Strategy Logic Review
Entry and exit rules read for internal contradictions, unstated conditions, and the gap between the method as described and the method as executed.
Write your rules down, then hand over the trades. The interesting part of this review is the distance between the two documents, and that distance is rarely small.
The comparison that does the work
Every trade in the record is tested against the rules as written. Entries that no rule permits, exits taken early, positions held past a stated invalidation, filters applied on some days and not others. The result is a compliance rate, and then the more useful number: what the method would have returned had every rule been followed exactly.
Both outcomes are informative and they point in opposite directions. If discipline cost money, the problem is behavioural and the rules are fine. If deviation earned money, the written rules are not the method being traded, and optimising them is wasted effort.
Also examined
- Conditions that cannot be evaluated in real time, such as a rule referring to the high of a session while the session is still open.
- Rules with no defined failure state, which is the most common structural fault in discretionary methods.
- Dependence on a small number of trades, tested by removing them and recomputing.
- Overlapping or contradictory filters that quietly reduce the sample to a handful of setups per year.
What to send
The rules in whatever form they exist. A document, a checklist, screenshots of a notebook, or a recorded explanation. Incomplete is normal and usually informative in itself: rules that cannot be written down are rules that cannot be tested.