach seems counterintuitive, but take a closer look. As stated earlier, equity is composed of profit and losses from trades. These trades are based on a system that you are following. Each system design is based on a certain type of market (trending, choppy, volatile, range bound, etc.). As the market changes from one type to another, a system can start to incur losing trades. These losing trades are bunched up during the period that the market is in the type of action that your system dislikes. As the losses occur, your equity will fall and eventually go below the moving average.

While the equity is below the moving average, for the most part the system is still producing losses (more losing trades than winning trades). Therefore, the signals are wrong. What would happen if you did the opposite of the system during this time frame? Common sense would say that you would have winning trades. Now granted, you will have some winning trades when your equity is below the moving average; and eventually as the market changes, the system will start picking up winning trades and the equity will rise above its moving average. Also, keep in mind that once your equity falls below the moving average, this situation can last for a long time. Weeks and months can come and go, as you watch your system take losing trade after losing trade. An example will give you a basic idea of how this technique can work using the volatility breakout system.

As one of the more popular trading systems, the volatility breakout system has developed a following because over time it has performed quite well. In addition, many people feel that markets trade in ranges, only to break out of these ranges to form a new range or trend depending on the fundamental and economic news. This system does not predict the future direction of prices but simply reacts to the current volatility and captures the major price moves. The volatility breakout system performs poorly when the markets are in a tightly congested range. It performs well in volatile markets that are trending in nature.

Figure 7.7 the equity curve of the modified system.

The volatility breakout system is always in the market either long or short. It moves from one signal to the next. It works by calculating the true range for the number of days specified by the trader. Calculating this range requires subtracting the low from the high for that particular day. True range takes into account gaps up and down in price activity. This is a more accurate measure of a markets range. True range uses yesterdays closing price instead of todays low, if todays low is greater than yesterdays close. Also, true range uses yesterdays close instead of todays high, if todays high is less than yesterdays close. You would select the number of days to calculate the average true range. A trader who chooses five days would calculate the average true range over the previous five days.

The second parameter for the system is the breakout percentage. If the market breakout is more than a specified percentage of the average true range either above or below the previous days close, than a buy or sell signal is generated. As a recommendation, look at 4 to 12 days to calculate the true range and 25 percent to 130 percent for the breakout. Experimentation and testing will enable you to come up with a good set of parameters. Many software packages have this system so you can just plug the numbers in and go.

We will look at one example with the Swiss Franc. This market is fairly volatile and like many of the other currencies tends to trend better and more often. Figure 7.8

Figure 7.8 200 days of the Swiss Franc.

SWISS FRANC im 03/13/95

is 200 days of the Swiss Franc ending on March 13, 1995. The market has a few up and down trends and ends with an explosive move to the upside. By looking at 200 days starting with the period ending March 13, 1995, we will use 7 days for the average true range and 65 percent for the breakout percentage. We will also take into account $32 per trade for commissions and no slippage.

Testing during this period produced results shown in the "Initial System" column in Table 7.2.

With the equity curve, we place a 25-day time series moving average. The equity curve just falters back and forth with the moving average. Where the equity is below the moving average, it generally coincides with the times that the Swiss franc is trading sideways and choppy. Can the results be improved without destroying the systems integrity? The modified system places opposite signals when equity is below the moving average. Performance results of this modified version are in the "Modified System" column of Table 7.2.

A comparison of the results shows that this simple technique can turn a mediocre system into a winner. Keep in mind that certain types of systems will not perform better by reversing the signals. This will be due to the type of system and market you are testing. Always keep your stops in place and have good exit rules. Lastly, you will need to keep separate charts to follow the original systems trades and equity while you deviate with your own alternative strategies.

Percentage Charts

As discussed earlier, it is important to know your systems percentage of winning and losing trades. These percentages can be used in chart form to provide valuable insight. The percentage charts are running totals based on the sum of all trades. You will find that your systems percent of winning trades (percentage profitable) will fall within a range of values, for example, 50 percent to 70 percent.

One way to apply this graph is to set your systems limits to, say, 50 percent profitable. If the percent profitable line of your system fails below 50 percent, all trading should cease or position size should be reduced. Conversely, when this line is rising or is above some value, say 70 percent, you could increase position size. You should be trading your heaviest when the odds of a win are in your favor.

Table 7.2 Comparing trading performance of two systems

Initial System Modified System