What is a systematic trading system?
A systematic trading system is a fixed, rules-based framework that defines exactly when to enter a trade, how much to risk, when to exit, and what to do when the market moves against the position. Every decision is specified in advance, so execution is mechanical rather than emotional.
The point is not prediction. The point is consistency — the same inputs producing the same actions, every time, so that performance can be measured against the rules instead of against a feeling.
- Systematic trading
- Trading executed against a written rule set where every entry, exit, and sizing decision is pre-defined.
- Discretionary trading
- Trading where the human makes the final decision in the moment based on judgment, context, or feel.
- Algorithmic trading
- Any trading executed by software. Often systematic, but a system can be systematic without being algorithmic.
The four required components
Every real system — manual or automated — contains the same four pieces. If any one is missing, what you have is a setup, not a system.
- Signal — the precise condition that triggers a trade. Not a vibe; a value, a level, or a confirmed pattern.
- Sizing — how much capital is committed per trade, expressed as a function of account size and risk, not as a fixed dollar amount.
- Exit — both the win-side exit (target, trail, or time-based) and the loss-side exit (stop). Defined before entry.
- Risk cap — the maximum loss tolerated per trade, per day, and per drawdown period before the system pauses.
Systematic vs discretionary trading
Discretionary trading isn't wrong — it's just unmeasurable at scale. The reason structure wins over time is that systems are debuggable. When a discretionary trader has a bad month, they have to interrogate themselves. When a systematic trader has a bad month, they interrogate the rules.
| Systematic | Discretionary |
|---|---|
| Rules written before the trade | Decisions made in the moment |
| Performance attributable to the system | Performance attributable to the operator |
| Repeatable by anyone with the rule set | Tied to one person's intuition |
| Emotion is structurally removed | Emotion is the primary risk |
How to build one
- 01Define the edge in one sentence
Before any chart, write the single observation you believe gives this strategy an advantage. If you can't say it in one sentence, the edge isn't clear yet.
- 02Translate the edge into a signal
Convert the observation into a measurable trigger — a level break, a moving-average condition, a volatility threshold. The signal must be objective.
- 03Set sizing as a function of risk
Size each position so that hitting your stop costs a fixed percentage of capital. Sizing is not a number — it is a formula.
- 04Pre-define exits
Decide both the profit-side exit and the loss-side exit before placing the trade. These are not adjusted mid-position.
- 05Cap aggregate risk
Set per-day and per-drawdown caps. When hit, the system stops. This is how a system survives the months that test it.
Common mistakes when starting
- Optimizing the rules to past data until the backtest looks perfect, then watching it fail live.
- Adding a discretionary override 'just for this one' — which becomes every one.
- Treating sizing as an afterthought instead of as the most important variable.
- Measuring the system over a sample too small to be meaningful.
Systematic trading is less about being right and more about being repeatable. The traders who last aren't the ones with the cleverest signal — they're the ones whose rules survive contact with a bad month.