Following a Rule You Currently Disagree With
There is a particular moment that decides whether a trading rule is real. The session is in front of you, the rule says do not take this, and everything you can see says the rule is wrong about this specific case. You are not confused and you are not being lazy. You have looked and formed a view, and the view conflicts with the instruction you wrote. What happens next is the only thing the word discipline usefully describes.
The Disagreement Is Built Into the Design
A rule exists because a decision made calmly and in general is expected to outperform a decision made under pressure and in particular. That is the whole premise. It follows directly that there will be occasions when the general rule is wrong about the particular case, and it follows just as directly that you will be able to see it.
So the disagreement is not a sign that something has gone wrong. It is the situation the rule was written to handle. A rule that only ever agreed with your in the moment judgement would be doing no work at all, since it would never change an outcome. The friction is the mechanism, not a defect in it.
You Cannot Grade Your Own Judgement Live
The uncomfortable part is that the feeling of seeing something clearly is not evidence that you are seeing it. That feeling is present when you are right and equally present when you are wrong, and it is amplified by having a position, by having missed the previous move, and by having had a bad week.
This is why an appeal to your own certainty is not a good reason to override. If confidence were correlated with correctness in a way you could rely on in the moment, the rule would be unnecessary. Your log is where the correlation can be checked, and it is checked afterwards, which is exactly when the answer is no longer useful for the trade in question.
Follow It Without Agreeing With It
A common mistake is thinking that compliance requires being convinced. It does not, and trying to talk yourself into agreement is usually a losing effort that leaves you resentful and half committed. The workable position is to keep the disagreement and act on the rule anyway.
Practically that means noting what you think the rule is getting wrong, in writing, at the time, and then doing what the rule says. The note takes a few seconds and does two useful things. It gives the objection somewhere to go other than into the decision, and it creates a record that can be tested later against what the session actually did.
Let the Record Settle It
Over a run of sessions those notes become a dataset about your own overrides, which is a rare and valuable thing to own. Each one records what you believed at the time, and each can be checked against the outcome that followed.
Most traders discover that the notes are considerably less impressive in aggregate than they felt individually. Some discover a genuine pattern where their objection is consistently right in one identifiable circumstance. Both results are useful. The second one is the legitimate route to changing the rule, and it is available only if the objections were recorded rather than acted on.
Amend the Rule Later, Never Now
If the evidence eventually supports your objection, change the rule. That is what revision is for, and a rule that survives contrary evidence indefinitely has become a superstition. The requirement is that the change happens away from the market, on the basis of the accumulated record, and applies from that point forward.
Changing it during the session it inconveniences is not revision, whatever it is called at the time. The reasoning may be identical and the conclusion may even be right, but the process has been replaced by improvisation, and once the document can be edited under pressure it no longer constrains anything. Keeping the disagreement alive and unresolved until the weekend is less satisfying than settling it immediately, and it is the entire difference between a plan and a preference.