How to keep a trading journal that actually works?
A trading journal is neither a hobby nor paperwork — it is the only instrument that answers the question "why am I losing" rather than merely "how much have I lost". Without one you cannot tell a bad strategy apart from a good strategy executed under emotional pressure, and those are two different problems with two different repairs. Your broker statement records the facts but says nothing about the circumstances in which those facts were produced. What follows: which fields are genuinely worth writing down, how to build a spreadsheet template, and what to ask the journal once a week.
Why keep a journal at all?
Without a journal the month ends like this. The account is eight percent under water and the question "why" has no answer. Memory will not help, because memory is selective — what survives is the two loudest trades, one win and one painful loss, not the thirty decisions the month was actually made of. On top of that sits the mechanism psychologists call retrospective rationalisation: a week later every entry you took feels justified, because the mind manufactures a reason for whatever it has already done.
With a journal the month looks different. You open the spreadsheet and thirty trades sit in front of you with their context attached: the hour, the pair, the reason for entering, the state you were in. Only then can you ask sensible questions and get sensible answers. This is not about impressive statistics; four or five simple breakdowns usually cover it.
- What share of your trades is opened on a Friday evening, tired and into thin liquidity — and how those trades compare with the rest of the week.
- How many entries followed immediately after a loss, the classic pattern of trading to get even, and what they add up to.
- How each strategy performs on its own — because one of them often earns while another gives it all back, and the balance shows only the difference.
- What your real, rather than planned, reward-to-risk ratio is: if the average stop is 28 pips while the average realised gain is 42 pips, you are trading at roughly 1 to 1.5, even though the plan said 1 to 2.
Each of those breakdowns leads to one concrete change in behaviour. Without a journal you will see none of them, because there is nothing to compute them from.
Ten fields worth logging on every trade
Ten fields sounds like a lot; in practice it is thirty to sixty seconds per trade. The hardest one is the ninth, because it demands honesty with yourself. Most beginners write "calm" beside every entry, unwilling to admit they went in out of fear of missing the move. After a month of honest logging, naming your own state gets far easier — you simply have the vocabulary for it.
A practical spreadsheet template
Start with an ordinary spreadsheet: columns A through J correspond to the ten fields in the table above. Add five helper columns that only start earning their keep after a few dozen entries.
- Result in account currency — the result in pips multiplied by the pip value and the position size.
- Result as a percentage of capital — the previous column divided by the account balance at the moment the position was opened.
- Strategy name — if you trade three setups, give them short labels and stick to them.
- Did it match the strategy? — yes or no, answered honestly. This is one of the most valuable columns in the whole sheet.
- Note after closing — one sentence on what you would do differently if the same situation turned up tomorrow.
That is the minimum and it will comfortably last several months. Later it is worth adding a pivot table that filters results by pair, hour and emotional state — at that level of detail the patterns surface on their own.
"Without a journal you trade emotions you have labelled intuition. With a journal you see what that intuition actually looks like — and the view is rarely flattering." — Jarosław Wasiński, forex-podstawy.pl, 2026.
Four questions for the weekly review
Once a week — Sunday evening, or Friday after the close — open the journal covering the last five days and put four questions to it. The whole review should take half an hour.
First question: what do the three largest losses have in common?
Pick the three worst trades of the week and compare their fields: pair, hour, state on entry, reason for entering, whether they matched the strategy. You are looking for the overlap. Very often it turns out all three fell on a Friday evening, all three had haste written beside them, and none of them followed the strategy. That is your weak spot for this particular week — a single one, named, and closeable with a rule.
Second question: what do the three best trades have in common?
The same procedure in the other direction. The three biggest wins: when they happened, under which strategy, in what state. The typical picture is the European session, calm, and the one setup you know best. That is your edge — not something to admire, but something to reproduce by deliberately recreating the conditions in which it shows up.
Third question: which emotional states cost you the most?
Add up the results grouped by the state you recorded on entry. Below is a sample month; the numbers are an illustration rather than a norm, but the proportions look like this for a surprising number of people.
The conclusion from a table like that is not "I must trade better" but "I am not allowed to open a position in these two states". That is the difference between a resolution and a rule.
Fourth question: what am I changing next week?
One change. Not five, not ten — one, written into the journal as a rule you can check with a simple yes or no. For instance: "no new positions on Friday after 19:00", or "after a loss larger than one percent of capital I take an hour off". The next review opens with the question of whether the rule was kept. If it was not, it stays for another week instead of having anything new piled on top of it.
What to avoid when keeping a journal
- Filling the journal in once a week, from memory. Memory supplies justifications. The entry about your reason and your state has to be written at the moment the position is opened, not on Sunday.
