How to keep a trading journal: what to record, how to review it, and how it becomes an edge
A practical guide to keeping a trading journal: the fields worth recording for every trade, the plan → execution → review loop, and the weekly numbers that show where your edge actually is.
PnLTrack teamPublished: 9 min read
A trading journal is a record of every trade you take: what you planned, what you did, and what happened. Most traders have heard they should keep one. Far fewer do, and the honest reason is simple: reviewing your losing trades is unpleasant. A journal makes the mistakes you would rather forget impossible to ignore.
That discomfort is also the whole point. Memory is selective — it keeps the wins and smooths over the losses — and a journal is the only thing that stops you from arguing with your own results. Over a few hundred trades it turns vague impressions (‘I think breakouts work for me’) into measurable statements (‘breakouts return +0.4R per trade; counter-trend entries lose money’).
This article covers how to keep a trading journal in practice: what a journal is (and is not), what to record for every trade, how to run the plan → execution → review loop, which numbers to look at each week, and whether a spreadsheet or an app makes more sense for crypto.
What a trading journal is — and what it is not
A trading journal is not a P&L spreadsheet. Your exchange already shows your profit and loss; copying those numbers into a table adds nothing. A journal records the decision, the context and the outcome of every trade — three things a P&L statement never captures.
The decision is why you entered: the setup, the plan, the stop and the target. The context is the state you were in: the market conditions, the time of day, your emotional state after the previous trade. The outcome is what happened — not only in money but in multiples of the risk you planned to take. Of the three, your exchange knows only the outcome.
That distinction matters because a losing trade can be a good decision and a winning trade can be a bad one. A journal that only records results cannot tell the two apart; one that records decisions can.
Why traders keep a trading journal
Everyone has their own reason, but four come up again and again:
- Systemic mistakes become visible. One revenge trade feels like an accident. Fifteen of them in a month, tagged and counted, is a pattern you can fix.
- You measure the edge per setup, not by feel. ‘My breakout strategy works’ is a feeling. ‘38 breakouts, 45% win rate, +0.3R expectancy’ is a fact — and it may show that only one of your three setups makes money.
- Discipline: the plan is written before the entry. Writing the stop and target down before you click gives the trade a definition of ‘wrong’. You can still break the plan, but now it is a recorded choice, not an improvisation.
- Emotions become data. ‘Tired’, ‘FOMO’, ‘revenge’ and ‘bored’ are states you can tag. After a few months you can see which of them cost you money and simply stop trading in them.
What to record for every trade
The most common question is what a trading journal should include. The answer: less than you think, but consistently. These are the fields that earn their place:
| Field | Why it matters |
|---|---|
| Entry, exit, size | The raw facts. If the journal syncs with the exchange, they are filled in automatically and cannot be misremembered. |
| Fees and funding | Easy to forget, and on perpetual futures they add up. A strategy that is marginally profitable before fees is often losing after them; an exchange sync records both. |
| Setup / strategy | The name of the pattern you traded: breakout, pullback, range fade. Without it you cannot compare strategies against each other. |
| Planned stop and target → risk in R | The stop defines how much you planned to lose (1R); the target defines what you expected for taking that risk. Every later result is expressed in these units. |
| Reason for entry | One or two sentences on why now. Weeks later this reveals whether you followed the setup or invented a justification. |
| Emotional state | A word or a rating before the trade: calm, tired, chasing, confident. It becomes measurable after enough trades. |
| Mistake tags | A short, fixed list: entered early, moved the stop, oversized, no plan, held past the target. Tags are what you count in the review. |
| Screenshot of the chart | A picture of what you saw at the moment of the decision — not of what the chart looks like afterwards. |
A good rule: everything the exchange knows (prices, size, fees, funding, time) should come from the exchange, and everything only you know (the plan, the reason, the state, the tags) should take under a minute to fill in. If it takes longer, the journal will not survive the first bad week.
Before, during, after: the plan → execution → review loop
A journal works best as a loop rather than a diary filled in afterwards. Before the trade you write the plan: setup, entry, stop, target, size. During the trade you execute it — or record the moment you deviate. After the trade you compare the two and tag what happened.
The unit that ties the loop together is R. 1R is the money you planned to lose if the stop hits: the distance from entry to stop, multiplied by the position size. Results are then written as multiples of it: a trade that made twice what it risked is +2R, a trade that hit its stop is −1R, a trade you closed early at a small loss might be −0.4R.
Position size follows from the plan, not the other way round. If 1R is 1% of your account and the stop sits 2.5% below the entry, the position is 40% of the account — with or without leverage, the dollar risk is the same. The position size calculator does this arithmetic for crypto, and the guide to crypto trading risk management covers how to choose what 1R should be in the first place.
The weekly and monthly review
Recording is half the job. The review is where the journal becomes an edge, and it need not be long — half an hour at the weekend, an hour at month end. Look at the same fixed set of numbers every time:
- Win rate and average R per setup. Not overall — per setup. The overall number averages your best and worst strategies into one meaningless figure.
- Expectancy in R. The average result of one trade, which is the single number that says whether a strategy is worth trading at all.
