1. What a plan contains
Six decisions, written down: market (which pairs), timeframe (how long positions live), setup (what pattern/condition you wait for), risk per trade (a number), exit rules (both directions), and a no-trade rule (when you sit out). Anything vaguer than that is a hope, not a plan.
2. The setup must be objective
"Buy when it looks strong" is not a setup. "Buy the first daily close above the 50-day MA with volume 20% above average, stop 3% below" is - a third person could check it. Objectivity is what makes the backtest and the journal meaningful.
3. Exits decide profitability
Both exits are planned before entry: the stop (risk side) and the target or trailing rule (reward side). A plan with only entries is half a plan - the asymmetry of drawdowns punishes missing exits brutally.
Remember: The plan's job is not to predict. It is to make your behaviour predictable when your emotions are not.
4. Writing it: a one-page discipline
One page, no essays. After every trade, the journal answers one question: did the trade follow the plan? Rule-following - not profit - is what the journal grades. A profitable rule-breaker is a future statistic.
5. When this fails
A plan fails when the market regime it assumed ends (high-volatility trend → chop), when rules stop being checkable, or when adherence is measured only in P&L. Review monthly; adapt deliberately.
Risk: Changing the plan mid-trade to "make the trade work" converts a small planned loss into an unplanned large one. The plan protects you from you - mid-trade edits break that.