Test rewarded rule breaks before replacing the system

Short-horizon results can retire the procedure

A carefully specified mechanical trading system can still post losses in the short run. A deliberate rule break can still be followed by a favorable short-horizon outcome. That pairing can start a pattern of abandoning the procedure.

A mechanical trading system is a single written procedure that turns rule inputs, market state and execution constraints into entry, exit or abstention signals over the system holding period. The trading psychology process is the habit loop in which short-horizon wins and losses train the operator to follow, outsource or sabotage that pre-specified procedure.

Peer advice and sequential breaks

A new operator who substitutes peer advice for a purchased, tested system's signals can receive intermittent favorable outcomes. Those outcomes can further reduce trust in the original rules, including unused system signals that later looked favorable.

A long-tenured operator who accidentally violates a major rule and receives an unusually large gain can then break remaining rules in sequence until the account is largely depleted. In both cases the first rewarded exception, a rule violation followed by a favorable short-horizon outcome, tempts the operator to treat the violation as a new method.

Specify the exception in planning mode

A candidate rule change should be designed in planning mode, with contingencies for the consequences of the change, while the incumbent rules continue to be followed. Planning mode is the separate design step in which that draft is specified without substituting it for live signals. The incumbent rules are the current written procedure that remains in force while any candidate change is planned and tested.

A favorable result after a bad decision is not sufficient reason to keep the exception unless that exception is tested and written into the rules.

Trust comes after the long sample

New rules earn trust through later losing stretches only after they have been backtested and shown to function over a long sample. Commitment through drawdown is the willingness to keep executing a rule set after testing has shown why that set is the procedure to follow when results turn unfavorable.

Keep decision ownership with the operator

The operator should retain responsibility for trading decisions, because insecurity about oneself or the system makes it easy to hand those decisions to other traders. Decision ownership is the requirement that the operator, not a more confident peer, issue the entry, exit and abstention signals.

Emotional discharge should happen outside the trading process. Following the rules can be paired with a personally valued reward so adherence carries a positive association.

When the incumbent rules were never tested

When incumbent rules were never seriously backtested, simultaneous testing of those rules and the recent exceptions can decide which procedure to adopt as the committed rule set. Robustness testing is that side-by-side historical testing of an incumbent rule set against a specified exception before the operator is asked to trust either set through later losing stretches.

A rule set should be believed because it has been tested over time and shown to be reliable. If it cannot pass that test, the operator should check whether the exceptions can themselves be written as committed rules.