Analyze Your Trading System

I'm going to share with you some of the factors that I examine when I am testing and analyzing potential new trading systems. First, you might be wondering what I use when developing new systems. There are a variety of different software packages available, but I use TradeStation most of the time. The graphic below shows a TradeStation summary of a system that is being tested. You can refer to that throughout this article for the different factors we will discuss.
If you do not want to pay for a software package, advanced users in a spreadsheet program such as Microsoft Excel could import data from Yahoo! Finance or some other data vendor and set up data tables to simulate trading. That works just as well as sophisticated software packages but is much more time-consuming and difficult to set up.
The Time Period you choose is important because you want to analyze your system during a variety of market conditions. I usually choose to start somewhere around January 1998 and test all the way to the current date for my systems. This way your system will be tested in strong up years, strong down years and several years that moved relatively less.
Net Profit is probably the first thing most traders look at when they develop a system. Unfortunately, some traders only look at their net profit, and that is why I am writing this article. Net profit is important, of course, but not necessarily the most important factor. A system might have a huge net profit, but if all that profit was made in a couple large trades during 1999 or 2002, that is not a good system. What you want to make sure you develop is a system that works consistently well in all market conditions.
Total Trades and Trade Length are important so you know for sure that your system will work well with what you are trading and again you also want to look for consistency. If you are developing a swing trading system for stocks you will probably want a few dozen trades per year that meet your conditions and most of the trades taking anywhere from 2 to 14 days. If you're going to be trading something else then those conditions will be different.
Another important factor in your new trading system is Percent Profitable. Of course we would naturally assume that the system that has the highest number of profitable trades would be the best. That is not always true though. This is where Win/Loss Ratio (average winning trade divided by average loosing trade) comes in. You may have a high percentage of profitable trades but the average loss might be much higher then your average gain. I have seen systems that have as many as 85% of their trades showing a profit, but if that same system has a Win/Loss Ratio of 0.50, that is their losses are on average twice as big as their wins, you still may not be able to make any money with this system. However there are also systems that may only be right 30% of the time but when they're right they're right by a lot and may have a high Win/Loss Ratio of 4.5 or greater and that you might actually be able to make money with. In general though, most good trading systems have a greater then 50% Percent Profitable and a great then 2.0 Win/Loss Ratio.
The final factor that we are going to examine is the Equity Curve. This is the ultimate test of a consistent trading system. The Equity Curve is simply a graph that shows the total profit or loss after each trade. The line of a good Equity Curve should steadily increase from the left side of the chart to the right. This shows that the system is making consistent profits in all market conditions. Bad systems, even ones who are winners in the end will have wild fluxations in their Equity Curve, having huge increases and decreases for long periods of time or periods of no movement at all.