Mechanical systems as a gated factory

Feelings move onto the rules

A mechanical trading system is a fully specified procedure that turns rule inputs, market state and execution constraints into entry, exit or abstention signals over the system's holding period. The archive presents mechanical trading as still emotional. The change is that feelings attach to following or ignoring a rule set, rather than disappearing once rules exist.

Once more than a handful of rule-based systems are live, tracking discretionary overrides becomes impractical. That limit is offered as a way to reduce ad hoc signal rejection. Each live system is supposed to carry a stop-trading rule: pre-set performance conditions that retire the system so retirement is not left to mood when results deteriorate.

Begin with written goals

The design sequence is described as seven to eight steps that begin with written goals, including targets for annual return and maximum drawdown. Vague aims such as large profits with small drawdowns are linked to extra rules, curve-fitting and over-optimization that fail to hold later. That pattern is overfitting: adding rules or free parameters until a historical test looks strong while the same procedure fails on later data.

Keep candidates as raw material

After goals are set, candidate ideas are treated as raw material and discarded if they do not remain acceptable over a test horizon typically given as five to ten years. Loading every rule onto the full history, optimizing all parameters and trading the best in-sample result is described as a common procedure that usually does not hold up.

Gate the procedure

Walk-forward analysis is a rolling out-of-sample check that re-estimates rules on earlier data and tests them on the next unseen segment so parameters are not locked to one historical sample. Monte Carlo simulation reshuffles an ordered trade or return sequence to produce a range of paths, especially drawdowns, instead of judging viability from one equity curve. The archive uses that reshuffling of trades to inspect drawdown outcomes that a single equity curve can understate, and it applies viability criteria at those simulation points.

Incubate, combine and watch for retirement

A candidate that looks viable is parked for later review. That pause is incubation: a planned wait during which the candidate is watched, with possible walk-forward reoptimization, and rechecked after about six months before it is treated as still intact.

Viable systems are then combined with others to avoid doubled risk and high correlation. System combination means running several mechanical procedures together across markets, styles or holding periods so risk is not concentrated in one curve. After combination, live procedures are monitored for retirement against their stop-trading rules.