Trading strategy
Family VII · Market & styles
Not to be confused with trading day, trading psychology, trading journal.
A trading strategy is a written rule set that determines when a position is opened, how large it is, how it is managed while open, and when it is closed. It converts a market view into repeatable decisions, so that results can be measured rather than attributed to individual judgement on the day. The rules may be discretionary, systematic, or a mixture of both.
What a strategy specifies
A complete strategy answers a fixed set of questions before any trade is placed:
- Universe — which instruments or contracts are eligible.
- Signal — the observable condition that triggers a trade, such as a price level, an indicator reading or a scheduled event.
- Direction and size — whether the trade is long or short, and how many units or contracts are committed.
- Risk limits — the maximum loss per trade, per day or per position, and any stop level.
- Exit — the conditions for taking profit, cutting a loss or closing on a time limit.
- Filters — conditions that suppress the signal, such as low liquidity or scheduled news.
Rules that are not written down cannot be tested, and a strategy that has not been tested cannot be distinguished from a series of guesses.
Worked example: expectancy of a rule set
A strategy is applied to 200 trades. The average winning trade gains 1.8 times the amount risked, the average losing trade loses 1.0 times the amount risked, and 45% of trades win.
The result is expressed in R, the amount risked per trade, so it is independent of account size. A positive expectancy does not guarantee a profit over any short run of trades; it describes the average outcome if the rules are followed over many repetitions.
Testing and limits
Strategies are usually examined in three stages: a backtest on historical data, a forward test on live data with no capital committed, and live execution at reduced size. Each stage has known weaknesses. Backtests are sensitive to the choice of period, to survivorship in the instrument list, and to assumptions about fills and costs. Forward tests are limited by the number of trades the market produces. Live results are affected by slippage, commissions and financing, all of which vary by broker, instrument and jurisdiction.
Past performance of a rule set is not evidence of future results. A strategy can also decay as other participants adopt similar rules or as market structure changes.
Discretion and automation
A strategy may be fully mechanical, with every decision produced by explicit conditions, or discretionary, where the trader interprets the same conditions in context. Discretionary strategies are harder to test because the decision rule is not fully specified. Automated strategies remove execution hesitation but introduce dependence on the software, data feed and connectivity. In both cases the strategy is the rule set, not the platform or the instrument it is applied to.
Often confused with
- trading day
- A trading strategy is a rule set for entering and exiting positions, whereas a trading day is a calendar period during which a venue accepts orders; the visible sign is that one is a set of instructions and the other is a span of time.
- trading psychology
- A trading strategy defines what actions the rules require, whereas trading psychology concerns the emotions and biases that affect whether those actions are carried out; the visible sign is that one is written in the plan and the other appears in the trader's behaviour.
- trading journal
- A trading strategy states the rules in advance, whereas a trading journal records what was actually done and the outcome afterwards; the visible sign is that the strategy is written before the trade and the journal entry after it.