Short answer
A useful trading journal records what you planned before a trade, what actually happened, and whether you followed your rules—then turns those records into a scheduled review. Start with a small set of consistent fields: setup, entry, initial stop, size, initial risk, actual exit, costs, net R-multiple, and rule compliance. The journal is not a scorecard for predicting the next trade; it is evidence for separating strategy, execution, and behavior.
A trading journal is not just a broker statement
A broker statement tells you what was filled and what the account gained or lost. It normally cannot tell you whether the trade matched your setup, whether the planned stop changed, or whether a profitable result came from breaking a rule. That is the job of the journal.
Keep three documents conceptually separate. Your trading plan defines what should happen. The trade log preserves objective transaction data. The journal compares the two and records the reason for any difference. Mixing all three into a long paragraph makes review slower and inconsistent.
The Corporate Finance Institute's trading guide describes a journal as a record of the entry, the initial reason, stop and target, subsequent market action, the trader's response, and the final result. Its useful point is not that every trader needs the same format; it is that the record must preserve both the transaction and the decision. See the record-keeping section of the CFI guide.
Build the journal backwards from questions
Every field should support a question you intend to review. Otherwise it becomes friction. Before building a spreadsheet or paying for software, write five to seven questions that a month of records should help answer.
Setup
Which written setup produced the trade, and which version of its rules applied?
Risk
Was the initial risk within the limit, and did actual loss exceed it?
Execution
How far did actual fills differ from the planned prices?
Costs
Did spread, commission, financing, or slippage materially change the result?
Compliance
Which rules were followed, broken, or impossible to verify?
Context
Do results differ by session, direction, instrument, or a defined market regime?
Use fixed tags for repeated categories such as setup, session, and mistake type. Reserve a short note for information that cannot be expressed as a tag. If "trend pullback," "pull-back," and "PB" describe the same setup, a later filter will treat them as three different groups.
The minimum useful trading journal fields
Start with the smallest version you can complete accurately. Add optional fields only after a review reveals a question the current data cannot answer.
| Group | Fields | Why they matter |
|---|---|---|
| Identity | Trade ID, account, date/time, instrument, direction | Reconcile records and segment results |
| Plan | Strategy version, setup, timeframe, planned entry, stop, target | Compare the trade with written rules |
| Risk | Position size, loss per unit, initial monetary risk, open-risk total | Normalize size and verify limits |
| Execution | Actual entry and exit fills, partials, stop changes | Measure deviation from the plan |
| Costs | Spread, commission, fees, financing, estimated slippage | Calculate a realistic net result |
| Outcome | Net P&L, net R, exit reason, maximum loss if known | Compare differently sized trades |
| Process | Rule-compliant: yes/no, violation tag, short lesson | Separate process from outcome |
| Evidence | Entry/exit screenshots or chart links | Preserve the visible context without relying on memory |
Start with a copyable structure
The free Strategy Archive trading journal template provides the full field list. Remove anything you will not review; do not remove the initial stop, size, risk, costs, or compliance fields if you want to calculate the metrics below.
Record each fact at the moment it can still be trusted
- 01
Before entry
Save the setup, planned entry, invalidation, target or exit rule, size, initial risk, and a screenshot. A plan reconstructed after the result is vulnerable to hindsight.
- 02
After the fill
Replace assumptions with the actual fill, time, size, spread, and order route. If the fill changes the risk, calculate it again before adding exposure.
- 03
During management
Record only decisions that alter the original plan: stop movement, partial exit, scale-in, cancellation, or a rule override. Add a time and a reason.
- 04
After exit
Import or enter the final fill, costs, net result, net R, exit reason, and an exit screenshot. Grade compliance before writing a lesson.
- 05
After the session
Reconcile the journal with the broker statement, flag missing trades, and record valid signals you deliberately skipped if that is one of your review questions.
- 06
At the scheduled review
Aggregate results, compare like with like, identify one issue worth testing, and record the change as a new strategy version rather than rewriting history.
Worked example: one result, three different conclusions
Consider a hypothetical stock trade entered at $100 with an initial stop at $99 and 50 shares. The initial planned risk is ($100 − $99) × 50 = $50. The position exits at $101.80. Gross profit is $90; after $4 of total costs, net profit is $86 and net R is $86 ÷ $50 = +1.72R. This is an arithmetic example, not a suggested trade or risk amount.
| Record | What happened | Review conclusion |
|---|---|---|
| A | Setup and every management rule were followed | Compliant +1.72R trade |
| B | Entry was outside the setup, but price rose | Profitable rule violation |
| C | Setup was valid, but the position was twice the allowed size | Valid idea, invalid risk execution |
P&L alone labels all three as winners. A journal preserves the difference. It also keeps planned risk fixed when calculating R: using the final stop after it has been moved would rewrite the denominator and make comparison unreliable. If you need help translating stop distance into size, use the Trading Calculator before placing the trade.
Metrics worth reviewing—and what they cannot prove
Net R-multiple
Net R = net result after costs ÷ initial planned risk. R helps compare trades with different monetary size, but it is only consistent if the definition of initial risk remains consistent.
Historical expectancy
Mean net R = sum of net R results ÷ number of trades. The equivalent grouped form is (win rate × average winning R) − (loss rate × average losing R), with break-even trades included consistently. It describes the sample; it does not promise the next result.
Rule-adherence rate
Compliant trades ÷ all reviewed trades. Publish the checklist used for the grade. A self-awarded score without binary rules can drift when results are good or bad.
