Most trading journals are diaries: a log of entries, exits, and profit or loss that the trader never opens again. That is not a journal, it is a receipt. Learning how to keep a trading journal that actually improves your trading means logging the two things a receipt never captures, why you took each trade and how well you followed your plan, and then doing the one thing that turns the log into insight: reviewing it.
This guide gives you the futures version, not the generic forex one. You will get the exact fields to log per trade (measured in R and tagged to your order-flow read), a worked ES example, and the weekly review that reveals which setups and sessions are your edge and which are quietly leaking money. Everything is illustrative, not advice. A journal reveals problems, it does not fix them; futures carry a substantial risk of loss, and most day traders lose money.
What is a trading journal?
A trading journal is a per-trade record of both what you did and why: the objective facts (instrument, entry, stop, exit), the result in R, the setup and order-flow read behind it, and whether you followed your plan. You review it on a schedule to separate what is working from what is leaking. It is not a P/L diary.
The distinction that matters is that the objective facts tell you what happened, but only the process fields tell you why, and why is the only thing you can improve. A log of just entry, exit, and P/L is a bank statement. The setup tag, the order-flow read, and the honest process grade are where the edge is found.
What should you track in a futures trading journal?
Split the fields into two buckets. The first is objective and should auto-import from your platform’s statistics export or a broker CSV, so you never hand-type it. The second is the process, which you hand-enter after the close, and which is the entire point of the exercise.
| Field | Bucket | Why it matters at review |
|---|---|---|
| Date, time, session (open / lunch / close) | Auto | Reveals which session is your edge and which is chop |
| Instrument, direction, size | Auto | Lets you compare ES, NQ, and Gold side by side |
| Entry, initial stop, exit, exit reason | Auto | The raw trade; the stop sets 1R |
| Result in R (P/L ÷ 1R) | Process | The linchpin: only R makes trades comparable and expectancy computable |
| Setup + order-flow tag (absorption, delta divergence, stacked imbalance) | Process | Lets you rank setups by expectancy; the field the whole SERP omits |
| In-plan? · stop before entry? · process grade 1-5 | Process | Turns discipline into a number you can review |
| Mistake tag (chased, moved stop, oversized, revenge) | Process | Totals the R that each bad habit costs you |
Two of those fields are what separate a futures order-flow journal from the generic templates that rank today: the result in R rather than dollars, and the order-flow tag that records what the footprint actually showed at your decision point. Here is one ES trade with every field filled.
Grading each trade 1 to 5 on how well you followed your plan is what lets a bad-process win show its true colours. A chased trade in the lunch chop that happened to close green is not a good trade, it is a grade-2 win that will cost you later; a textbook setup that hit its stop is a grade-5 loss and still a good trade. That is the process-over-outcome discipline, and the journal is where you record the score.
The weekly review: find your edge and your leaks
The log is worthless until you review it, and the review is not admiring your winners. It is arithmetic. Because every result is in R, your expectancy is simply the average of the R column, and the whole edge-versus-leak picture comes from grouping the same trades two ways: by setup, and by session. Take an illustrative month of 40 trades, 45% winners at an average of +1.8R with losses at 1.0R, an expectancy of +0.26R per trade.
| Setup | Trades | Win% | Expectancy | Net | Read |
|---|---|---|---|---|---|
| Sweep + absorption reversal | 16 | 56% | +0.69R | +11.0R | The edge |
| Delta-divergence continuation | 12 | 50% | +0.30R | +3.6R | Fine |
| Chased breakout, no confirmation | 12 | 25% | −0.35R | −4.2R | The leak |
That is the whole payoff of a journal, and it is invisible in a P/L total. The chased-breakout setup lost 4.2R over the month; drop it and the same trader nets +14.6R over 28 trades, a jump from +0.26R to +0.52R per trade while trading less. Group the identical month by session instead and a second leak appears: the 9:30 to 10:30 open carried the account at +0.94R per trade, while the lunch window bled at −0.50R. The action writes itself: do more of what has positive expectancy, cut or fix what does not. When you should run this review, daily log and weekly read, belongs to the trading plan and its cadence; here the point is what to compute.
Ten trades are noise. A single setup’s expectancy needs roughly 20 to 30 instances before it means much, and ideally more. A positive-expectancy setup still strings four or five losers in a row as a matter of ordinary variance, so do not kill a setup on three losses or fall in love with one on three wins. Change one variable, then re-measure over a fresh sample.
