Systems - Buy Them Or Build Them?

Since we offer systems for sale one might assume that we would suggest
that systems should be purchased rather than built from scratch.
Actually we are very strong advocates of building your own system if you
are capable of doing so.

One of the reasons that we make this recommendation is that we believe
it is important that trading systems should fit the needs and
preferences of the user. A system may appear to have excellent
performance data but another trader may consider one trader's favorite
system as completely worthless. It seems that everyone has a different
level of comfort when it comes to following a particular system and the
comfort level increases dramatically if you are personally involved in
the planning and research that produces the system.

In addition to comfort levels and other psychological issues,
preferences for systems may vary because the resources of traders will
vary. For example some traders have very large amounts of capital to
work with while other may have much smaller amounts to put at risk. Due
to differences in capital or risk tolerances these traders may require
different systems. In addition to the amount of capital available there
are other resources that may influence the choice of systems. Some
traders have many hours a day available and want to trade for a living
while others have very demanding occupations and limited time to devote
to trading. Many systems traders have extensive experience with
computers and a great deal of knowledge about programming. But on the
other hand, many of us will have little knowledge and experience in this
technical field. Some systems traders have many years of actual trading
experience and others are admitted novices. Some traders have access to
the newest and most expensive hardware and prefer to receive real time
price data while many traders will find they can trade well using less
sophisticated equipment and simple end of day data.

We could continue with additional examples of important differences
among traders that might influence their preference in trading systems
but the point has been made. The advantages of building your own
trading system might be compared to the advantages of owning a tailor
made suit as opposed to buying one off the rack. If you build your own
system you can tailor it to your exact specifications so it fits your
preferences and resources.

However, just as most of us would not consider making our own clothing,
designing and testing a trading system is a task that may not be
appropriate or realistic for everyone. Assuming that you have the
necessary knowledge and resources to create your own system there are
still many obstacles to overcome. For one thing there is the
frustration of investing in equipment, software and data only to find
that most of our best ideas don't work. It would not be an exaggeration
to reveal that ninety nine percent of our research winds up in the
wastebasket. Also keep in mind the value of your time because until a
system is completed you will be working hundreds of hours for less than
minimum wages. There are no shortcuts so you had better enjoy the
entire process.

However the benefits of building the system yourself can be substantial
and we think the benefits are worth the effort. You will know and
understand the logic behind each step in the system. The knowledge
gained in the testing process should give you confidence in the system
so that you will have the necessary discipline to follow the system
correctly when it comes time to start trading it with real money. The
testing process should also give you a "feel" for the system that will
help you to sense when it is not performing as intended. If necessary
you will also have the ability to fine tune the system based on your
real time experience. We believe that over the long run the discipline
and confidence of the system operator have a great deal to with
determining the ultimate success of any system. By creating the system
yourself you are most likely to trust it and operate it correctly.

We suggest that you should seriously consider building your own systems
if you can satisfy most of the following criteria:

1. You enjoy challenges and have a knowledge of the commitment required
to excel at anything. If you have been an outstanding success at any
endeavor then you already understand commitment. In particular we
believe that success at competitive games like bridge, chess,
backgammon, poker and other pastimes requiring discipline, patience and
strategy would be a good indication that you can build a good system and
operate it successfully.

2. You own or are prepared to purchase a testing platform (like
TradeStation or something similar) and will spend the time required to
learn how to operate it.

3. You understand how to program the software you will be using or you
are willing to pay a professional programmer to assist you. If you do
not have a programming background you will find that there is nothing
easy about Omega's so-called Easy Language. If you have no programming
experience and decide to hire a programmer you should study enough about
the programming to enable you to communicate with your programmer. As a
minimum you need to be able to express your ideas in "if / then"
statements.

4. You need to have studied some basics on testing procedures and know
how to analyze and evaluate the test results. Some training in
statistical analysis would be very helpful.

5. You will need a source of clean historical data and the software to
convert the data to the format that will be used in testing. For
example you may want to convert the data to continuos back-adjusted
contracts. There are other popular formats for historical data and you
should know which format is appropriate for the testing and type of
system you will be creating.

6. You have spent some time clearly defining exactly what you want from
your system. You need to list and prioritize these desired
characteristics because they will guide your research.

7. You must have an abundance of time and patience. Picture yourself as
Thomas Edison setting out to invent the light bulb. Will you have the
commitment and patience to keep trying one element after another to find
the one that works? (As an example, we have been working on S&P trading
systems for years and we have yet to come up with one that we would be
willing to trade. We are still trying.)

8. Can you be truthful with yourself and analyze your test results
without letting your natural optimism cloud your judgment. (However, if
you are not an optimist you are automatically disqualified.)


Chuck Lebeau is the co-author of Computer Analysis of the Futures Market, and the former co-editor of Technical Traders Bulletin. Chuck is currently operates a website devoted to trading topics; www.traderclub.com.