Anatomy of a Trading Plan

The short answer

A trading plan is a written document specific enough that a stranger could read it and take the same trades you would. That is the whole test, and most documents people call trading plans do not pass it.

The test

Read each rule and ask: would two competent people acting on this take the same trade?

• "Enter on a strong breakout" — no. Whose definition of strong?
• "Enter long when price closes above the prior swing high on the 1-hour chart, with the swing high defined as a candle whose high exceeds the three either side, while price is above the 200-period moving average" — yes.

The second is longer and less elegant. It is also testable, and untestable rules cannot be improved, because you can never establish whether a losing run came from the method or from your interpretation of it that week.

Every soft word — strong, clean, obvious, healthy, clear — is a place where your judgement will vary with your mood. Find them and replace them with conditions.

The sections

1. Scope

Which markets, which instruments, which sessions. "Everything, whenever I'm free" is not scope; it is availability. Scope narrow enough that you can become familiar with how those instruments behave.

2. Timeframes

Context and execution, fixed. Plus the commitment not to consult others while in a position — see timeframes and context.

3. Setups

Each one gets: a name, the conditions that must be present, the conditions that disqualify it, and where invalidation sits. Two or three setups is plenty. More than that and you will not accumulate enough samples of any one to learn anything about it.

4. Risk

• Maximum risk per trade, as a percentage of current equity
• Maximum risk open at once, accounting for correlation
• Daily loss limit — and what you do when it is hit
• Weekly loss limit — and what you do when it is hit
• Sizing method (see position sizing)

The bolded parts are the ones most often missing. "Daily limit: 3%" without a stated action is a number, not a limit. "Daily limit: 3%; at 3% I close the platform and do not trade again that day" is a limit.

5. Execution

Order types for entry and exit. Whether stops are submitted with the entry. Whether you scale in or out, and if so, the exact schedule. Maximum trades per day.

6. Exits

Invalidation for each setup. Target logic, or the rule for trailing. What happens if neither is hit by the end of the session.

7. Review

When you review execution, when you review the method, the minimum sample before a method conclusion, and who or what has to be true before an amendment is made.

8. Amendment

The rule about changing the rules. It should say: amendments only at scheduled reviews, outside market hours, with no position open, and recorded with a date and a reason.

That last section is what stops a plan from decaying into a diary of whatever you did.

What a plan does not do

It does not make a method profitable. A precisely specified bad method is still a bad method — it is just a bad method you can now identify as such.

That is the actual value. Without a plan, a losing period is uninterpretable: you cannot tell whether the method failed, or your execution of it did, or whether the sample is simply small. With one, those are separable questions, and separable questions can be answered.