Trade anatomy — case studies from setup to exit

By Jarosław Wasiński • Published: May 3, 2026 • Updated: Oct 31, 2026 • 65 min read

A trade's result tells us how much was made or lost, but does not explain the quality of the decisions. A profit can occur despite an error, and a loss after the plan was followed correctly. This article follows five illustrative trades from setup evaluation and preparation and position sizing through entry and management to closure and journal review. We examine execution prices, costs, initial risk and price movement while the position remained open. You will also see how to distinguish slippage from a departure from the plan and when a result in R cannot be calculated reliably. These are educational models designed to demonstrate a review method for your own records, rather than establish a strategy's effectiveness from individual outcomes.

ANATOMY OF A TRADE. Educational illustration; the article explains the detailed rules and assumptions.

Open the image at full size

Conceptual overview: the hero groups several actions into five panels. The article separates modification from monitoring and uses six stages; their complete sequence appears in Section 7.

Reconstruct the fills before judging the decision

“One losing trade” can contain several executions at different prices. Reviewing it against only the first entry can misstate the price result, while dividing by a convenient risk number can distort the R-multiple. Begin with the actual quantities, prices and separate cash charges, then inspect the decisions that produced them.

Our hypothetical EUR/USD long has two entries and one complete exit. All quantities are EUR units, all prices are USD per EUR, and the account is in USD. The first entry buys 10,000 EUR at 1.1020; the second buys 5,000 EUR at 1.1032. The full 15,000 EUR is later sold at 1.1018. These are executable fill prices, not midpoint quotes.

The first entry has a 0.70 USD commission, the addition 0.35 USD, and the complete exit 1.05 USD. There is no financing, conversion or other charge in this model. Spread and any price effects already reflected in the supplied fills must not be deducted again as a separate estimate. This invented example establishes bookkeeping, not the wisdom of adding to a position.

Weight the entry price by the executed quantity

Hypothetical execution ledger
Execution	EUR units	USD per EUR	Separate commission in USD
First buy	10,000	1.1020	0.70
Additional buy	5,000	1.1032	0.35
Complete sell exit	15,000	1.1018	1.05

The combined entry price is (10,000 × 1.1020 + 5,000 × 1.1032) / 15,000 = 1.1024. It is nearer the first fill because that fill is twice as large. The unweighted average of the two prices would assign equal influence to unequal quantities.

Against the weighted entry, the price result is 15,000 × (1.1018 − 1.1024) = −9.00 USD. Check it by leg: the first quantity contributes −2.00 USD and the added quantity −7.00 USD. Both methods produce the same total. Treating the entire position as entered at the first price would instead show −3.00 USD, understating this loss by 6.00 USD.

Keep cash costs and the R denominator explicit

Total commission is 0.70 + 0.35 + 1.05 = 2.10 USD. The net result is −9.00 − 2.10 = −11.10 USD. The official MQL5 deal-property reference separately identifies executed volume, price, profit, commission, swap and fees. A real export must be checked for what each field includes; a column labelled “profit” is not evidence that every charge is already inside it.

For this case only, define a fixed price-risk reference after both entries, using their common recorded stop at 1.1000. The first quantity contributes 20.00 USD of reference risk and the second 16.00 USD, giving 36.00 USD. This reference excludes commissions and does not guarantee a stop fill. At the stated stop price with the same total commission, the modeled loss would be 38.10 USD.

Using that declared reference, the actual net result is −11.10 / 36.00 ≈ −0.3083R, rounded to four decimals. Do not present the 36.00 USD price reference as an all-in maximum loss. Another journal may define R differently; comparison requires the same stated convention rather than quietly changing the denominator after seeing the exit.

Attach the arithmetic to a sequence of decisions

Record the initial decision, permission for the addition, actual fills, stop state, exit instruction and final confirmation separately. The calculations assume both entries and the complete exit are confirmed and that their quantities reconcile. They do not establish that either entry satisfied a playbook or that the addition fitted the account’s exposure limits. Those require contemporaneous evidence.