- Leaving out losing trades. It is tempting to treat an "unfair" loss as though it never happened. A journal missing even a handful of entries stops being data and becomes a story you tell about yourself.
- Vague entries in the "reason" field. "Setup", "intuition", "nice move" are of no use to anyone. Something concrete reads: "bullish engulfing on the hourly, RSI above 50 on the daily, price at weekly support". Without that you cannot measure how a setup actually performs.
- Skipping the weekly review. Logging without analysis is work with nothing collected at the end. The review is what turns data into knowledge — and it, rather than the column-filling, decides whether the journal changes anything at all.
What to do this week
- Build the spreadsheet today. Export the last thirty trades from your platform to a CSV file, open it in a spreadsheet and add the four missing columns: reason for entering, state on entry, strategy name and the note after closing. Fill in the historical rows only where you genuinely remember; leave the rest blank, because an empty field is more honest than an invented one.
- Set up a note template on your phone. In any notes app, save five labels to fill in right after you click the order: instrument, reason, state, risk in percent, hour. It takes twenty seconds, and that habit is what decides whether the journal survives its first month.
- Fix a permanent slot for the review. Put a recurring thirty minutes in the calendar — Sunday at 19:00 works for most people. A review without a fixed time does not happen at all, and the review is where the whole effect comes from.
- After the first review, write down one rule. One, drawn straight from the data, phrased so that compliance can be checked. Keeping it for a week is worth more than a list of five resolutions none of which survives. If you want to set that within a wider daily order, we describe it in the piece on building a system of trading discipline.
After three months of consistent entries you will start catching your own mistakes before you click the order — you will simply recognise the situation you have already described a dozen times in the "what would I do differently" column. The journal does not become redundant at that point, but it stops feeling like effort. That is a good moment not to abandon it.
Sources & bibliography
-
Van Tharp Institute About Van Tharp — position sizing and trader development · IITM founder profile www.vantharp.com ↗
-
CFA Institute Behavioral finance — performance attribution and journaling · CFA Program curriculum overview www.cfainstitute.org ↗
-
Brett Steenbarger The Psychology of Trading · 2003, klasyka analizy psychologicznej tradera en.wikipedia.org ↗
Frequently asked
An ordinary spreadsheet or a dedicated journaling app?
Start with an ordinary spreadsheet, and not out of thrift. Building your own template forces you to work out which fields you genuinely need and which ones you are copying because you saw them somewhere else. A ready-made application makes that decision for you, before you have had the chance to understand what you are looking for in your own trading.
Paid journaling tools do exist and some of them import your platform history automatically, which saves a few minutes a day. Prices and features change often enough that they have to be checked with the provider before you buy. A sensible moment to move across is the third or fourth month of consistent entries — by then you know what you expect and can judge whether a given application delivers it. For most retail traders a spreadsheet with a well-thought-out template lasts a long time.
Isn't MT5 itself a journal?
No. The platform records facts — the opening price, the closing price, the result — but says nothing about the context: why you entered, what state you were in, and how you judged the trade once it was closed. Without that context the line "100 USD loss" is neutral information: you cannot tell whether the strategy failed, its execution failed, or you simply went in hastily where you had no business being.
The journal exists precisely to add the layer the platform has no access to. The simplest arrangement: once a week export your trade history into a spreadsheet and add four context columns by hand. The export takes under a minute, and filling in the context — provided you noted it as you went rather than from memory — takes a few minutes more.
How much time does journaling take?
Two minutes per trade, three at the most: half a minute to record the context as the position is opened and about as much again for the note after closing. Add half an hour a week for the review. At thirty trades a month that comes to roughly two hours a month.
That is very little next to the time most people spend hunting for another indicator or another strategy. The journal gives you information you will not find in any course or on any forum, because it concerns nothing but your own decisions and your own circumstances. It is worth adding that those two hours are spread across the whole month in chunks of under a minute — the difficulty is not their length but their regularity.
What if I don't remember the emotion from a trade a week ago?
Do not reconstruct emotions from memory — it simply does not work. The mind manufactures a justification for a decision it has already taken, so a week later every entry is remembered as considered, including the one that came out of haste. A record rebuilt after the fact is a mirror showing a version more comfortable than the truth.
The only solution is to note it as you go. Set up a note template on your phone with five labels — instrument, reason for entering, emotional state, risk in percent, hour — and fill it in as soon as the order is placed. It takes about twenty seconds, and it is the only moment when the honest answer is available to you. If an entry was missed, leave the field blank rather than guessing: missing data is honest, an invented entry ruins the whole set.