Write and Change Your Rules Only When the Market Is Closed

“Enter only on a break of the morning high, with the stop under the opening range.” Say that is rule three in your plan. You take two trades on it before lunch, both stop out, and by 12:30 rule three has become “wait for a pullback after the break,” typed into your notes between two charts with a third setup forming on the screen.

That edit was made by the two losses. It was not made by you.

The last trade writes the mid-session edit

A rule exists to take a decision away from the version of you that is in a trade. That is its entire value. When the market is open you are holding positions, watching prices tick, and carrying the feeling of whatever just happened, and every one of those things pulls on judgment in a direction that has nothing to do with whether the rule works.

After a loss, a rule looks broken. After a win, it looks like it could be pushed harder. Neither impression comes from evidence. Both come from the most recent outcome, which is the smallest and noisiest sample available, and a rule rewritten on the strength of it tends to be rewritten again after the next outcome, until the plan is a record of your moods.

Blunt version: if the market is open, the rule stands.

Two losses prove almost nothing

Put numbers on rule three. Suppose, hypothetically, that your records for it over a longer run look like this.

On those numbers, two straight losses turn up in about a third of all consecutive pairs. They are the rule behaving normally. A trader who abandoned it at 12:30 would be walking away from a hypothetical edge of 0.2R a trade at exactly the moment its ordinary losing streaks are doing their job, and would then be trading a replacement rule with no record at all.

The figures here are invented for the example. Your own will differ. The point carries over anyway: a judgment about a rule needs a sample, and the sample must be decided before the results arrive, or the losses will decide it for you.

Write rules that leave nothing to interpret

The closed market is also the time to write the rules in the first place, and to write them tightly. A vague rule is a rule you will end up changing mid-session without admitting it.

Take “cut losers quickly.” During the session, “quickly” gets defined by how the position feels, so the rule is renegotiated on every trade and never technically broken. Compare “exit at the stop entered with the order; the stop may be raised and never lowered.” There is nothing in that sentence for a nervous trader to reinterpret at 11:15, which is the whole reason for writing it that way, and it can be checked afterward against the order history, line by line, without any argument about what was meant.

A useful test: could someone else follow the rule from the written words alone? If not, rewrite it tonight.

Keep a rule-change log

Every change goes into a log with four fields:

• Date of the change.
• Rule, old wording and new wording, both written out in full.
• Reason, in one sentence, of the kind you would use for a one-sentence trade thesis.
• Evidence: the trades, the count, and the result that justify it.

The evidence field is the one that does the work. If you cannot fill it with something larger than today’s session, the change waits. Trades you did not take belong in the evidence too, since a rule that keeps you out of losers is working even when it feels idle, and a skipped-trades log is where that record lives.

Over months the log becomes its own check. You can see how often you change rules, which ones you keep reversing, and whether changes made after bad days look any different from changes made after quiet ones.

Changes start next session at the earliest

Write the change after the close. Read it again before the next open. It takes effect then, and not a minute before. If the morning read leaves you less sure, leave the old rule in place and let the change sit in the log as a proposal until the evidence field is fuller.

The delay does two things. It puts a night between the outcome and the decision, which is often enough for a change that seemed urgent at 3:55 to look unnecessary at breakfast. And it keeps each session internally consistent, so every trade in a day was taken under one set of rules and your records can be read cleanly afterward. After a truly bad day, the questions in what to do after a big losing day come first. Rule changes come later.

The one exception

A rule that is plainly broken can be suspended at once. If your platform is misreporting positions, your data feed is frozen, or a rule refers to an order type your broker has stopped accepting, stop using it now, trade smaller or not at all, and write the proper fix after the close. Log the suspension like any other change, with the date and the fault, so the record shows why the rule was set aside and when it came back. Everything short of that waits for the bell.