For a review, preserve the entry and addition as separate rows even if the platform displays one combined position. Keep any later amendments with their timestamps. A blended price helps reconcile the result, but it should not erase which decision introduced the added exposure or which facts were available then.

The five longer cases below examine successful and unsuccessful outcomes, compliance and behavioral errors. Use this compact ledger to keep their two questions distinct: what happened financially, and which decisions complied with the declared process? A correctly reconciled loss is useful evidence; one isolated result still does not prove that the strategy or the addition has a positive expected return.

From a reconciled execution ledger to the full trade anatomy

Keep quantities, cash costs and reference risk consistent while examining each stage of the five detailed cases below.

1. What trade anatomy means and what it teaches

Trade anatomy means breaking one trade into individual decision stages, each with its own decision, metrics and errors to investigate. “Bought, sold, +1.3R, next” treats it as one indivisible outcome. A more useful review recognizes six to eight smaller decisions, each potentially correct or mistaken independently of final P&L.

Why does that matter? A single financial outcome does not establish decision quality. A profitable C-setup may have succeeded through favorable randomness; an A+ setup executed correctly may still lose after an unpredictable market move. Reinforcing behavior solely by profit and loss rewards some random outcomes instead of repeatable process. Over hundreds of trades, the resulting habits can become a mixture of noise and accidental feedback rather than a deliberate method.

How stage-by-stage analysis differs from a trade review

A conventional review may include an entry/exit chart, short comment and star rating. Anatomical analysis evaluates each stage separately: pre-trade, entry, in-trade, modification, exit and post-trade. Record a numerical value or sortable category for each. Section 7 details the tasks, metrics and traps; Section 8 provides the full field list.

Those recorded values build initial statistics about your own habits after several dozen trades, not just statistics about the market. That is an exploratory description, not a sufficient-sample rule. Sometimes a valid setup is damaged along the way by incorrect size, entry into a news window, an unplanned manual exit or a missing stop.

How many observations are needed, and what they cannot decide
A trade count alone proves nothing. Report it with an interval, method and assumptions: after n trades, the estimated win-rate interval or expectancy interval is such-and-such. A normal-approximation 95% expectancy interval is E ± 1.96 × std(R)/√n; std(R)/√n alone is a standard error, not an interval. Independence is an assumption; serial dependence and shared events can make this interval too narrow.
At a 50% win rate, the Wald normal-approximation half-width is about ±9.8 percentage points for n = 100 and ±5.7 for n = 300. The Wilson interval, generally preferable to Wald near small samples or boundary rates, gives about ±9.6 points at n = 100 and 50% wins. These describe measurement precision, not an admission threshold or reliability guarantee.
Operational implication: if the interval spans the decision threshold, defer the evidence-based decision: observe, use the applicable lower size and review again rather than declaring either a confirmed edge or a dead setup. Operational-risk stops, such as a loss limit or execution error, need not wait for statistical resolution. 15.9 develops metrics and intervals; 15.10 defines setup decisions.

Illustrative example — synthetic observations, not an account record: a promising backtest versus early live trades. A London breakout strategy's backtest looks good, but the first dozen or so live trades do not. Before concluding that the strategy stopped working, inspect (a) look-ahead bias: a historical intrabar decision using ATR(14) calculated from the entire same candle uses information unavailable until that candle closed and can make A+/B classification overly favorable. Reading the still-forming ATR live is not itself look-ahead, but without recording that provisional value the classification may not be reproducible later. Also inspect (b) modeling data: M1 candles omit the ordering of intrabar false breakouts, which materially affects an MT4 breakout test. A dozen live trades cannot settle whether an edge exists, but can reveal a data-processing defect. Check the scanner's closed-bar logic and test-data provenance before replacing the strategy. 15.4 covers test methodology.

Six stages of one trade — Profit and compliance with the plan are different assessments.

Open the diagram at full size

What this article will teach you

Five cases follow a well-executed winner, a correctly executed loser, a loser with an error, a random trade and a winner impaired by behavioral mistakes. Each includes concrete pips, holding times, slippage and net R-multiples where R is defined, alongside practical complications. Even a good-looking outcome can hide a process error.

