Most traders who keep a journal are recording a ledger. Entry, exit, profit or loss. Their broker already has all of that.
A ledger tells you what happened. It cannot tell you why, and it cannot change anything, which is why most journals are abandoned after five weeks.
A journal changes behaviour when it records the two things your broker cannot see: what you decided, and what you did not do.
The two columns nobody logs
Start here, because these two do most of the work.
1. The trades you did not take
Every setup that met your criteria and that you skipped. Time, setup, reason for skipping, and what it went on to do.
This is the single most valuable column in a trading journal and almost nobody keeps it.
Your broker statement contains only trades you took, so if hesitation is your problem, it is completely invisible in your records. You can spend a year “fixing your trading” while the actual leak, skipped setups, never appears in any number you look at.
Two weeks of this produces an honest figure: what hesitation costs you per month, in R. For many traders it exceeds what their losing trades cost.
2. Where the trade went after you exited
One number: the maximum favourable move after your exit.
This makes the cost of cutting winners visible. Normally that cost is invisible by design, because you never see the 3R you did not get. You just see a green trade and feel fine.
Uncomfortable reading, and it is the only feedback loop that makes early exits real.
The columns that matter
Keep it short enough that you will actually fill it in. Ten fields, most of them a single character.
| Field | What goes in it |
|---|---|
| Date and time | Entry time specifically. Patterns cluster by hour. |
| Instrument | Nifty, Bank Nifty, stock, strike. |
| Setup | Which of your named setups. If none, write “none”. |
| All criteria met? | Y or N. The compliance column. |
| Risk in rupees | Planned risk at entry. |
| Stop as live order? | Y or N. |
| Planned exit | Target and stop, written before entry. |
| Actual exit | Where you actually got out. |
| Result in R | Not rupees. R makes trades comparable across sizes. |
| Max favourable after exit | In R. The cost of early exits. |
Plus one free-text line, written before entry: setup, stop, why now.
That pre-entry line is doing double duty. It creates the record, and writing it is itself a filter, because it is nearly impossible to complete on an impulsive trade.
Record R, not rupees
Worth being firm about this.
Log every result as a multiple of the amount you risked. Risk ₹2,000 and make ₹6,000, that is +3R. Risk ₹2,000 and lose it, that is -1R.
Rupees are not comparable across trades because your size varies. R is. It lets you compare a Nifty trade from March to a Bank Nifty trade from August, and it lets you compute the only two numbers that actually determine profitability: average R won and average R lost.
If your average winner is under 1.5 times your average loser, that is your biggest leak, and nothing else in your statistics matters until it is fixed.
Logging state, briefly
Three quick fields per day, not per trade:
- Hours slept
- Money pressure outside the market, Y or N
- Carrying anything from yesterday, Y or N
After a month, cross-reference these against your worst days. Most traders find their bad sessions cluster against specific state answers, usually sleep.
This matters because the warning signs typically precede the losing day rather than following it. A journal that only records trades cannot show you that. One that records state can, and it turns “I had a bad day” into a condition you can check tomorrow morning.
The weekly review
The journal does nothing without this. Thirty minutes, same time each week, market closed.
Five numbers:
- Compliance rate. Trades meeting all criteria, divided by trades taken.
- Average R won and average R lost.
- Skipped setups and what they would have produced.
- Total left on the table from early exits.
- Rule breaks by type and by time of day.
Then one question: what single change would have improved this week most?
One change. Implement it next week. Traders who identify six improvements implement none.
Study your best trades too
Most journalling advice focuses on losses. That is half the method, and the smaller half.
Brett Steenbarger’s work makes the case for reviewing your best trades with the same rigour. Once a month, pull your five largest winners and look for what they share. Time of day, setup type, market condition, how long you held, what you were doing that morning.
The pattern is usually clearer than the pattern in your losses, and it is far more actionable, because you can do more of it. Most traders discover their best trades cluster in a narrow window and come from one or two setups. That finding alone often justifies cutting half of what they trade.
Making it survive
Journals fail from being too elaborate. Some things that help:
- A spreadsheet, not an app. Ten columns you control. Apps import broker data automatically, which sounds convenient and quietly removes the decision-logging that makes the journal work.
- Fill it at the time, not at the end of the day. The pre-entry line has to be written before entry or it becomes fiction. Memory reconstructs reasons that were not there.
- Under two minutes per trade. If it takes longer you will stop.
- Screenshot only the ones that surprised you. Screenshotting everything is how people quit in week three.
- Review weekly even if you traded twice. The habit matters more than the sample.
Frequently asked questions
How long before a journal shows me something useful?
Two weeks for time-of-day and rule-break patterns. About 30 trades before the R statistics mean anything. Three months before you can see whether an intervention worked.
Should I journal emotions?
Briefly and specifically. “Felt rushed, entered before the candle closed” is useful because it links a state to an action. “Felt anxious” is not, because there is nothing to do with it. Log the behaviour the emotion produced, not the emotion alone.
Is a journalling app better than a spreadsheet?
Apps are better at statistics and worse at the part that matters. They auto-import fills, which means the two most valuable columns, skipped setups and pre-entry reasoning, are absent. A simple spreadsheet you fill in manually beats an app you sync.
I journal but nothing changes. Why?
Almost always because there is no weekly review, or because the journal records outcomes rather than decisions. Recording that you lost ₹4,000 changes nothing. Recording that you entered without your third condition met, at 2:15pm, for the fourth time this month, changes something.
Do I need to journal every trade?
Yes, including the small ones and the ones you regret. Selective journalling produces a flattering record and useless data. The trades you least want to log are the ones the journal exists for.
The bottom line
Your broker records what happened. A journal is for what you decided, what you skipped, and what the trade did after you left.
Ten columns, filled in at the time, reviewed weekly with five numbers and one change.
Journal trade ka record nahi hai. Decision ka record hai.
Related reading:
- The ACE Framework: Aware, Control, Execute Explained
- How to Build Habits That Actually Stick as a Trader
- The Evening Review That Fixes Tomorrow’s Trades
Want a review of your own numbers? I run a free live session twice a week for traders with two or more years of experience who know the setups but still cannot execute under pressure. Register for the next free session here.
I am Samir Dash, founder of Mindful Trading Hub. I work with experienced traders on live market decision-making. More about my story here.
