Why Can't I Follow My Trading Plan?

You wrote the plan on Sunday. It was a good plan. You were calm, you had the whole week in front of you, and every rule in it seemed obviously correct.

On Tuesday at 10:14 you did something else.

Then you did the thing everyone does. You wrote the plan again, slightly stricter, with a line at the top in capitals about discipline. And on Thursday, at roughly the same time of day, you did something else again.

The plan was written by someone who was not there

Here is the structural problem, and almost nobody names it.

The plan is written by a version of you with no position on, no P&L moving, no clock running, and no gap between where your account is and where you expected it to be. That version is thoughtful, patient, and completely unbothered. That version is also not the person who has to execute it.

The person executing is in a different state entirely. Not a worse person. A different one, running the same machinery under load. Sunday-you writes rules for a trader who does not exist at 10:14 on Tuesday, and then Tuesday-you gets blamed for not being him.

This is why the standard fix fails so reliably. You rewrite the plan harder. But rewriting is a Sunday activity. You are sending an even more demanding set of instructions to the same operator, under the same load, and expecting a different result.

Three reasons a rule does not hold

They feel identical from the inside. They are not, and the fix for each is different.

The rule was never yours. You took it from a book, a course, or a trader whose results you admire. It fits their wiring. It may be actively wrong for yours. "Never take a trade in the first fifteen minutes" is excellent advice for someone whose edge is patience and terrible for someone whose entire edge is decision speed at the open. Copying a rule is copying the output of someone else's self-knowledge. What made it work was the fit, and fit does not transfer.

The rule is right but the cost is wrong. Some rules are correct and expensive. A rule that requires you to override an instinct every single time is a rule that will be paid for out of a budget that runs out. It holds for nine sessions and fails on the tenth, and the tenth is the one you remember. That is not a character arc. It is a rule doing damage every time it works.

The rule is not actually a rule. "Take good setups." "Do not overtrade." "Be patient." These are wishes wearing a rule's clothing. They contain no condition, so at the moment of decision they can mean anything, and whatever feeling is loudest gets to define them. You did not break that rule. It was never load-bearing.

What the research points at

Brad Barber and Terrance Odean spent years on the question of what actually separates traders. Their finding, across enormous samples of real accounts, was not about strategy selection. The reliable predictor of underperformance was how much people traded and the confidence behind it. Same instruments. Same information. Different operators.

That is the whole thesis in one sentence: the instrument was rarely the problem. The operator was.

Which reframes the plan question completely. If the differences between traders live in the operator rather than the method, then a plan written without reference to which operator is running it is missing the only variable that mattered.

The test that separates them

Next time you break a rule, do not write "need more discipline." Ask one question instead:

Would a trader wired differently from me have found that rule easy?

If yes, you have a fit problem, not a discipline problem. The rule is asking your machinery to do something it is bad at, and no amount of resolve makes a weakness into a strength. Change the rule, or build scaffolding so the rule does not depend on you in the moment: brackets placed at entry, alerts instead of watching, a checklist that carries the load your working memory cannot.

If no, and it would have been hard for anyone, then it is probably a cost problem. The rule is correct and it is expensive, so it needs to be made cheaper rather than mandatory. Fewer decisions. Shorter session. Automate the part you are paying for.

And if you cannot state the rule as a condition and an action, it was never a rule. Fix that before anything else, because it is the cheapest problem on the list.

The plan that survives Tuesday

A plan that holds is not a stricter plan. It is a plan built for the operator who will actually be sitting there, with the machinery he actually has, under the load he will actually be under.

That is a harder plan to write, because it requires knowing something about yourself that is not visible from the inside. You cannot introspect your way to your own decision speed or your own risk dispositions any more than you can introspect your own blind spot. That is what makes them blind spots.

But it is the only version that does not need you to become a different person by Tuesday.