The previous version called these “real” entries and attributed them to anonymized TSB users, then attached unsupported claims about revenge-trade reduction, emotion-linked performance, review thresholds, and the behavior of “most profitable” users. Those claims are withdrawn. The five entries below are explicitly constructed examples: their prices, outcomes, states, and notes are teaching data, not customer records or historical market events.
What a Useful Trading Journal Example Shows
A useful entry preserves four things separately: the plan known before the order, the execution facts, the outcome, and the review written afterward. That separation lets you ask whether the decision followed a defined process without allowing profit or loss to rewrite what was knowable at entry.
The five examples cover a planned winner, a planned loss, a revenge trade, a valid setup exposed to an unresolved timing rule, and a rule-breaking winner. None proves that a setup works or that a behavior always loses. Each shows how to record enough evidence for a later comparable sample.
| If this happened | Read | The decision question |
|---|---|---|
| Plan followed, positive result | Example 1 | Can the process be repeated without treating one win as validation? |
| Plan followed, negative result | Example 2 | Was it normal strategy risk, an execution issue, or a missing rule? |
| Loss followed by an impulsive trade | Example 3 | Which observable trigger and response should become an if–then control? |
| Valid signal near scheduled risk | Example 4 | Did a timing rule already exist, or is this only a hypothesis to test? |
| Rules broken, positive result | Example 5 | Would the decision still be acceptable if the outcome were hidden? |
If the terms “setup,” “planned risk,” and “R multiple” are still new, use the beginner trading-journal guide first. Then return here and copy the evidence structure rather than the example trades.
Why Most Trading Journals Fail—and What Good Entries Preserve
A trade log can state what filled. A journal adds what was intended, why the decision qualified, what changed during execution, and what should be tested next. The distinction is not “few fields versus many fields.” It is whether every important value has a source and a time.
| Layer | Examples | Captured when | Evidence label |
|---|---|---|---|
| Pre-trade plan | Setup version, trigger, invalidation, intended order, stop, target, maximum risk, scheduled-event rule | Before order submission | Contemporaneous plan |
| Execution facts | Order and fill IDs, timestamps and time zone, side, quantity, fill price, fees, currency, partial fills | Imported or reconciled after execution | Broker or venue record |
| Context and state | Session, volatility regime, news-calendar state, focus or emotion on a defined scale | Before or during the decision | Trader self-report / external context |
| Derived values | Net P&L, planned and realized R, excursion, holding time, rule-compliance flags | After raw facts reconcile | Calculated, with formula/version |
| Review | Process verdict, alternative explanation, lesson, proposed if–then rule, evidence needed | After the trade is closed | Post-outcome interpretation |
For securities transactions, Investor.gov says a confirmation commonly identifies the security, amount, and price and can be checked against the account statement. That is a useful evidence pattern even when another market exports different fields: preserve the source record, reconcile it, and never let a handwritten note silently replace the execution record.
Before reading P&L, ask whether the pre-trade trigger, risk, and execution matched the versioned plan. Baron and Hershey's decision experiments found that people rated otherwise identical decision processes more favorably when outcomes were favorable. Their study was not about traders, so applying an outcome-blind review here is a risk-control inference—not proof that it improves returns.
Example 1: The Clean Winner
This synthetic FX example ends positively, but the entry does not say “good trade because it won.” It records the plan version and the net result separately.
| Field | Illustrative entry | Why it is useful |
|---|---|---|
| Session / instrument | Sample session A · EUR/USD · long | No claim that this was a real date or market event |
| Plan version | BR-2.1 · close above range, then first retest holds | The setup can be reproduced rather than renamed after the result |
| Planned levels | Entry 1.0852 · invalidation 1.0827 · target 1.0902 | The initial distance defines 1R; target is 2R before costs |
| Actual execution | Entry filled as planned; target filled; recorded costs 0.06R | Gross outcome and friction are not blended |
| Net outcome | +1.94R | Comparable across differently sized hypothetical trades |
| State / exception | Distraction 1 of 5 (5 = severe); no checklist exception recorded | Uses a defined scale and a separate compliance flag |
| Review sentence | “Repeat waiting for the defined retest; do not infer that retests outperform from one result.” | Actionable without promoting a one-trade conclusion |
What the entry supports: this execution matched BR-2.1 and produced a positive net observation. What it does not support: that BR-2.1 has positive expectancy, that the retest caused the win, or that the same levels should be copied. Those questions require a defined, comparable sample.
