A trading journal is the difference between ten years of experience and one year of experience repeated ten times. Without a record, you remember your wins, quietly forget your losses, and “learn” a story that has nothing to do with your actual results. With one, every trade becomes a data point about your reading, which is the only thing you can genuinely improve. This guide covers how to keep a journal built specifically for order flow, where the goal is to grade the flow you read, not just the money you made.
Why most journals are useless
The typical trading journal is a P&L spreadsheet: entry, exit, profit, loss. That tells you that you lost, never why. It can’t distinguish a good process that lost from a bad process that won, and those two are the whole game. A good read that lost is a trade you should take again. A bad read that won is a habit that will eventually bankrupt you. A columns-of-money journal treats them the same and teaches you the wrong lesson from both.
For order flow, the useless journal is even worse, because the entire edge lives in the read, the aggression you saw, the absorption you judged, the context you weighed. If none of that is recorded, you have thrown away the only information worth reviewing.
What to actually log: the read, not just the result
For every trade, record what you saw and decided, in enough detail that you can grade the decision later independently of how it turned out. A workable set of fields:
| Field | What to capture |
|---|---|
| Instrument & time | ES, 09:48, so you can pull the replay later |
| Level / context | “Prior day VAL 5,378” — the context that made the spot matter |
| Flow read | “Heavy bid volume absorbed, cumulative delta diverging” |
| Pattern | Absorption reversal / trapped traders / delta divergence |
| Decision & reason | “Long 5,379, stop 5,374 below absorption” |
| Result | +8 ticks, stopped, scratched |
| Read quality (1–5) | Was the read right, regardless of the money? |
| Note to self | “Entered two ticks early, before price lifted off” |
The two fields that make this an order flow journal are flow read and read quality. That read-quality grade, scored on its own merits and separately from P&L, is what lets you find the pattern in your mistakes.
Anchor each entry to the actual footprint and delta you were looking at. A screenshot of the setup at entry is worth a paragraph of description, so grab one.
The four-box review that finds your real edge
Once you have a couple of weeks logged, sort every trade into one of four boxes using two axes: was the read good, and did the trade win.
- Good read, winner. Your bread and butter. Do more of this. Note what the setup had in common.
- Good read, loser. The market’s variance, not your error. These are correct trades, take them again. Beginners wrongly “fix” these and break their edge.
- Bad read, loser. The pile to attack. What did you misread? Wrong context, delta traded in isolation, an imbalance with no level behind it?
- Bad read, winner. The most dangerous box, because the money rewards a bad habit. Flag these hard. They feel great and teach the worst lessons.
This split is the whole reason to journal the read. A pure P&L journal can only see winners and losers, so it praises the bad-read winners and punishes the good-read losers, exactly backwards. Grading the read is what lets you reward process instead of luck.
How to run the weekly review
Journaling without reviewing is just data entry. Set a fixed weekly slot, 30 to 45 minutes, and do this:
- Read every entry back with the chart open. Pull up the replay of the moment and judge the read fresh, now that you know the outcome. Was it genuinely good, or does hindsight just make it look that way?
- Tally the four boxes. Where is the leak? Usually one type of bad read repeats.
- Name one thing to fix next week. One. “Only take delta divergences at marked levels, skip mid-range ones.” A single, specific, testable change.
- Confirm your good-read losers were correct. Resist the urge to tinker with trades that were right and just didn’t pay. Protecting your correct process is as important as fixing the broken one.
The output of every review is one behavior change, tested the following week and reviewed again. That loop is how the journal actually makes you better instead of just documenting your decline.
The mistakes journals expose fastest
A read-based journal surfaces the classic order flow errors quickly, because they show up as repeating bad-read boxes. You will typically catch yourself:
- Trading delta as a signal instead of reading it against price’s response.
- Acting on flow in the middle of nowhere, with no volume profile level behind it.
- Overtrading, visible as a cluster of tiny scratches and bad-read entries in a short window.
These and the rest of the usual suspects are catalogued in common order flow mistakes; the journal is what turns that list from abstract warnings into “I did this four times last Tuesday.”
Tools: keep it simple
You don’t need software to start. A spreadsheet with the columns above and a folder of entry screenshots does the job for months. Dedicated journaling apps add automatic stats and P&L import, which is nice later, but the automation can tempt you back toward P&L-only thinking. The read fields are manual by nature, nobody can log “I saw absorption at VAL” for you, so whatever tool you use, the discipline is the same: write the read, grade the read, review the read.
If you are early on the path, the journal isn’t an optional extra you add once you are good. It is the mechanism that makes you good, which is why the learning roadmap places it at the center of the practice phase, right alongside replay. Practice generates the reps; the journal extracts the lesson.
Frequently Asked Questions
What should I log in a trading journal beyond profit and loss?
The read itself: the level and context, the flow you saw (footprint pattern, delta behavior), your decision and its reasoning, and a 1–5 grade of whether the read was good regardless of the money. Those fields let you separate correct trades that lost from lucky trades that won, which pure P&L can never do.
How often should I review my trading journal?
Once a week, in a fixed 30-to-45-minute slot, with the charts open so you can re-judge each read. The goal of every review is to name one specific, testable behavior change for the following week. Daily is usually too noisy to see patterns; monthly is too slow to correct them.
Why grade the read separately from the result?
Because a good read can lose and a bad read can win, and treating them by outcome teaches the wrong lesson from both. Grading the read lets you keep taking correct trades that happened to lose and stop taking bad trades that happened to win, which is how you protect a real edge.
Do I need a special app to keep a trading journal?
No. A spreadsheet with the read fields plus a folder of entry screenshots works for months and forces you to write the read manually, which is the part that matters. Dedicated apps add automatic stats later, but the manual read-and-grade discipline is what actually improves your trading.