A journal you only fill in after wins can only say you are doing fine

A trader opens his journal after a good day. A good day is when you feel like writing: the screen is green, the setups were clean, and the entry nearly writes itself.

After a bad day he closes the laptop. He will fill it in tomorrow. Tomorrow has its own session.

Nothing in the journal is false. Every row is accurate. That is what makes it dangerous.

What the log says, and what the account says

Take twenty sessions for one trader, written in R. Thirteen winners and seven losers. He logged all thirteen winners and the two smallest losers, and let the five worst sessions go unrecorded.

The journal reads fifteen sessions, thirteen of them winners, up 8.9R. The account reads twenty sessions, thirteen of them winners, down 1.5R.

Both sentences are true of the same twenty sessions.

Why the bad days drop out first

Writing is a reward after a good day and a chore after a bad one. The worst sessions are also the ones that leave you with the least energy and the most reluctance to look, so the gaps are not random. They gather exactly where the lessons are. The stop that was moved, the size that was too big and the trade that was never on the list tend to live on the days nobody wrote up.

So a journal with its bad days missing is not a thin version of the truth. It is a tilted one. Every figure worked out from it, the win rate, the average loser, how often you followed the plan, is an average of the days you were willing to write down.

What a log like that can never tell you

It cannot tell you how big your worst loss was, because that is the session most likely to be missing. It cannot tell you whether your rules hold on a bad day, because a bad day is where they get broken. And it will tell you, with complete confidence, that you follow your plan nearly every time you write it down.

That is a fair description of a trader who is doing fine. It is a poor description of one who is not.

A blank row is not a missing row

The fix is not a promise to write more. A promise relies on the same tired person who closed the laptop.

The fix is a row that exists before the result does. When you open the platform, the session gets a line in the journal with the date on it and nothing else. When the session ends you fill in the result, and if you cannot face it that evening you write a question mark and leave the line there. A blank row is a fact on the page. A missing row is a gap nobody can see.

The question mark costs ten seconds. It is also the only thing that stops a bad Tuesday from quietly becoming a Tuesday that did not happen.

The check to run tonight

Count the trading days on your broker statement for the last month, then count the rows in your journal, and subtract. For each day with no row, read the statement and write down what that day made or lost.

Then compare the total of the missing days with the total of the logged ones. If the two have the same sign, your journal is a fair sample. If they do not, you have found the most important thing in your log, and it is the thing the log was not holding.

Then put those days back. For each one, write a row from the statement: the date, the result in R, and one line on what happened. It is not a diary entry. It is the worst sessions rejoining the sample, so that the next time you work out a win rate or an average loser, the number is about the account you have and not the one you wrote down.

It is a close cousin of a green day that was still a bad day, which is about grading the sessions you do record, and it needs a good run of rows before your numbers mean anything.

If the journal says up 8.9R and the statement says down 1.5R, no entry in the journal was a lie. The journal was silent about five days, and those five days were the account.