Example 2: The Normal Loss (Good Execution)
A “good loss” means the recorded decision complied with the frozen process—not that the setup is already proven and not that nothing can be learned.
| Field | Illustrative entry | Review |
|---|---|---|
| Session / instrument | Sample session B · GBP/USD · short | Constructed record, not a historical fill |
| Plan version | REJ-1.4 · close-and-reject rule at a premarked level | All required checklist fields were recorded before entry |
| Planned levels | Entry 1.2715 · invalidation 1.2745 · target 1.2655 | Planned reward is 2R before costs |
| Actual execution | Stop filled with 0.02R adverse execution; other costs 0.01R | Execution effect is visible rather than blamed on the setup |
| Net outcome | −1.03R | One valid losing observation |
| Process verdict | Compliant under REJ-1.4 | Keep in the valid sample; do not change the rule from this outcome alone |
| Review sentence | “I would take the same decision under the same written plan; verify the execution deviation across later fills.” | Separates strategy variance from a possible routing issue |
The counterfactual question—“Would I take it again with only the information available then?”—is useful for detecting hindsight. It is not validation by itself. The entry belongs in the planned sample with the loss and costs intact.
Example 3: The Revenge Trade (Emotional Mistake)
This example follows the planned loss above. Its classification comes from observable process violations, not from the second loss.
| Field | Illustrative entry | Evidence |
|---|---|---|
| Trigger | Submitted 18 minutes after Example 2 closed | Timestamp sequence |
| Decision | Long GBP/USD after the short stopped; no setup ID | Blank setup field and order record |
| Risk | 1.4× the plan's normal risk; stop chosen after entry | Size, timestamped plan, and edit history |
| Self-report | Frustration 4 of 5 (5 = severe); note says “make it back” | Contemporaneous note, not inferred from loss |
| Net outcome | −1.44R | Constructed fill and cost calculation |
| Process verdict | Impulsive: no setup, size exception, post-entry risk definition | Would remain a violation if the trade had won |
| If–then control | “If I close a loss with frustration ≥4, then lock order entry and run the re-entry checklist before another order.” | Prospective rule to test, not a guarantee |
The corrective value comes from naming the trigger, the prohibited response, and the replacement action. General implementation-intention research tests this kind of when–then planning outside trading; applying it here is a reasonable experiment, not evidence that a timer or tag will eliminate revenge trading. The revenge-trading guide expands the interruption workflow and its limits.
Example 4: The “Perfect” Setup That Lost
“Perfect” is shorthand, not a defensible rating. The old example said everything was correct and then retroactively invented a rule to avoid a scheduled release. This version records the ambiguity directly.
| Field | Illustrative entry | Review |
|---|---|---|
| Session / instrument | Sample session C · micro equity-index future · long | No real contract date or release is claimed |
| Setup | ORB-3.0 · opening-range break criteria all recorded as met | Signal compliance is distinct from timing policy |
| Known context | Scheduled high-impact release in 15 minutes | Calendar status existed before entry |
| Plan gap | ORB-3.0 had no rule for scheduled releases | Cannot call the entry a rule breach retroactively |
| Net outcome | −1.05R after adverse stop execution | Outcome does not prove that releases should be avoided |
| Hypothesis | Compare ORB trades inside versus outside the declared event window | Predefine window, event source, outcome, costs, and minimum precision before testing |
| Temporary control | Mark event-window entries “not permitted” only if risk policy requires caution while evidence accumulates | A safety choice, not a backtested edge claim |
If an event-exclusion rule already existed, the same trade would be noncompliant. If no rule existed, record a candidate variable and test it; do not rewrite the old plan to make the loss look avoidable. That distinction protects the dataset from hindsight labels.