Mistake cost
Compare the actual result with the result produced by the written rule only when the counterfactual price and execution are available. Label estimates; do not convert a chart's best possible exit into money you claim was “lost.”
Van Tharp's lesson defines R from the initial risk and describes expectancy as the average R-multiple produced by a set of trades. It also demonstrates why win rate and expectancy are different questions. Read A Short Lesson on R and R-Multiples.
Segment cautiously. Looking at setup, session, instrument, direction, weekday, emotion, volatility, and ten other tags creates many opportunities to find a pattern by chance. Start with a question written before the analysis, compare rules and costs on like-for-like trades, and test any proposed change on new data. Our backtesting versus forward testing guide explains why a result that fits old data may not survive live conditions.
A weekly review that ends with one decision
Logging is collection; review is the feedback loop. Use the same sequence each week so the review is not driven by the most recent win or loss.
01
Reconcile
Match every journal row to broker fills. Resolve missing trades, duplicated partial fills, fees, and timezone differences before interpreting performance.
02
Freeze the period
Do not rewrite old setup labels or stops to fit a new theory. Corrections need an audit note.
03
Separate process and outcome
Count compliant wins, compliant losses, non-compliant wins, and non-compliant losses. A losing compliant trade is not automatically a mistake.
04
Review one main question
For example: did late entries increase slippage, or did one setup generate most rule violations? Avoid searching every tag for a flattering result.
05
Inspect the evidence
Open a small set of representative screenshots: the cleanest execution, the clearest violation, and any outlier that materially affected the week.
06
Choose one action
Keep the rules, clarify one ambiguous rule, test one hypothesis, or pause a setup. State the evidence, effective date, and what would reverse the decision.
A defensible weekly conclusion
“This week contained 12 trades. Ten followed the checklist and two were early entries. The two violations cost -1.4R net, but the sample is too small to change the setup. For the next 20 eligible signals, the entry alert will be tied to candle close; all other rules remain unchanged.”
This conclusion reports the evidence, its limitation, one operational change, and the next observation window. “I need more discipline” does none of those things.
Common trading journal mistakes
Recording only losses
The journal needs compliant wins and losses to show what normal execution looks like.
Writing essays for every trade
Long free text is hard to compare. Use stable tags plus one concise decision note.
Grading the outcome
A profitable rule violation remains a violation; a planned loss can still be compliant.
Ignoring costs
Gross results can conceal spread, commission, financing, and slippage that remove a small edge.
Changing several rules together
If entry, stop, and target all change, the next sample cannot identify what mattered.
Treating a small segment as proof
A striking weekday or emotion pattern may be noise, especially after many comparisons.
Reconstructing the plan later
Screenshots and notes created after the exit can absorb information that was not available at entry.
Collecting sensitive data carelessly
Protect account identifiers, statements, API keys, and screenshots before uploading them to any journal service.
Research method and sources
On October 1, 2026, we reviewed five relevant first-page guides returned for the search intent around “trading journal guide” and “how to keep a trading journal.” Common coverage included objective trade data, subjective context, screenshots, fixed tags, R-multiples, and weekly review. This article adds a stricter separation between plan, log, and journal; a data-timing workflow; a net-R example; and safeguards against hindsight and small-sample overfitting.
- TradingJournal.com: What Is a Trading Journal?
- TradeLuma: Trading Journal Guide
- TradeHarbor: How to Use a Trading Journal
- The Trading Terminal: Guide to Trade Journaling
- Trading Journals: Five-Step Process
Definitions and calculations were checked against the CFI record-keeping material and Van Tharp's R-multiple lesson linked above. Competitor inclusion reflects search relevance, not an endorsement. Search results vary by country, language, device, and date.
Frequently asked questions
What is the difference between a trade log and a trading journal?
A trade log records transaction data such as instrument, entry, exit, size, costs, and result. A trading journal adds the plan, setup, market context, rule compliance, screenshots, and review notes needed to explain the decision. A useful journal normally contains both.
What should I record in a trading journal?
At minimum, record the instrument, setup, direction, planned entry, initial stop, position size, initial monetary risk, actual fills, costs, net result in money and R, rule compliance, and one short review note. Add fields only when they answer a recurring review question.
When should I fill in my trading journal?
Record the plan and initial risk before entry, actual fills and management decisions while evidence is fresh, and the final result after exit. Run a separate weekly review so one emotional trade does not cause an immediate strategy change.
How do I calculate R-multiple in a trading journal?
Define 1R as the initial planned monetary risk. Net R equals the net trade result after costs divided by that initial risk. For example, a net gain of $75 on an initial $50 risk is +1.5R. If actual loss exceeds the plan, the result can be worse than -1R.
How often should I review my trading journal?
Reconcile trades after each session and perform a structured review weekly or after a predefined number of trades. Longer monthly or strategy-version reviews can examine broader patterns. The schedule should be written in advance and kept consistent.
How many trades do I need before changing a strategy?
There is no universal number. It depends on trade frequency, variation in outcomes, market regimes, costs, and the size of the suspected effect. Treat small samples as diagnostic clues, not proof, and distinguish a broken rule from a tested strategy change.
Does keeping a trading journal improve profitability?
A journal does not create an edge or guarantee better results. It can make decisions, costs, rule violations, and recurring patterns auditable. Improvement still depends on data quality, a valid strategy, realistic testing, and acting on evidence without overfitting it.
Use the journal to preserve decisions, not manufacture certainty
A complete journal can reveal how you traded; it cannot guarantee that an observed pattern will continue. Keep raw records, label estimates, preserve strategy versions, and test a proposed change on new data before increasing risk.