Writing down a strategy
A strategy only becomes one once a stranger could execute it without asking any questions. Everything else is a collection of intentions.
The test is simple: hand your strategy to someone with no expertise. If they can apply it without a single follow-up question, it's defined. If not, it isn't.
A complete strategy answers six questions. What do I trade? When do I enter? How big is the position? Where do I exit at a loss? Where at a gain? And when do I not trade at all?
That last question gets forgotten almost every time, and it's the most important one. A strategy has to say when it doesn't apply, or it gets applied to everything.
Write down why you believe this works. If you can't state the reason in two sentences, you don't have a strategy, you have a rule with no foundation, and you'll break it at the first sign of trouble.
For an overview of common strategy families and what the research finds, see Strategy Types Overview.
A testable strategy definition requires precise statements on universe, signal definition, position size, exit rules, timing, and cost assumptions. Any under-specification creates degrees of freedom during later evaluation that invite after-the-fact adjustment, destroying the value of any test.
The rationale should rest on a nameable cause, such as a risk premium, a structural market constraint, or a documented behavioral bias. Without such an explanation, a pattern found in hindsight can't be distinguished from a random finding, and there's no basis for expecting it to persist.
It also makes sense to define stop conditions before deployment: at what drawdown, what deviation of realized metrics from expected ones, or what period without confirmation the strategy gets paused. Without predefined stop criteria, that decision gets made under emotional strain, right when a loss is happening.
Summary
- Six questions: what, when in, how big, where out at a loss, where at a gain, when not at all.
- Without a nameable cause, a rule can't be told apart from chance.
- Stop criteria belong before the first trade, not during the losses.
Did you get it?
When is a strategy fully defined?
When a stranger could execute it without asking a single question.
Which question gets forgotten most often?
When the strategy doesn't apply.
Why does a rule need a rationale?
Without a nameable cause, a hindsight finding can't be told apart from a random one.
Related
- Capital preservation before profitStage 2
- The one-percent ruleStage 2
- The trading journalStage 2