Ask a trader why they lost money last quarter and you will get a story. The story will be coherent, plausible, and largely constructed after the fact.
Memory does not store trading experience accurately. It keeps the dramatic ones — the big win, the disaster — and quietly discards the ordinary majority that actually determine your results.
A journal is the only fix. It is also the single highest-return habit available to a trader, and almost nobody keeps one.
What memory does to your record
It keeps the extremes. You remember the trade that ran 400 points. You do not remember the eleven that stopped out for small losses in between.
It rewrites the reason. After a win you recall the analysis that justified it. The doubt at entry, and the fact that you nearly did not take it, are gone.
It converts luck into skill. A trade that worked for reasons unrelated to your thesis gets filed as evidence the thesis works.
The consequence is that a trader's beliefs about their own edge drift steadily away from their actual results — and drift in a flattering direction.
What to log
The temptation is to record everything, which produces a journal too tedious to maintain. These fields do most of the work:
| Field | Why |
|---|---|
| Date and instrument | Basic |
| Setup name | Lets you group trades and find which actually work |
| Direction and structure | Long call, bull spread, futures |
| Entry, stop, target | The plan, as it was |
| Size, and the risk in rupees | Reveals sizing drift |
| Reason for entry — one line | The most valuable field |
| Exit price and reason | Did you follow the plan or improvise? |
| Was the plan followed? yes/no | The only true measure of process |
| Result | Last, deliberately |
Two of those matter far more than the rest.
Reason for entry, written before the trade. This is the field memory corrupts most reliably. Written down, it cannot be revised later.
Was the plan followed? A simple yes/no that separates process from outcome. Over fifty trades this single column tells you whether your problem is the strategy or your execution of it — and those need completely different fixes.
What not to log
Feelings, in detail. "Felt nervous" adds little. A single tilt flag is enough.
Market commentary. Your journal is about your decisions, not your opinions about the market.
Every price tick. Your broker has that.
A journal you actually maintain beats a comprehensive one you abandon in week three. Keep it to a spreadsheet row per trade, under ninety seconds to fill in.
The review, which is the entire point
Logging without reviewing is just filing. The review is where the return comes from.
Weekly, ten minutes. Read the entries. Count how many followed the plan. That percentage is your real process score.
Monthly, thirty minutes. Group by setup name. For each: how many trades, win rate, average win, average loss. Now you know which of your setups actually work rather than which you enjoy.
Quarterly. Look for drift. Is average size creeping up? Are you trading more often than three months ago? Has the losing side of one setup quietly become most of your damage?
What the data usually reveals
Three findings come up repeatedly, and all three are invisible without records.
One setup carries the account. Most traders have several setups and one that works. The others are roughly break-even after costs, and they consume most of the screen time.
Losses cluster. Not randomly — around specific conditions. Particular times of day, particular market regimes, the day after a big loss. Clusters are actionable in a way that scattered losses are not.
Adherence predicts results better than analysis does. The correlation between "followed the plan" and profitability is typically stronger than between any analytical refinement and profitability.
Making it stick
Log at entry, not at the end of the day. The reason field is worthless if written after you know the outcome.
Same place, every time. A spreadsheet is fine. Consistency beats sophistication.
Put the review in your calendar. Ten minutes on Friday. Unscheduled reviews do not happen.
Track the streak. Days logged in a row. It is a small trick and it works.
Check yourself
0 of 4 answered1.Which journal field is most corrupted by memory, and therefore most valuable to write down?
2.What does a 'was the plan followed?' yes/no column tell you that P&L cannot?
3.After ten trades your new setup shows a 30% win rate. What should you conclude?
4.What do trading records most commonly reveal about a trader's setups?
What to take away
- Memory keeps the extremes, rewrites the reason, and converts luck into skill.
- The two fields that matter: reason for entry (written first) and was the plan followed.
- Result goes last — knowing it bends everything else.
- Logging without reviewing is just filing.
- Expect to find: one setup carries the account, losses cluster, adherence beats analysis.
- Fifty trades before concluding anything.