Illustrative example — synthetic observations, not an account record: the trader whose best results came from C-trades. A journal of several hundred trades is profitable, so the trader doubles size. The following months deteriorate by more than size multiplication alone would explain. Reclassification reveals that much of the earlier profit came from bored afternoon C-trades rather than the A+ playbook setups. Doubling size itself scales gains and losses symmetrically; it does not create an edge or cause an extra deterioration by arithmetic alone. In this example, C-trades have small winners and full-stop or worse losers, so larger size magnifies the monetary consequences of that unfavorable asymmetry when losses occur. The changed sequence or distribution, not mere multiplication, explains damage beyond a simple scale-up. If you cannot identify the sources of your results, do not scale. Classify first, journal by category, then decide size under 15.11.

2. Case 1 — Profitable A+ trade: London breakout, EUR/USD, +1.21R

The first case follows a trade whose main decisions remained within the plan. A completely faultless six-stage trade is a demanding standard; here the process was broadly preserved, but small operational defects reduced the result and must remain visible in the assessment.

Illustrative composite, not a historical result or account record. Times, prices and balances are synthetic teaching inputs, not one identified session, account or person. All five cases use USD/PLN = 3.80. The purpose is journal structure and management mechanics, not a market forecast. “PL” times mean Polish local time, Europe/Warsaw; convert dated events to your own time zone. The 09:14 observation below refers to the latest completed M5 bar, not a claim that a standard five-minute bar closes at that minute.

Market context

A working-day London-session scenario. EUR/USD traded in a narrow overnight Asian range from 1.0832 to 1.0851, 19 pips. Daily ATR(14) is 87 pips, within the setup's required 60–120 range. The event window defined in Section 4 is clear of high-impact releases; the next comes in the afternoon. DXY remains in a weekly uptrend but has begun a small daily retracement, a context the scenario treats as supportive of an EUR/USD long.

Setup: binary A+ classification

London Open Breakout condition	Requirement	State at 09:14 PL	YES/NO
Pre-London range, 08:00–09:00 PL	12–35 pips	19 pips, 1.0832–1.0851	YES
Daily ATR(14)	60–120 pips	87 pips	YES
Macro calendar under Section 4's event-window table	No high-impact release in the exclusion window	Clear; next release in the afternoon	YES
Correlations and existing EUR/USD exposure	No conflict	No open positions	YES
Time of day	09:00–10:30 PL	09:14 PL	YES
Completed M5 breakout-candle body	At least 0.3 × M15 ATR(20)	0.4 × M15 ATR(20)	YES
Breakout-candle tick volume	At least 1.2 times the previous 20 M5 bars' average	1.4 times	YES

All seven conditions are checked individually, giving A+. Class definitions, mandatory conditions and sizing by class, illustrated here as 1% of equity, belong in the playbook card in 15.2. This case uses that class to describe a specific planned trade.

Decision

At 09:14 PL, the scanner reports the completed breakout bar closing at 1.0863, body 0.4 × M15 ATR(20), and tick volume 1.4 times the 20-bar M5 average. Confirmation filters are green. The decision takes 23 seconds after the 09:14:08 alert: all seven items are checked separately, including calendar, correlations and sizing. A scanner alert does not waive that review. There is no discretionary increase in lot size merely because the breakout looks strong.

Sizing

Account equity: PLN 28,400. The 1% budget is PLN 284.
Stop: the opposite range edge, 1.0832, minus a five-pip buffer = 1.0827. Distance from planned entry at 1.0863 is 36 pips.
EUR/USD value at 0.1 lots is PLN 3.80 per pip under USD/PLN = 3.80, or PLN 38 per pip per standard lot.
Required size = PLN 284 / (36 pips × PLN 38 per pip per lot) ≈ 0.208 lots. Equivalently, divide by the 0.1-lot pip value and multiply the result by 0.1; do not lose the unit conversion.
Round down to 0.20 lots: R0 = 36 × PLN 7.60 = PLN 273.60, or 0.96% of equity. Fix that R denominator before sending the order and retain it throughout.

