Keeping an investment decision journal that is worth re-reading

Why the outcome is the wrong thing to grade

A position that went up was not necessarily a good decision, and one that went down was not necessarily a bad one. Markets are noisy enough over any single holding period that outcome alone tells you almost nothing about process.

The only way to grade a decision honestly is to compare what you expected at the time against what actually happened. That requires a record of the expectation, written before you knew the answer, and human memory is not that record. Memory quietly edits itself to match the outcome, which is why an investor who sold at the bottom will often recall having had doubts all along.

What to write down at the moment you act

Four things, and it should take two minutes. If it takes longer than that you will stop doing it by the third entry.

• What you did. Bought, sold, added, trimmed or deliberately held. Holding through something is a decision and belongs in the record.
• The price and the size. Both, not one. A price without a quantity cannot become a position, and a year later you will want to know whether the conviction matched the sizing.
• Why, in your own words. One or two sentences. The test is whether a stranger reading it could tell what you were relying on.
• How sure you were. A number on a fixed scale. Its value is not the number itself but the series: seeing conviction drift up while evidence drifted down is the pattern worth catching.

The three fields most journals leave out

What would change your mind. Written at the time of the decision, not afterwards. This is the field that converts a diary into something you can be held to, and it is the one people skip. See kill criteria for how to write ones that actually work.

What you were reading. The filing, the call, the article. Six months later the question “did I actually check this or did I absorb it from somewhere” is unanswerable without it.

What you chose not to do. The decision not to sell during a drawdown is the one you will most want to review, and it leaves no trace in a brokerage statement. A journal that only records transactions misses every occasion where the right answer was to sit still.

Reading it back

A journal earns its keep at review time, not at writing time, and the review is a specific question rather than a general re-read: for each entry, was the reasoning wrong, or was the reasoning fine and the outcome went the other way?

Those two failures need opposite responses. Faulty reasoning means changing how you decide. An unlucky outcome from sound reasoning means changing nothing, which is the harder discipline. Without a written record of the reasoning you cannot tell them apart, and you will end up rewriting a good process because it produced a bad quarter.

The other thing worth reading back is the conviction series against the evidence. Conviction that rose while the business deteriorated is the single most useful pattern a journal can show you about yourself.

Where the format falls down on its own

A journal is only as good as the reviewing, and reviewing depends on you remembering to do it. In practice the entries get written and never opened again, because nothing prompts a re-read at the moment it would be useful, which is when the company publishes something that bears on what you wrote.

That is the gap Okidor is built to close: the journal sits beside the thesis, and new filings, earnings calls and material news are read against the reasoning automatically, so the prompt to re-read arrives when there is a reason to. You can see a worked example without an account, or read how the engine decides.

None of which requires the tool. A notebook and a calendar reminder will do the same job if you keep to it.