A complete trading system consists of the following elements:
- what markets to trade: stocks, futures, currencies, options
- a filter to decide what specific assets to trade
- signals that determine the type of trade to be executed: long, short, spread, etc.
- risk management: position sizing, maximum loss per position, use of stops (hard or mental)
- exits: how to exit winning and losing trades
- an edge: you must have some reason for believing that your approach is profitable
Let me give a high level view of my system. In the future I will write more about the various concepts that led to my current trading system, but for now, I will mention the two most important ideas. The first is trend following. Trend following has a long and distinguished pedigree in trading circles. Here is a definition of trend following trading from Michael Covel‘s excellent book “Trend Following“:
‘One author described it succinctly: “Let’s break down the term ‘trend following’ into its components. The first part is ‘trend.’ Every trader needs a trend to make money. If you think about it, no matter what the technique, if there is not a trend after you buy, then you will not be able to sell at higher prices … ‘following’ is the next part of the term. We use this word because trend followers always wait for the trend to shift first, then ‘follow’ it.”‘
There are many different types of trend following systems, break outs (such as the Turtle system), moving average cross overs, Bollinger bands, etc. Each system has free parameters that a trader can set to make a system short, medium, or long term oriented, the nature of which depends on the frequency of the data (for instance, one can create a long term day trading system based on minute bars). My system is a long term system (winning trades should last for months) that uses a variation of a classical dual moving average cross over system. William Dunn of Dunn Capital Management is probably the most well known and successful user of this type of system.
Dunn is profiled in Covel’s book. After reading the book and seeing his performance numbers, I decided to investigate his methods by running simulations on Trading Blox using futures data. I discovered that a simple dual moving average cross over system has some problems. While good returns can be generated, the draw downs can be unpleasant. A good rule of thumb is that the annual % returns will match the maximum draw down. Looking at the performance records of Dunn Capital Management compared to simulation results, it was evident that Dunn has added some elements to mitigate down side risks in choppy markets (the type of markets that cause trouble for all trend following approaches).
After much tinkering, reading, and thinking I realized one other problem with the classical approach to trend following. Trend following has traditionally been used for futures and currency trading. Although less common, it has also been used for stocks (see “Does Trend Following Work on Stocks?” by Wilcox and Crittenden). What these ideas have in common is the identification of trends for what ever markets are being considered combined with adjustments for correlations. In my search for trading ideas, I read about various asset allocation schemes. This prompted me to think about markets as not distinct entities, but rather as belonging to a set of assets. The idea here is to spread risk among various asset classes (equities, bonds, currencies, real estate, private equity, commodities, countries, sectors, etc.).
My next conceptual breakthrough was due to reading about cybernetics (see links; I have a number of posts planned to discuss this subject). Here is a brief description, “So a great variety of systems in technology and in living nature follow the feedback scheme, and it is well known that a new discipline, called cybernetics, was introduced by Norbert Wiener to deal with these phenomena. The theory tries to show that mechanisms of a feedback nature are the base of teleological purposeful behavior in man-made machines as well as in living organisms, and in social systems.” What was most important about the concepts of cybernetics was that it lead to me think about trading in a different way. A cybernetic view of my dual moving average trend following system can be conceptualized as the long term trend as a goal that the shorter term trend attempts to follow. When the short term trend deviates too far from its goal, a feedback mechanism attempts to correct the error and redirect it towards its goal, the long term trend. From this description, it should be evident that volatility must be handled carefully so that the deviations are not too large or too frequent. My system can be explained without cybernetics, however, for me, the concepts of cybernetics were crucial in understanding the nature of my system.
Thus I arrived at a system that combines a modified version of classic dual moving average systems and asset allocation.
In my next post, I will describe asset allocation in detail.
 Bertalanffy, Ludwig Von. “Causality and Teleology.” General System Theory: Foundations, Development, Applications. New York: Braziller, 2009. 44.