Execution

09:14:31 PL: market buy 0.20 lots EUR/USD.
Displayed bid 1.08620 / ask 1.08635, a 1.5-pip spread. The scenario uses a standard account without commission; Section 10 separately models a commission-based raw/ECN cost variant, whose spread assumptions must also be specified.
Fill 1.08638: 0.3 pip adverse slippage versus the displayed ask, within the scenario's tolerance.
Click-to-confirmation time: 180 milliseconds.

Trade timeline: information available at each moment

Illustrative chronology, not historical performance. The available-information column prevents hindsight from rewriting the reason for a decision.

Time, PL	Information then available	Planned decision	Actual action	Order / position state	Cost / metric	Evidence	Deviation
09:14:08	Completed M5 bar at 1.0863; body 0.4 × M15 ATR(20); tick volume 1.4 times average.	Check all 7 conditions individually.	7/7 checked; A+.	No positions or orders.	—	Checklist record and scanner alert log.	None.
09:14:31	Bid 1.08620 / ask 1.08635; 1.5-pip spread.	Buy 0.20 at market with SL 1.0827 in the order ticket.	Market buy of 0.20 lots submitted.	Sent, not yet filled.	—	Order ticket.	None.
09:14:31 +180 ms	Fill confirmed at 1.08638.	Fill near displayed ask 1.08635.	Filled at 1.08638.	Long 0.20; server-side SL 1.0827.	0.3-pip slip = PLN 2.28.	Execution confirmation.	Below the 1.0-pip tolerance.
09:16	1.08584, the lowest observation after entry.	No action below management triggers.	No action.	Long 0.20; SL 1.0827.	MAE −5.4 pips = −0.15R.	Recorded price path reconciled with account history; ordinary order history alone need not contain excursions.	None.
09:51	Price 1.08998 = the +1R price-distance level (entry fill plus 36 pips).	Sell-limit 0.10 at 1.08998; remaining stop to entry.	Manual market sale of 0.10, fill 1.08993.	Long 0.10; price break-even SL 1.08638.	0.5-pip shortfall on 0.10 = PLN 1.90.	Order history.	YES: market order instead of the preplaced limit.
10:15 M15 bar 10:00–10:15 closes	Completed-bar M15 SMA-20 is above entry plus 1R; trailing activation condition met.	SL = SMA-20 on M15 minus 5 pips; ratchet upward only after completed M15 bars.	No action; sound alert missing.	Long 0.10; SL still at entry.	—	M15 chart with SMA-20; no modification record.	YES: one M15 bar, 15 minutes, late.
10:30 M15 bar 10:15–10:30 closes	SMA-20 = 1.0903; price 1.0928.	SL 1.0903 minus 5 pips = 1.0898.	SL moved to 1.0898, 34.2 pips above entry.	Long 0.10; SL 1.0898.	—	Order-modification history.	No new deviation; follows the missed 10:15 action.
12:14 (observation); 12:15 (modification)	Price reaches 1.0938, the highest observation while the position is open; SMA-20 on M15 = 1.0921 after completion of the 12:00–12:15 candle.	Trail behind SMA-20 on M15, updating only after completed bars.	SL moved to 1.0921 minus 5 pips = 1.0916.	Long 0.10; SL 1.0916.	MFE +74.2 pips = +2.06R.	Price-path record and modification history.	No action breach; thought of adding size was recorded.
13:01	M5 price crosses below 1.0916.	Exit through trailing stop.	Fill 1.09155, closing the remaining 0.10 lots.	No position or pending orders.	Separate ECN commission sensitivity: PLN 4.94 for the complete 0.20-lot round trip, not an extra cost in the standard-account case.	Broker report reconciled with journal.	None.

