A trading journal is a written record of every trade — the setup, the trigger, the risk, the outcome and whether you followed your own rules. Its real job is not bookkeeping. It is catching the patterns your memory edits out, because traders remember their reasoning far better than their behaviour.
Most journals get abandoned in three weeks, and the reason is that they were built to record profit and loss. Your broker already does that, perfectly, for free. A journal that duplicates the statement is genuinely a waste of time. A journal earns its place only when it captures the things the statement cannot see: what you were waiting for, what actually made you click, and whether the trade you took was the trade you had planned.
What a journal is actually for
Human memory of one's own decisions is reconstructive rather than archival. Ask a trader in March why they took a trade in January and they will give you a coherent, confident answer assembled largely from knowing how it turned out. Winners get remembered as skill and losers as bad luck, and neither the sample nor the story is reliable.
A journal replaces recollection with evidence. Its output is not a number — it is a sentence like "I take my best setups at 09:45 and my worst at 11:30, and I break the entry rule roughly one trade in four." That sentence is invisible from inside any single session and obvious across thirty logged ones.
The bias a journal is built to catch
There is one pattern worth naming up front, because almost everyone has it and nobody notices it without a record. It is the disposition effect: the tendency to realise gains quickly while holding losses.
Studying trading records from 10,000 accounts at a large discount brokerage between 1987 and 1993, Terrance Odean measured how often investors sold positions that were up versus positions that were down. Outside December, they realised about 14.8% of their available gains but only 9.8% of their available losses — a winning position was roughly 50% more likely to be sold than a losing one. In December the pattern reversed, when tax considerations gave people a reason to take losses (Odean, "Are Investors Reluctant to Realize Their Losses?", Journal of Finance, 1998).
That behaviour destroys risk-to-reward arithmetic quietly. Winners cut at 0.6R and losers held to 1.4R will produce a losing account at a 60% win rate, and the trader will spend months trying to improve their entries — the one thing that was already working. You cannot see it in your P&L. You can see it immediately in a journal that records planned R alongside realised R, which is the point made in the risk-to-reward ratio.
The twelve fields worth logging
Keep it small enough to complete in ninety seconds, or it will not survive a busy week.
| Field | Why it earns its space |
|---|---|
| Date & time of entry | Surfaces performance by hour and by session |
| Instrument | Shows which markets you are actually good at |
| Setup name | Lets you group trades — the single most useful cut |
| Trigger | Separates "the level was there" from "the condition was met" |
| Entry, stop, target | The plan, recorded before the outcome is known |
| Risk in $ and % | Exposes size creep after wins and after losses |
| Planned R | What the trade was supposed to be worth |
| Realised R | Compared with planned R, this is your disposition-effect detector |
| Exit reason | Target, stop, trail, or "I got nervous" — be honest |
| Rule compliance | Yes / no. The most valuable field in the table |
| Chart screenshot at entry | Stops you from re-imagining what the chart looked like |
| One line of note | State, distraction, anything unusual. One line, not an essay |
The compliance field is the one people leave out, and it is the reason most journals produce nothing. Without it you can only analyse outcomes; with it you can analyse behaviour, which is the only variable you control.
The metrics that tell you something
Once you have thirty logged trades, a handful of cuts do almost all the work:
- Expectancy in R — average R per trade across the sample. This single number tells you whether the system makes money over repetitions, independent of size.
- Average winner vs average loser, in R. If winners average less than 1R while losers average more, you have found the disposition effect in your own data.
- Compliance rate. The percentage of trades taken exactly to plan. Below about 80% there is no point analysing anything else — you are measuring a system you are not running.
- Expectancy by setup. Most traders find one setup carries the account and one quietly bleeds it. Cutting the second is usually the fastest available improvement.
- Expectancy by hour and by weekday. Frequently the sharpest edge in the whole journal, and it usually argues for trading less — see the best time of day to trade.
Three numbers mislead beginners more than they help. Win rate alone says nothing without average R beside it. Daily P&L is noise at any realistic sample size. And your best trade is a story, not evidence — it is usually the trade that taught you a habit you will pay for later.
How to keep one you'll actually maintain
- Log within minutes of the close, not at the end of the week. Accuracy on the "why" field decays within hours, and the why is the whole asset.
- Start in a spreadsheet. Twelve columns. Dedicated journaling software is fine later; it is not the thing standing between you and a review habit now.
- Screenshot at entry, not at exit. A chart captured after the fact shows you what happened, which you already know. Captured at entry, it shows you what you were actually looking at.
- Review on a schedule. Two minutes at the end of each session; a proper analysis every 20 to 30 trades. Reviewing daily invites rule changes based on noise, exactly the failure described in the ten mistakes that blow up new accounts.
- Write one change, not five. Each review should produce a single adjustment for the next block, so you can attribute the difference to something.
A journal and a plan are two halves of the same instrument: the plan states what you will do, the journal records what you did, and the gap between them is where nearly all improvement lives. If you have not written the first half yet, start with how to build a trading plan you'll actually follow.
Frequently Asked Questions
What should you record in a trading journal?
The setup and trigger, the entry, stop and target prices, the risk in both currency and percentage terms, the result expressed in R, a screenshot of the chart at entry, and one honest line on whether you followed your own rules. The compliance field is the one most people omit and the one that produces the most improvement.
Does keeping a trading journal actually improve results?
A journal does not improve results on its own; reviewing it does. Its value is that it records behaviour rather than reasoning, which is what memory distorts. Grouping trades by setup, by hour and by rule compliance surfaces patterns that are invisible from inside a single session.
What is the disposition effect in trading?
It is the documented tendency to sell winning positions too early and hold losing ones too long. Studying 10,000 accounts at a discount brokerage from 1987 to 1993, Odean found investors realised about 14.8% of their available gains but only about 9.8% of their available losses — roughly 50% more likely to take a profit than a loss. A journal is how you find out whether you do it.
How often should you review your trading journal?
Log within minutes of the trade closing, review briefly at the end of each session, and do the real analysis every 20 to 30 trades. Single sessions contain too much noise to draw conclusions from, while a block of 20 to 30 trades is large enough for patterns by setup and by hour to appear.
Bottom line
A trading journal is not a diary and it is not a second copy of your broker statement. It is the only instrument that measures your behaviour rather than your reasoning, and behaviour is where the money goes. Odean's 10,000 accounts realised gains at 14.8% and losses at 9.8% — a bias none of those investors would have reported having, and one that shows up in a journal within thirty trades. Log twelve fields, record compliance as ruthlessly as you record profit, review every twenty to thirty trades, and change one thing at a time. Where this sits in the wider beginner path is laid out in how to start day trading, and the questions we get asked most often are answered on our FAQ page.