Example 5: The Bad Setup That Won (The Dangerous One)
A positive result can reward a decision that had no defined risk. This synthetic equity example uses a fictional symbol and round numbers so it cannot be mistaken for a historical recommendation or live price.
| Field | Illustrative entry | Process finding |
|---|---|---|
| Instrument | Fictional ABC · 100 shares | Not a real issuer or recommendation |
| Entry | $100.00 late in the session after an unverified tip | No setup or independent thesis |
| Risk plan | No invalidation, stop, or maximum loss recorded | Planned R is undefined |
| Exit | $103.20 next session · +$320 gross before costs | Positive outcome; still noncompliant |
| Gap exposure | Downside between sessions was not bounded | A stop, if entered, would not guarantee the stop price after a gap |
| Process verdict | Rule-breaking winner | Same verdict with P&L hidden |
| Review sentence | “Do not add this outcome to the validated setup sample; log it in the exception ledger and prohibit tip-only entries.” | Protects the strategy sample from contamination |
FINRA explains that a stop order becomes a market order when triggered and may execute at the available price; a limit can control price if filled but may not fill. The general lesson is narrower than “always use a stop”: predefine allowable exposure and understand what the selected order can and cannot guarantee.
The Universal Journal Entry Template
No template is literally universal across markets. A stock confirmation, futures execution report, OTC statement, and crypto fill export can expose different fields. This is a minimum decision-and-evidence schema; extend it without deleting raw source fields. The field-selection guide shows how to choose optional fields by a specific review question.
| Group | Required fields | Rule |
|---|---|---|
| Identity | Account alias, instrument, market, direction, order and fill IDs | Use pseudonymous account labels; never paste credentials or secret keys |
| Time | Order, fill, exit timestamps; explicit time zone | Do not mix venue, local, and chart time silently |
| Plan before entry | Setup/version, trigger, invalidation, intended order, planned size/risk, stop/target, context exclusions | Lock or timestamp before reading the outcome |
| Execution | Actual quantity and fills, fees/credits and currency, partial exits, financing/funding where applicable | Reconcile to original export or statement |
| Result | Gross and net outcome, planned and realized R, exit reason | Do not invent R when initial risk was undefined |
| Process | Checklist result, exceptions, intervention, plan-compliance verdict | Keep verdict independent of P&L sign |
| State | Optional predefined emotion/focus scale and capture time | Treat as self-report, not a causal diagnosis |
| Review | Repeat/avoid sentence, competing explanation, proposed rule, evidence needed, reviewer | Label hypotheses and version any plan change prospectively |
Use a chart image when it preserves decision context that the raw export does not, but keep it supplemental. Capture before-and-after states under a consistent template, mark source and time zone, and remove account identifiers. A screenshot cannot substitute for fills, fees, or edit history.
For planned-risk arithmetic, use the TSB risk/reward calculator, then store its assumptions with the entry. The calculator does not make an undefined stop valid or predict the fill.
How to Review These Entries
- Freeze source evidence. Preserve the original export or statement, then reconcile journal totals and missing fields before analysis.
- Separate pre-trade from post-trade text. An edit timestamp should reveal whether a “reason” was entered before or after the result.
- Audit completeness first. Report missing setup IDs, risks, fees, exits, or states; a clean-looking chart built on selective entries is not evidence.
- Judge process with outcome hidden. Apply the same versioned checklist to wins and losses, then reveal P&L.
- Use net outcomes. Reconcile fees, financing, funding, and partial fills before calculating expectancy, profit factor, or R.
- Compare like with like. Hold setup definition, market, sizing logic, and relevant regime fields stable enough for the question.