Exit and calculation

First 0.10-lot tranche: 35.5 pips × PLN 3.80 = PLN 134.90, approximately PLN 135.
Second 0.10-lot tranche: 51.7 pips × PLN 3.80 = PLN 196.46, approximately PLN 196.
Total: PLN 331.36, rounded to PLN 331.
Original rounded-volume risk denominator: R0 = 36 × PLN 7.60 = PLN 273.60.
Full-position result: 331.36 / 273.60 ≈ +1.21R. Actual entry and exit fills already incorporate spread; do not subtract it again. This excludes commission, swap and tax. Holding these fills fixed and deducting the separate PLN 4.94 ECN commission sensitivity gives approximately +1.19R.

Holding the entire position to the same 13:01 exit, with no partial TP1 and otherwise unchanged execution, would give approximately +1.44R. The highest observed price, 1.0938, represents MFE +74.2 pips = +2.06R. The article's working capture rate, realized R divided by MFE in R, is 1.21 / 2.06 ≈ 0.59. This is a local journal measure, not an industry standard; compare it against your own consistently defined benchmark. Partial exits lower this case's average realized R but can reduce give-back when a move reverses; whether they stabilize a strategy's distribution requires testing across trades.

The source's playbook labels the profitable A+ case above +1R as a three-star reference for weekly review. Preserve the useful setup and entry process, but do not let that overall label erase the two management deviations recorded below.

Review: practical defects inside a winner

Not everything was ideal. Three journal observations:

09:51: TP1 was entered manually four seconds after the level was touched. The intended limit should have been placed at entry. The 0.5-pip shortfall on 0.10 lots cost PLN 1.90, revealing missing mechanical execution.
10:15–10:30: trailing activation was one M15 bar late. SMA-20 exceeded entry plus 1R at 10:15, but no sound alert was set. The measurable process cost is 15 minutes without the planned tighter protection; a specific lost pip amount is not established.
12:14: euphoria. The trader recorded a thought about changing trailing or adding size but did not act. Record it as a possible behavioral warning, not an executed breach or proof that the next trade will fail.

Binary journal conclusions:

Plan matches the playbook? YES.
Entry execution matches the plan? YES, with 0.3-pip slippage within tolerance.
Modifications match the plan? NO: trailing activated one M15 bar, 15 minutes, late because the required sound alert was absent.
TP1 executed as planned? NO: manual market instead of preplaced limit; 0.5 pip or PLN 1.90 shortfall.
Behavioral-discipline breach? NO: the thought of increasing size was recorded but not acted on.

Proposed repairs: (a) an alert for the completed-M15 trailing activation condition and (b) TP1 as the tested preplaced limit mechanism instead of a manual market action.

Case 1 · EUR/USD: two exits — USD/PLN 3.80; baseline excludes commission. ECN variant: PLN 4.94 fee.

Open the diagram at full size

Illustrative mechanism, not a measured account test: intervention bias can damage a system. A trader starts manually closing whenever MAE approaches 60% of the stop distance, deciding that the position is not working. This can truncate losses before −1R but also remove trades that would later recover to break-even or +1R. Different interventions change different parts of the distribution; a high or rising win rate can coexist with a losing account if payoff asymmetry is poor. No particular win-rate increase is implied by closing losers early, and no such test was performed here. Record every off-playbook manual intervention explicitly. Repeated entries trigger a temporary restriction to the tested SL/TP/trailing process while the behavior is reviewed. Determine whether the effect exists in your own activity by comparing R distributions before and after with uncertainty, using 15.9.

3. Case 2 — Losing A+ trade with correct execution, −1.06R

This case shows a losing trade without a process breach. The setup is A+, execution matches the plan, no unauthorized management changes occur and the stop still fills. Under an illustrative 55% win-rate assumption, not a measured property of this strategy, 45% of trades lose; any of those losses can be plan-compliant. The loss is part of the modeled distribution rather than proof of failed discipline.

Illustrative composite, not historical performance or an account record. The numbers demonstrate a GBP/USD London-opening scenario. Equity continues from Case 1 using the source's whole-PLN rounding. PL timestamps retain the Polish local-time context; 09:21 is the observation/decision time for an already completed M5 bar, not a standard M5 close boundary.

