A trading journal is not an account statement. Your broker already does that, for free and better than you. If your journal just copies out your entries, your exits and your P&L, you have built an archive: something you look at once and never open again.

The journal that actually moves you forward answers one precise question: what is it in my behaviour that produces my results? Not "how much did I make", but "what state was I in when I made money, and what state was I in when I gave the day back".

This article lays out a concrete method for keeping that kind of journal, based on three years of iterating on my own.

Why most trading journals are abandoned within three weeks

The pattern is always the same. You open a spreadsheet, you create fifteen columns, you fill it in religiously for twelve sessions, and then one Monday you don't. Tuesday neither. The file dies.

Three causes come up every single time.

The journal is too heavy. Fifteen fields per trade means twenty minutes of typing after a session that has already drained you. The cost of keeping it outweighs the perceived benefit, so you stop. A journal that takes more than five minutes a session does not survive.

The journal gives nothing back. You fill it in, you fill it in, and a month later you have data but no answers. A journal that never produces a finding is a chore with no reward.

The journal only logs numbers. And this is the big one. Your P&L, your average R, your winrate are consequences. They don't tell you *why*. Two sessions can both show −€450: in the first you followed your plan and the market simply didn't agree, in the second you took three trades outside your plan after a loss. Those are radically different sessions. A numbers-only journal displays them as identical.

If your journal can't tell a clean loss from a dirty one, it can't teach you anything.

What a trading journal really needs to record

Split it into two layers. They serve different purposes and get filled in at different moments.

The execution layer: what you did

This is the factual minimum, per trade. Keep it short:

  • The instrument and the direction
  • Entry time and exit time
  • The setup traded, picked from a closed list (if you can write anything you like, you will never aggregate anything)
  • The risk taken, in R rather than in euros
  • The result, in R
  • One binary box: was this trade in the plan, yes or no

That last box is worth ten columns on its own. It is what will later let you separate your "clean" performance from your actual performance.

The trader layer: what state you were in

This is the one almost nobody keeps, and the one that produces the real discoveries. Once per session, not per trade, in thirty seconds:

  • Mood before the session, out of 5. Rested, tense, rushed, still angry about yesterday.
  • Mood after the session, out of 5.
  • Execution discipline, out of 10. Did I take my entries as planned, or early, or late?
  • Plan adherence, out of 10. How many trades outside the plan, how many stops moved?
  • Emotional control, out of 10. Did I get dragged around by my emotions or did I work with them?
  • My personal rules: three to five rules, ticked or not. "No more than 3 trades", "No trades before the cash open", "I stop at −2R on the day".

These scores are subjective. That's not a flaw: it's exactly the point. A subjective score given honestly, session after session, by the same person, becomes a consistent series and is perfectly usable.

The method: three moments, five minutes a day

A journal doesn't get kept "when I have time". It gets kept at fixed moments, or it doesn't get kept at all.

Before the session — 2 minutes

You log your mood, you run your pre-session checklist, you reread your rules. The point isn't administrative: it's to make you aware of your state before you put money at risk. A trader who wrote "2/5, slept badly, wound up" at 8:55 rarely makes the same decisions at 9:30 as a trader who never asked himself the question.

During the session — 10 seconds per trade

Execution layer only, and above all the "in plan / out of plan" box. Tick it at the moment of the trade, not in the evening. By the evening your memory has rewritten the story: an impulsive trade that happened to win becomes, in hindsight, "a good read of the market".

After the session — 3 minutes

The trader-layer scores, then three to five lines of free text. Not an essay: one observation. "I got back in 40 seconds after my stop because I didn't want to finish red." That's the kind of sentence that, reread thirty times over two months, eventually changes a behaviour.

How to read your journal: the four cross-checks that matter

Keeping it is only half the job. Without a structured review you accumulate without learning. At the end of every week, and in more depth at the end of every month, make your data talk by cross-referencing it.

1. Results of "plan respected" sessions versus "plan broken"

The founding cross-check. Split your sessions into two groups: the ones where every rule is ticked, and the rest. Compare the average result.

The gap is almost always brutal. Plenty of traders discover at this point that they are profitable on their clean sessions and that the whole of their annual losses comes from a minority of out-of-plan sessions. That finding doesn't call for a new strategy: it calls for eliminating one behaviour.

2. The cost of each broken rule

For each rule, compare the results of sessions where it's ticked with those where it isn't. You get a price, in R, for every rule you break.

This is the most actionable piece of information a journal can give you. It turns "I should be more disciplined", which is not an objective, into "my 3-trade maximum rule costs me 1.8R every time I break it, and I break it one session in three", which is.

3. Pre-session mood versus result

Group your sessions by starting mood. Plenty of traders find a clear threshold: below a certain level, the session is statistically a losing one. This isn't armchair psychology, it's your own data series. And it leads to a simple, measurable rule: below the threshold, reduced size or no session at all.

4. Your weakest axis

Of your three behavioural scores, one is systematically lagging. That's your work item for the next eight weeks. One, and only one. Working on three axes at once means working on none of them.

The mistakes that make a journal useless

Logging cold, three days later. Your scores will be smoothed by the outcome. A trader who ends the week green retroactively scores every session higher than it really was.

Scoring yourself generously. A journal is read by you and nobody else. If you give yourself 8/10 for discipline on a session where you moved two stops, you have simply manufactured false data, and one day you'll pay the price of your own leniency.

Changing format every month. Data only becomes usable as a long series. A journal reworked after every disappointment never builds enough history to reveal anything. Keep the same format for six months, whatever happens.

Confusing volume with quality. Fifty lines of commentary per session are worth less than three honest lines, logged every single day for six months.

How long before a journal produces results?

Allow around twenty sessions for the first cross-checks to become readable, and two to three months for a behaviour to actually change. The sequence is always the same and there's no shortcut: first you notice it, then you watch yourself do it again knowing full well what you're doing, then you stop before you do it.

That timeline is precisely why keeping the journal has to be cheap. A journal that takes five minutes will last six months. One that takes twenty will be abandoned before it has served any purpose at all.

Spreadsheet or dedicated app?

A spreadsheet works, and it's perfectly fine to start with. Its limit shows up at review time: the cross-checks in the section above need pivot tables that you will not rebuild every week. In practice, the review stops happening, and a journal you don't review is just an archive again.

That's exactly the problem Edgio solves: the trader layer — mood, discipline, plan adherence, personal rules — is logged in thirty seconds, and the cross-checks between those scores and your results are computed continuously. You don't have to build the analysis: you just read it.

Whatever tool you use, the principle stays the same, and the principle is what counts: log the trader, not just the numbers. You already have the numbers. What you're missing is what explains them.