- The costliest mistake tags. Sort the tags by the total R they cost you. Usually one or two of them explain most of the damage.
- Results by time of day and session. Many traders discover that a specific hour or session is consistently negative for them — late-night entries after a losing day are a classic.
- The P&L calendar. A month view shows streaks, the day after a big win (often sloppy) and the day after a big loss (often revenge) at a glance.
Expectancy is simple to compute: win rate × average win (in R) − loss rate × average loss (in R).
Monthly, add one more question: did anything change in the market or in you? A setup that worked in a trending month can lose in a ranging one. The journal cannot predict that, but it shows it before your account balance does.
Spreadsheet vs journal app
A spreadsheet is a perfectly good start, and many traders never need more. The question is what it costs you per trade and what it leaves out. The honest comparison:
| Excel / Google Sheets / Notion | Journal app with exchange sync | |
|---|---|---|
| Cost | Free (or already paid for) | Usually a free plan with limits, then a subscription |
| Effort per trade | Several minutes: copy prices, size, time and fees by hand | Under a minute: prices, size, fees and funding are already there; you add the plan and the tags |
| Fees and funding included | Only if you copy them every time; most people stop | Yes, from the exchange’s own records |
| Error risk | High: typos, forgotten trades, ‘I’ll add it later’ | Low for the facts; your notes are still yours to keep honest |
| Analytics | Whatever you build yourself: pivot tables take real work | Per-setup statistics, R-multiples, mistake tags and a calendar out of the box |
| What you give up | Nothing | A read-only API key for each exchange (no withdrawal or trading rights) |
PnLTrack is an example of the second column, built for crypto. It syncs trades about every minute from Binance, Bybit, OKX, MEXC, BingX, Hyperliquid and Propr via read-only API keys (Hyperliquid by wallet address), aggregates the individual fills into positions with entry, exit, PnL, fees and funding, and adds the parts an exchange does not have: a diary, plans with stop and target, R-multiples, mistake tags, statistics and a P&L calendar. The FREE plan is free forever with one exchange and one month of history; the full list is on the exchanges page and the plans are on the pricing page.
Common journaling mistakes
- Recording only the result. A journal of P&L numbers is a copy of your exchange history. Without the plan and the reason there is nothing to review.
- Skipping the losers. The trades you least want to write down are the ones with the most information in them. A journal with the losses left out is fiction.
- Too many fields. A 30-column template dies within a fortnight. Eight fields filled in every day beat thirty filled in once.
- Never reviewing. Data that is never looked at changes nothing. Put the weekly review in the calendar like any other appointment.
- Changing the rules after the fact. Moving the stop in the journal, relabelling a mistake as ‘market conditions’, quietly deleting a trade. The journal only works if it is allowed to disagree with you.
A 10-minute daily routine
- Before the session (2 min): write down which setups you will trade today, the size of 1R, and the maximum number of trades or the daily loss limit.
- Before each entry (1 min): record the setup, the stop, the target and the reason. If the exchange sync fills in the prices, you only write the plan.
- After each exit (1 min): note the result in R, tag any mistake and rate your state in one word.
- End of the day (3 min): scan the day’s trades once. Did execution match the plan? Attach a chart screenshot to anything worth revisiting.
- One sentence for tomorrow (1 min): the single thing to do differently — ‘no entries in the first 15 minutes’, ‘size down after two losses’.
- Weekly, not daily: leave the statistics alone until the review. Reading your win rate after every trade is noise, not feedback.
Frequently asked questions
- Is a trading journal worth it for crypto specifically?
- Yes — arguably more than elsewhere. Crypto markets run 24/7, perpetual futures add funding and fees that are easy to overlook, and leverage magnifies every undisciplined decision. A journal is the cheapest tool for catching all three.
- How long until a trading journal shows results?
- Honestly: a few dozen trades per setup before the statistics mean anything, and a few months before you trust them. The discipline effect — planning before entering — is immediate; the analytics take time.
- Do I have to record every trade?
- Yes. A journal with a selection of trades measures your selection, not your trading. Automatic import removes the excuse: the facts of every trade are recorded whether you felt like it or not.
- Can trades import automatically from Binance or Bybit?
- Yes. PnLTrack syncs from Binance, Bybit, OKX, MEXC, BingX, Hyperliquid and Propr via read-only API keys (Hyperliquid by wallet address), which can read your trade history but cannot trade or withdraw. Fills are aggregated into positions with fees and funding included.
- What is an R-multiple?
- R is the amount you planned to lose if your stop hit — the distance from entry to stop multiplied by the position size. An R-multiple expresses a trade’s result in those units: +2R means you made twice what you risked, −1R means the stop was hit.
Start a free trading journal
PnLTrack syncs your trades from Binance, Bybit, OKX, MEXC, BingX, Hyperliquid and Propr via read-only API keys (Hyperliquid by wallet address) and adds plans, R-multiples, mistake tags and statistics. The FREE plan is free forever: one exchange, one month of history.
Free plan with no time limit. All it needs is a read-only API key.