Market context

GBP/USD's 08:00–09:00 PL pre-London range is 1.2914–1.2937, or 23 pips, within the required 12–35. Daily ATR(14) is 105 pips, within 60–120. No high-impact event falls inside Section 4's exclusion window. DXY is ranging intraday with no strong USD bias. Dates and prices remain illustrative.

Setup: structurally the same A+ conditions

Every playbook condition passes, including range, ATR, calendar, correlations and time of day, alongside the breakout-body and volume checks below.
Class A+; the scenario allocates 1% of equity. Rounded equity after Case 1 is PLN 28,731, giving a budget of approximately PLN 287.

Decision, sizing and execution

At 09:21 PL, the latest completed M5 breakout bar has closed at 1.2945, eight pips above the 1.2937 pre-London high. Tick volume is 1.3 times the M5 reference average, above 1.2, and body size is 0.55 × M15 ATR(20).

The decision takes 18 seconds with individual checklist checks. SL = 1.2914 minus five pips = 1.2909; planned-entry distance from 1.2945 is 36 pips. PLN 287 / (36 × PLN 3.80 × 10) ≈ 0.2098 lots. Rounding down to the supported 0.01-lot step gives 0.20, so R0 = 36 × PLN 7.60 = PLN 273.60. As in Case 1, the denominator uses the planned stop distance and actual rounded volume.

09:21:42: market buy 0.20 lots GBP/USD. The scenario spread is 1.8 pips; it is not asserted as a measured universal opening spread. Fill 1.29465 is 1.5 pips worse than the planned 1.29450 reference. This is total entry shortfall against the plan; without the contemporaneous ask, it cannot all be labeled slippage against the executable quote.

In-trade: what developed differently

09:21:42, entry: fill 1.29465; SL 1.2909.
09:35 PL: price 1.2952, MFE +5.5 pips or about +0.15R. The move is weak.
09:48: price 1.2940, back toward the range; adverse excursion −6.5 pips. Uncomfortable, but within the setup's permitted behavior.
10:02: price 1.2929, in the range's lower half; adverse excursion −17.5 pips. It resembles a false breakout, but the plan does not authorize an early discretionary close.
10:21: price 1.2918; adverse excursion −28.5 pips.
10:34: price 1.2912.
10:38: price crosses below 1.2909 and the stop fills at 1.29085, 0.5 pip beyond the stop. All 0.20 lots close within the scenario's execution tolerance.

Result

Entry-to-exit loss: 38.0 pips × PLN 7.60 = PLN 288.80, rounded to −PLN 289.
R-multiple = −288.80 / 273.60 ≈ −1.06R. Beyond the planned 36 pips are 1.5 pips of entry shortfall and 0.5 pip at the stop. Do not revise R0 to hide the excess. A loss beyond −1R can arise from execution even when the procedure was followed.
Holding time: approximately one hour 16 minutes, specifically 1:16:18 with the displayed timestamps.

Review

Process-error classification: none. Decision, rounded size, execution tolerance and management follow the plan. The structurally defined stop is triggered. The source awards three stars for process, independent of negative P&L; unlike Case 1's favorable overall label, no late trailing or manual-TP deviation needs qualification here.

Possible market context to investigate in weekly review, without treating one trade as causal proof:

A media comment from an FOMC member during London trading, recorded as contextual information rather than a demonstrated explanation.
False breakouts during the first 60–90 minutes of London trading: a hypothesis to measure in your own breakout data, not an asserted market constant.
The assumed outcome distribution: with a hypothetical 55% win rate, nearly half of A+ trades lose around their planned risk; this synthetic case represents one such outcome.

The journal still records three warnings: at 10:02, a thought of giving the position five more pips; at 10:21, switching attention to news; and at 10:38, considering an immediate reversal after the stop. None becomes a trade or order change. Recording these observations helps identify behaviors that might produce a process breach under different conditions, without treating the thought itself as one.

A correctly executed losing A+ is not by itself a strategy-change signal. With independent trades and the assumed 55% win rate, five losses in one specified five-trade window have probability 0.45⁵ ≈ 1.85%. That is different from seeing at least one five-loss run anywhere in the next 100 trades: the exact finite-state calculation gives approximately 65%. Dependence can alter that probability; persistent loss clustering can make it higher. A five-loss run therefore calls for the predefined review of data processing, regime and classification rather than automatic reconstruction of the strategy. A single plan-compliant loss requires documentation and continued adherence to the playbook, including its operational stop limits.

