How to keep a forex trading journal

Most forex trading journals fail for the same reason. They are written after the fact, by the person who already knows how the trade turned out. That person is not a neutral witness. A loss becomes “bad luck, the setup was fine.” A win becomes “good read,” even when you moved the stop twice to get it.

A journal that records what you thought after seeing the result is not a record. It is a story. And you cannot improve against a story.

This is a practical guide to keeping a forex trading journal that avoids that: what to log, when to log it, how to review it, and which numbers actually tell you something. It assumes you want the journal to change how you trade, not to sit in a folder.

What a trading journal is for

A trading journal is a record of every trade you take, with enough context attached that you can later group those trades and compare them. That last part is the whole point. A list of entries and exits is a statement — your broker already has one. A journal earns its name when it lets you ask which of the things I do actually make money and get an answer that is not a matter of opinion.

So the test for any field you consider logging is simple: will I ever group by this? If you will never sort your trades by it, it is decoration.

What to record on every trade

Two categories. The first is what happened, which your platform already knows. The second is what you decided, which only you know, and which is where every useful insight comes from.

The mechanical fields

Pair, direction, entry, exit, stop, target, position size, date and time, and the resulting profit or loss. If you trade MT4 or MT5 you can export these rather than type them — a statement export gives you months of history in one go, and there is no virtue in retyping what a CSV already contains.

The decision fields

These are the ones that matter, and they are the ones people skip:

• Setup — the name of the pattern you traded. Not “it looked good.” A name you use consistently, so trades with the same name are genuinely the same thing.
• R multiple — the result expressed in units of the risk you took, not in currency. A £200 win on a £100 risk is +2R.
• Plan adherence — did you follow your own rules, yes or no. One field, brutally binary.
• Session — London, New York, Asia, or the overlap.

Those four carry most of the signal. We argued that case at length in the four columns that actually change how you trade, and the short version is that a journal with thirty fields gets abandoned in three weeks while a journal with six gets kept.

Log the trade before you take it

This is the single change that separates a journal that works from one that does not.

Write the entry before you click buy. Setup name, entry, stop, target, size, and the reason. Thirty seconds. Then take the trade, and when it closes, add only the outcome.

Doing it this way makes the record falsifiable. You wrote your stop down before you knew whether it would be hit, so “I would have held” is off the table. You named the setup before the outcome could colour it. And plan adherence becomes a fact you can check rather than a feeling — you either did what the note says or you did not.

There is a second benefit that is harder to measure and possibly larger. Having to write the setup name down before entering will occasionally stop you entering. If you cannot name it, you are not trading a setup. You are clicking.

The numbers worth reading

Once you have fifty or so trades logged consistently, the journal starts to answer questions. These are the figures that carry weight, roughly in order of how often they are misread.

Win rate — necessary, and almost useless alone

Win rate is the most quoted number in trading and the least informative. A 30% win rate is excellent if your winners run at 4R and your losers stop at 1R. An 80% win rate is a slow death if you let losers run. Win rate only means something next to your average win and average loss, and you should be suspicious of anyone quoting it on its own.

Average R, and the distribution behind it

Average R per trade is your expectancy in risk units. Positive means the strategy makes money at any size. Look at the distribution as well as the average — one +9R outlier can carry an otherwise losing set of trades, and a system that depends on outliers is a system you will abandon during the drought before the outlier arrives.

Profit factor

Gross profit divided by gross loss. Above 1.0 you are making money; below, you are not. Somewhere around 1.5 is a real edge, and anything much above 3.0 over a small sample usually means the sample is too small.

Maximum drawdown

The largest peak-to-valley fall in your equity curve. This is the number that tells you whether you can actually trade the strategy, because it is the one you have to live through. A system with a good expectancy and a 40% drawdown is a system most people abandon at 25%.

Read those together rather than one at a time. The win rate alone says very little; the win rate beside a +0.45R average and a 2.28 profit factor says the edge is real and the winners are meaningfully bigger than the losers.

How to review it

A journal you write but never read is just slower note-taking. Two cadences are enough.

Weekly, about ten minutes

1. Read every trade you took, in order.
2. Mark the ones where plan adherence was no. Do not rationalise them. Just count them.
3. Ask one question: what was the most expensive thing I did this week? Not the biggest loss — the biggest mistake. Those are frequently different trades.

Monthly, about half an hour

This is where grouping earns its keep. Split the month’s trades by each of your four fields in turn and look for the gap:

• By setup — which named setups are actually positive? Most traders find they have one that pays for two that do not.
• By session — are your London trades and your New York trades the same business? Usually not.
• By plan adherence — the one that stings. Compare average R on trades you followed the plan on against trades you did not. If the gap is large, you do not have a strategy problem.
• By pair — worth checking, though it is more often noise than the other three until you have a few hundred trades.

The output of a monthly review should be exactly one change for next month. Not five. One change, applied consistently, produces a readable result; five changes at once produce a mess you cannot attribute.

Five ways journals go wrong

• Too many fields. Thirty columns feels thorough and guarantees abandonment. Six you fill every time beat thirty you fill for a fortnight.
• Writing it after the close. Covered above, and it is the big one. Hindsight rewrites the reason.
• Logging only the interesting trades. The boring ones are the sample. Skipping them biases everything downstream, usually in a flattering direction.
• Screenshots instead of fields. A chart image is lovely and cannot be grouped. Take the screenshot, but put the decision in a field as well.
• Reviewing by reading, not grouping. Scrolling your trade list feels like review. It is recall, and recall is what the journal exists to replace.

Spreadsheet, template, or app

A spreadsheet is a completely legitimate place to start, and if you have one you actually maintain, do not let anyone talk you out of it. Its limits show up in two places: pre-trade entry on a phone is painful, and the grouping you want monthly means building pivot tables you will eventually stop rebuilding.

Notion and similar templates solve the entry problem and are worse at the arithmetic. You will end up maintaining formulas.

A dedicated journal earns its place when the grouping is built in — so the monthly review is reading a breakdown rather than constructing one. That is the part people stop doing, and it is the part that changes behaviour.

PipFlo is ours, so treat this as disclosure rather than a recommendation: it logs those four fields by default, imports MT4 and MT5 statements, and splits performance by setup, session, direction and plan adherence without you building anything. The free plan covers 50 trades, which is roughly the point at which the breakdowns start to mean something. There are also free position size and risk/reward calculators if you just want the arithmetic before a trade.

Starting from nothing

If you have never kept one, do not design the perfect journal. Do this instead:

1. Pick four fields: setup, R multiple, plan adherence, session.
2. Write the entry before each trade. Thirty seconds.
3. Add the outcome when it closes.
4. Do not review anything for the first month. Just build the sample.
5. At fifty trades, group by each field and find the one gap worth acting on.

Fifty trades is not a statistically robust sample and it is enough to see something obvious, which is usually all you need at the start. The goal is not certainty. It is to replace “I think I overtrade on Fridays” with a number.

Your next hundred trades are going to happen whether you write them down or not. The only question is whether you will be able to learn anything from them afterwards.