Trading Strategies: How to Build a Trading Approach

How to build a trading approach: the five components every strategy needs, the difference between a strategy and a setup, why the exit and the size matter more than the entry, and how to test an approach before risking money on it.

A trading strategy is a written set of rules that answers five questions the same way every time: what you trade, when you enter, where you exit if wrong, where you exit if right, and how large the position is. Anything short of all five is a setup, not a strategy — and a setup without the last three is a guess with a chart attached.

Key Takeaways

• Five components: market and timeframe, entry, exit if wrong, exit if right, position size.
• The entry gets most of the attention and matters least. The exit and the size decide the outcome distribution.
• A strategy must be specific enough that two people reading it would take the same trade.
• No strategy works in all conditions. Knowing which conditions yours needs is part of the strategy.

The Five Components

1. Market and timeframe. Which instruments, and on what chart. A range approach that works on EURUSD in London hours is a different thing on an exotic pair overnight — see volatility and liquidity. Narrow is better than broad at the start.

2. Entry. The conditions that must be present. Written so they can be checked as true or false, not interpreted. "Price breaks above the range high on the London open" is a rule; "price looks strong" is not.

3. Exit if wrong. The stop-loss level, and how it is derived — a chart level, a volatility multiple. This has to come from the market, not from the amount you want to risk.

4. Exit if right. A target, a trailing rule, or a condition. Without one, the exit is a decision made under pressure, which is where most discretionary error lives — see loss aversion.

5. Position size. Derived from the stop distance and the risk per trade. See position sizing. This is the component that decides whether a run of losses is survivable.

Why the Entry Matters Least

Most published trading material is about entries. But an entry only decides where the trade starts; the exit decides what it produces, and the size decides what it costs. A mediocre entry with a defined stop, a defined target and correct sizing produces a workable distribution. An excellent entry with no stop and an arbitrary size does not — and the second is far more common.

The arithmetic is in risk-to-reward ratio: the ratio and the win rate together determine expectancy, and both are properties of the exits, not the entry.

The Main Families

Every row has a condition and a failure mode. A strategy that does not name both is incomplete.

Systematic or Discretionary

A systematic approach applies fixed rules with no judgement at the moment of the trade. It can be backtested and it removes most psychological error. A discretionary approach uses rules as a framework and judgement within it, which allows adaptation and admits confirmation bias.

Neither is superior. What matters is that a discretionary approach still has all five components written down, so that "judgement" applies to the entry and not to the stop.

Worked Example: A Strategy Written Properly

The right column can be tested, followed and reviewed. The left column cannot, and a month of it produces no information about whether the idea works.

Testing Before Risking

Before live money: define the rules, backtest them on historical data with the rules fixed before the data is examined, then run them on a demo account with the position size you will actually use, keeping a journal from the first trade. The journal's compliance column is what tells you whether a live result is about the strategy or about you.

Key Risks

• A strategy with no defined failure condition. If nothing would make you stop using it, it cannot be evaluated.
• Changing rules mid-sample. A rule adjusted after three losses is a new strategy with no track record.
• Testing on too few trades. A few dozen trades tells you little about a distribution.
• Ignoring costs. Spread, swap and slippage are part of the result and are usually absent from a backtest.
• A strategy that requires conditions that are rare. Correct and untradeable is still untradeable.

No Approach Guarantees a Result

Every strategy on this site describes how a method works, when it is used and how it fails. None of them is presented as profitable, and no arrangement of rules removes the risk of loss. What a written strategy does is make results interpretable: when it loses, you can tell whether the method failed or the execution did.

What is the difference between a strategy and a setup?

A setup is an entry condition. A strategy adds the exits, the size and the conditions under which it applies.

How many strategies should I have?

One, until it is understood well enough to be reviewed meaningfully. Multiple approaches at once make it impossible to tell which is producing the result.

How long before I know if a strategy works?

Long enough for the sample to be meaningful — many dozens of trades, including a losing run. Fewer than that is noise.

Should I use indicators?

A fixed, small set chosen in advance is a rule. A growing set consulted until one agrees is confirmation bias.

Can I use someone else's strategy?

Only if it specifies all five components. Most published "strategies" specify one — the entry — and leave the parts that decide the outcome undefined.

Related Guides

Trend trading · Range trading · Breakout trading · Price action · Swing trading vs day trading · Multi-timeframe analysis · Position sizing · Risk-to-reward ratio