Introduction to Trading Systems – Part I
Traders have many wise and pithy sayings which illustrate market wisdom and correct practices that might serve them well in their speculative operations in the markets. Two of those which appear to be in conflict with each other are:
1) “You never go broke taking a profit”; and
2) “Ride winners, cut losers, manage risk”.
In this article I attempt to answer the question of what systematic trading is by introducing what goes into a complete trading system. I also introduce a simple yet effective moving average crossover system which over a series of articles I will develop, adding each component to make develop a complete system that can be back tested and traded for any type of market or instrument.
Systems trading, systematic, mechanical or quantitative trading is nothing more than a set of hard coded rules programmed in a computer and applied to a time series data set. More specifically systems developers program certain logic which has a set of rules of when to buy or sell exchange traded instruments. One might argue that systems traders are data miners who aim to develop rules that produce profits when applied to actual markets in live trading conditions.
A good trading system seeks to automate the entire trading process and eliminate any discretionary decision making by the trader. A complete trading system covers a number of aspects and provides clearly defined answers to the following questions:
- which markets to buy or sell;
- how much to buy or sell;
- when to buy or sell;
- how to buy or sell;
- when to get out of a losing position; and
- when to get out of a winning position.
A Complete Trading System
Academics often argue that markets are random and efficient. Systems developers and quantitative traders have an opposing view in that markets are not efficient and that price patterns repeat themselves and it is possible to gain an advantage in the markets, through the development of computerised trading systems that scan for and capture repeatable price patterns in real time. These systems do not necessarily need to be 100% accurate.
To demonstrate this we can run a quick and simple “coin toss” exercise. To keep this simple we assume that every time you get heads you win €1.00 and every time you get a tails you lose €0.10. Assuming the coin is balanced over a sufficient number of coin tosses the results are evenly split between heads and tails. Providing your losses are much smaller than your winners, as in our example then your system is said to have positive expectancy or positive expected value.
Assuming you start off with €1.00 you can withstand a series of 10 consecutive losing trades before your system wipes out. In a very simplified manner this is the mathematics of risk management when it comes to developing trading systems or even trading discretionary. This directly also relates to the second quote at the beginning of our article where you ride winners, cut losers and manage risk.
Using a trading system with clearly defined rules is possibly the best way to make money trading. If a system has a positive expectancy and makes money over the long run then the only job for the trader is to follow the system religiously and take all trading signals irrespective of outcome on each individual trade. Traders that rely on their own discretionary judgement when trading may find themselves being inconsistent and emotionally wrapped to their trades.
Typically discretionary traders particularly novice traders tend to be courageous when they should be fearful and fearful when they should be courageous. This type of behaviour typically leads to the outcome in the first quote where they take profits far too soon and cut losses too late.
In the next article of the series I introduce a basic moving average crossover system which seeks to provide clear answers to the questions of when to buy and sell. In subsequent articles I will introduce concepts that seek to provide answers to the questions that a complete trading system might include.
Find more: Contributing Authors