Case 2 · loss and execution — A 1.5-pip plan shortfall does not establish slippage against the current Ask.

Open the diagram at full size

Illustrative example — synthetic observations, not an account record: changing strategy after every losing A+. After each pair of losses, a trader adds a filter, such as requiring a matching DXY trend. Months later, a dozen extra filters reject most signals, each invented after two observations. No versioned specification or test validates the combined new system. That is the problem: an untested strategy change without an evidence-based decision rule. A correctly executed A+ loss belongs to the distribution, not automatically to a repair request. 15.10 and 15.3 explain when and how to version a justified change.

4. Case 3 — Losing FOMO C-trade without a recorded stop: R undefined

The third case breaks the process at several stages. Size is selected without a risk calculation and losses exceed the account's usual one-percent budget. Because no stop level was recorded before entry, the loss cannot be expressed as exceeding 1R: R is undefined. This example illustrates a cascade of pre-trade errors rather than establishing how frequently such losses occur across retail traders.

Illustrative composite, not a historical result or account record. A synthetic US high-impact release occurs at 16:00 Polish local time, corresponding to 10:00 New York under the assumed six-hour offset. US/European daylight-saving mismatch weeks require a dated conversion. AUD/USD prices and reactions are teaching inputs. Equity continues from Case 2 using the source's displayed rounded-account sequence.

Market context

On the US-release day, the trader missed the EUR/USD London open because of a work call. The call ends at 15:18 PL. Opening the platform, the trader sees AUD/USD up 35 pips over 90 minutes and a large body on the latest completed H1 candle. The immediate thought is “there is my setup”, even though the planned London strategy's time window has passed.

The actual setup class is C, not A+

The trader checks filters hurriedly:

Pre-London AUD/USD range: not monitored for this session; London opening is long past.
Daily ATR(14): 78 pips, within the numerical filter.
Calendar: the US high-impact release is 38 minutes away at entry, so the calendar condition fails. This falls inside the governing ±2-hour exclusion in article 15.6 and also fails the narrower legacy example window discussed below.
Correlations: no other open positions.
Time of day: 15:21 PL, outside the London-breakout window and close to the US release.

The calendar and time-of-day conditions fail: class C, reject and record the reasons. Instead, the trader rationalizes that the H1 candle is strong and the release might reinforce the move.

Decision: the process breach

At 15:22 PL the trader buys 0.30 lots AUD/USD at market. Pip value is PLN 3.80 per 0.10 lots or PLN 11.40 for 0.30 under USD/PLN = 3.80. Volume is chosen without a risk calculation. No stop level is recorded or submitted before entry: “I will handle it manually after the release.” There is no pre-entry R0 or defined stop-based risk percentage, only volume 1.5 times Case 1's at similar equity.

Four errors occur in this decision:

Labeling the setup A+ instead of C through FOMO.
Selecting volume without a stop-based risk calculation.
Entering without a server-side SL.
Entering 38 minutes before a high-impact release without an event plan.

Execution

At 15:22:08 the fill is 0.7218, with the scenario's 0.8-pip adverse slippage and 2.3-pip spread. The source treats these as deteriorated pre-release conditions, not a verified normal-spread comparison. If the quoted bid is unchanged while the fill slips beyond the displayed ask, the immediate mark-to-bid shortfall is 3.1 pips: 2.3 spread plus 0.8 slippage. This is an entry snapshot, not another cost to subtract after calculating P&L from actual fills.

