The 100-Trade Laboratory

One profitable trade proves almost nothing. Ten trades can still be a flattering accident.

The purpose of a 100-trade laboratory is not to make the number magical. It is to create a sample large enough to reveal behaviour, costs and regime dependence while preventing constant reinvention.

Freeze the protocol:

Market and session.
Setup definition.
Entry and exit rules.
Instruments allowed.
Risk per trade.
Maximum portfolio heat.
Data to record.
Conditions that invalidate the experiment.

Do not change the rules because the first twelve trades disappoint you. A change creates a new experiment and resets the sample.

For every trade, record:

Date and market condition.
Screenshot before entry.
Exact reason for entry.
Planned and severe risk.
Execution price and costs.
Exit and result in R, where one R equals planned risk.
Whether every rule was followed.
One short emotional note.

Separate strategy result from operator result.

Strategy P&L asks whether the rules produced a positive expectancy.
Execution P&L asks what the results would have been if every rule had been followed.
Behaviour gap is the difference.

A profitable strategy with poor execution may be salvageable through process. A losing strategy with perfect execution needs a new hypothesis. A profitable result created by repeated rule-breaking is dangerous because it rewards the wrong behaviour.

Review in blocks of twenty, but do not optimise every block. Examine expectancy, average win, average loss, win rate, maximum drawdown, costs, time of day, market regime and rule adherence. Look for one or two large outliers. If removing one trade destroys the result, the evidence is fragile.

Use out-of-sample thinking. If the setup was discovered in past data, test it on a later period or different but relevant market without modifying it. The aim is not to find a perfect equity curve. It is to discover where the idea stops working.

At trade 100, choose:

Reject — expectancy remains negative after realistic costs.
Refine — a specific, economically plausible weakness is visible; begin a new protocol.
Continue small — expectancy is positive, execution is stable and the result does not depend on one outlier.

Do not scale because you are excited. Scale by a written rule after evidence, and increase risk gradually enough that psychology remains part of the same experiment.

The laboratory produces an asset even when the strategy fails: a record of how you behave under uncertainty. That knowledge can improve investing, business and decision-making far beyond trading.

The Field Note

Create the journal before the next trade. Write the protocol at the top and number rows 001–100. If you cannot commit to recording the losing trades with the same care as the winning ones, do not begin.

CHOOSE THE NEXT MOVE

01If rule adherence is below 90%, do not evaluate or scale the strategy; repair execution.

02If expectancy is positive after costs and evidence is robust, increase risk only under a new written scale rule.

Sources & Swiss context

ESMA retail CFD evidence: https://www.esma.europa.eu/pl/press-news/esma-news/esma-agrees-prohibit-binary-options-and-restrict-cfds-protect-retail-investors | FINMA public investor protection: https://www.finma.ch/en/finma-public/

Editorial education, not personalised investment, legal or tax advice. Swiss rules, limits and product terms can change; verify current information before acting.