- Report sample size and uncertainty. NIST's statistical guidance makes the core point visible: estimate precision depends on sample size and variability. There is no universal “20-, 30-, or 50-trade rule.”
- Change one defined control. Version it prospectively, specify what would count as improvement or harm, and preserve the old rule.
Choose a cadence that matches activity: a session close can catch missing entries, while a longer review needs enough comparable observations to answer its question. The full trade-review workflow covers segmentation, expectancy, outliers, and change logs without treating a weekly ritual as proof of improvement.
Journal Entry Mistakes That Corrupt the Sample
- Calling a constructed example “real.” Label synthetic, composite, anonymized, and historical records accurately; they imply different evidence.
- Logging only memorable trades. A winner-only, loser-only, or screenshot-only subset cannot represent the full strategy.
- Writing the thesis after exit. Preserve a timestamped pre-trade plan, then put hindsight in a separate review field.
- Judging quality from outcome. A planned loss may be compliant and an unplanned win may still violate risk policy.
- Mixing gross P&L, net P&L, currency, points, and R. Store units and formulas; never compare unlike values under one “result” column.
- Inventing one sample-size threshold. Required observations depend on variance, effect size, dependence, regime, and desired precision.
- Treating emotion tags as causes. A correlation can reflect size, volatility, time, losses immediately before the trade, or selective tagging.
- Changing multiple rules at once. You will not know which change affected execution or whether the market regime changed instead.
- Publishing sensitive screenshots. Remove names, account numbers, balances, tokens, QR codes, and order links before sharing.
The goal is not a literary diary. One precise exception and one testable next action beat a confident story. If no pattern is supported, write “not enough comparable evidence yet.”
Use TSB to Preserve the Evidence Chain
Trader's Second Brain is our product. Its canonical registry recognized 328 broker, exchange, and platform source profiles when checked on September 7, 2026. Full Access has a lifetime-access route and includes the Prop Firm Challenge Tracker. In this guide, its useful role is importing execution facts, attaching process fields, and filtering the same schema across planned, impulsive, winning, and losing entries.
Recognition is not the same as perfect field coverage. An export may omit pre-trade intent, quote history, missed orders, edits, funding, or a reliable setup label. TSB cannot know what you planned unless it was recorded, prove causation from an emotion tag, manufacture statistical power, or validate a strategy from five examples.
Check the exact source and field route in the TSB import directory, then reconcile a small sample to the original records. If the evidence survives, open TSB and build the five-entry schema.
About These Examples
All five records are constructed solely to demonstrate journal structure. They are not TSB customer rows, audited broker statements, live prices, investment recommendations, or claims that the described setup has an edge. The equations are internally checked; the outcome-bias, order-type, confirmation, planning, and uncertainty notes use the sources below. Behavioral studies cited here were not tests of trading profitability, so their application remains an editorial inference.
- Investor.gov — trade-confirmation fields and statement reconciliation
- FINRA — market, limit, and stop-order execution trade-offs
- NIST — sample size and confidence-interval precision
- Baron and Hershey — outcome-bias experiments in decision evaluation
- Webb and Sheeran — experimental test of implementation-intention processes
Examples are educational and deliberately incomplete as market advice. They do not model liquidity, leverage, tax, gap size, margin, contract specifications, counterparty rules, or every cost. Use the rules and records of the exact account and venue; never place a trade merely to reproduce an example.
Final Verdict: Make the Entry Auditable
The best trading-journal example is not the most dramatic one. It is the one another reviewer can reconstruct without guessing which facts were known before the trade. Preserve plan, execution, outcome, and review as separate layers. Label synthetic examples as synthetic. Keep losses, rule-breaking wins, costs, missing values, and revisions in the record.
Do not promote one winner into an edge, one loss into a broken setup, or one emotional tag into a cause. Build a comparable sample, report its size and uncertainty, and version changes prospectively. If you need a blank copy, the trading-journal template guide provides formats you can adapt without copying these hypothetical trades.
