Module 10: Trading systems

A trading system is a set of rules written down in advance: when to enter, when to exit and how much to trade. Writing rules makes them testable, and testing is what separates a system from a hunch.

01Why systems

Rules remove the decision from the moment. They also make results measurable, because the same rules can be checked on past data and followed in the future.

02The parts of a system

Market and timeframe, entry rule, exit rule, stop, position size and the conditions under which the system does not trade at all.

03Trend-following systems

These buy strength and sell weakness, accepting many small losses for occasional large wins. Moving-average crossovers are the classic example.

04Mean-reversion systems

These trade the return to an average after a stretch. They win often in ranges and suffer when a trend runs.

05Backtesting

A backtest runs the rules over past data. It must use only information that was available at each moment, include charges and slippage, and cover enough market conditions to mean something.

06Avoiding overfitting

Rules tuned to fit the past perfectly usually fail on new data. Fewer parameters, out-of-sample testing and common sense guard against it.

07Measuring a system

Win rate alone says little. Expectancy, drawdown, profit factor and the longest losing run describe whether a system can be lived with.

08Running it live

Paper trade first, then trade small. Most systems fail in execution, not on paper, because the trader overrides the rules at the worst moment.

09Where this leads

On A1Stock, replay lets you step through past sessions minute by minute with the data as it was, which is a practical way to check a rule by hand.