How to journal during the session without disrupting your trading
Do not write essays between trades; you will miss the next one. Use two layers:
- In-session: capture only seconds of data, a hotkey chart screenshot of the footprint at entry and exit, the setup name, and any rule you broke. For an order-flow trader that screenshot is the single most valuable field, because the read is the thesis.
- After the close: while it is fresh, do the full write-up, the R result, the process grade, the mistake tags, and a one-line lesson.
Let your platform do the boring half: ATAS ships a statistics and journal module that logs fills and computes win rate, average win and loss, and drawdowns, and exports to Excel, as does a broker CSV, so the objective bucket fills itself and you only hand-enter the process fields.
Spreadsheet or trading journal software?
Start with a spreadsheet. It is free, and defining your own columns is itself part of the learning. A handful of formulas does everything: win rate, average win and loss, expectancy as the average of the R column, profit factor, and a pivot table that gives you the by-setup and by-session expectancy automatically.
| Spreadsheet | Journal software | |
|---|---|---|
| Cost | Free | Subscription |
| Data entry | Export plus hand-entry | Auto-imports fills |
| Stats and breakdowns | Your own formulas and pivots | Computed and charted for you |
| Custom fields | Fully yours; forces you to define them | Mostly presets |
| Best for | Starting out, lower volume | High volume, once manual stats are the bottleneck |
Tools like Tradervue, TradeZella, and TradesViz auto-import and chart the breakdowns, which is genuinely worth it at volume. But be honest about what you are buying: the software does not create the value, the weekly review does, and no tool can grade whether you followed your plan or tag the order-flow read for you. Start with a sheet, and upgrade only when computing the stats by hand becomes the actual bottleneck, never to avoid the work.
Do trading journals actually improve performance?
Only if you review the journal and act on what it shows. The log itself changes nothing; it is a measurement instrument that makes invisible leaks visible and turns “I need more discipline” into an adherence number you can move. Be honest about the evidence: there is no trading-specific study proving a journal lifts returns. What exists is the general research on structured reflection, where taking time to review what you did measurably improved later performance in other fields, plus the plain logic that you cannot fix a leak you cannot see. A losing method logged flawlessly still loses, garbage in is garbage out, and a journal you never open is just a diary. It reveals; you still have to act.
The order-flow reads you tag each trade with, the absorption, the delta divergence, the stacked imbalance, are exactly what the Order Flow Suite is built to surface on ES, NQ, and Gold. The tools make the read visible so you can record it; the logging, the honesty, and the weekly review stay your work, and none of it removes the risk.
Frequently asked questions
What should I track in a trading journal?+
Two buckets. The objective facts your platform can export: date, session, instrument, direction, size, entry, stop, exit, and P/L. And the process fields you hand-enter, where the value is: the setup, the order-flow read behind it, the result in R, whether it followed your plan, a process grade, and a mistake tag. Facts plus process, not just P/L.
Is a spreadsheet enough, or do I need trading journal software?+
A free spreadsheet is enough to start, and defining your own fields is part of the learning. Software auto-imports fills and computes the stats and breakdowns, which is worth it once volume makes manual entry the bottleneck. But the software does not create the value; the weekly review does. Start with a sheet, upgrade only when you must.
How often should I review my trading journal?+
Log every trade the day you take it, review the week’s trades on the weekend, and revise your actual plan only on a real sample of trades, never after a single loss. The daily log captures the data, the weekly review turns it into insight, and monthly changes keep you from over-tuning on noise.
Do trading journals actually improve trading performance?+
Only if you review it and act on what it shows. The log itself changes nothing; it is a measurement instrument that makes your leaks visible. There is no trading-specific study proving it lifts returns, but structured reflection has improved performance in other fields, and you cannot fix a leak you cannot see. A losing method logged perfectly still loses.
How do I journal trades during the session without disrupting my day trading?+
Use two layers. In-session, capture only seconds of data: a chart screenshot at entry and exit, the setup name, and any rule you broke, then get back to trading. After the close, while it is fresh, do the full write-up: the R result, the process grade, and the notes. Let your platform export the objective facts automatically.
Where to go next
The journal is the feedback loop for the rest of the cluster. The R and expectancy it measures come from the risk management pillar, the rules whose adherence it grades come from the trading plan, and the grade-the-process discipline behind it is the subject of trading psychology. The order-flow reads you tag each trade with run through the order flow curriculum, from delta divergence to absorption. Log the trade, then read the log.