In-trade: escalation

15:22, open: 0.7218.
15:29 PL: 0.7224, MFE +6 pips; no action.
15:46: 0.7211, adverse excursion −7 pips.
15:58: 0.7223, a small recovery before the release.
15:59: 0.7220.
16:00 release and the ensuing first minute: the M1 bar develops a 28-pip range and price reaches 0.7195, −23 pips from entry; spread reaches 7.5 pips. The full first-minute range is known after that minute, not at 16:00:00.
16:01: 0.7187, adverse excursion −31 pips. With no submitted stop, the position remains open without the planned automatic exit mechanism.
16:02: 0.7172, adverse excursion −46 pips. The trader requests manual closure; 1.2 pips of additional adverse slippage produces a fill at 0.71708.

Result

Loss: 47.2 pips × PLN 11.40 = PLN 538.08, rounded to −PLN 538.
R-multiple: undefined. No stop level was fixed before entry, so no pre-trade R0 exists. Do not invent a hypothetical structural-stop denominator after the loss. Record currency and equity percentage: approximately PLN 538, or 1.9% of the source's PLN 28,443 displayed equity after Case 2. Its whole-PLN teaching sequence continues at PLN 27,905. Exact fill cash flows for Cases 1–3 give PLN 27,904.48; the small difference is intermediate rounding, not extra profit.
Reference budget: one percent at entry is approximately PLN 284. The loss is almost twice that budget; this comparison does not make it a defined R-multiple.
Holding time: approximately 40 minutes.

Review

Error classification: FOMO, missing SL, uncalculated size and a breached calendar filter. Each is independently sufficient to record a process error. Together they make a useful negative case to revisit in monthly review.

The journal reconstructs the FOMO sequence:

No captured setup for three days: macro filters excluded Monday and Tuesday London sessions, then a work commitment caused the Wednesday miss.
15:18: the call ends with accumulated frustration.
15:19: the first visible AUD/USD H1 candle looks strong.
15:20: “Something is finally happening” and “I must make up for the week.”
15:22: the order is submitted.

The mechanism is three inactive days → desire for action → rationalizing the first plausible-looking setup → skipped filters. This article calls it opportunity hunger as a working label, not a recognized literature term.

Corrective proposals belong in the change register, not automatically in tomorrow's strategy. Changes to setup filters or thresholds require a versioned specification and appropriate evidence review; several similar incidents can reveal a problem without establishing a statistically reliable new trading edge. Purely operational repairs, such as a missing alert or correct order type, can be implemented immediately:

Proposed anti-FOMO control: after three days without a setup, cap planned risk at 0.5% for the next five days. Review and document this candidate rule rather than asserting it is universally effective.
The source proposes a 60-minute pre-release ban for standard entries around FOMC, Employment Situation/NFP, CPI and ECB/BoE decisions. When using the current article 15.6 checklist, its broader ±2-hour exclusion already covers that proposal and remains binding. Dedicated event strategies require a separately specified and validated card, not an improvised exception.
A submitted SL at entry is mandatory. If a protected-entry request is rejected, do not knowingly proceed with an unprotected entry; if a position has already filled, follow the emergency confirmed-state procedure rather than assuming no exposure exists.
If every required checklist condition has not been checked individually, classify C. Scanner assistance does not waive checks. Decision duration alone is not a quality filter: Cases 1 and 2 completed their checks in 23 and 18 seconds.

News protocol: the calendar filter

Position status	Time relative to high-impact release	Action in this playbook
No position	More than 60 minutes before	The legacy example allows an A+ candidate beyond 60 minutes. The current checklist (15.6) also requires more than 2 hours from the relevant release on either side and all other conditions; 60–120 minutes before remains excluded.
No position	30–60 minutes before	No standard breakout/range entry. A dedicated event setup requires its own playbook card and permitted window.
No position	Less than 30 minutes before through 15 minutes after	No entry under this ordinary setup card. This shorter legacy window lies wholly inside the current (15.6) ±2-hour exclusion and does not define its boundary.
No position	15–60 minutes after	Still no entry under the current checklist (15.6): its 2-hour post-release exclusion remains active even if spreads stabilize and M15/M5 direction confirms. Those observations matter only after the governing window expires.
Planned position open	More than 15 minutes before	Continue with the original stop unless another plan trigger applies.
Planned position open	Within 15 minutes before	Apply the prewritten decision: a permitted break-even stop if profitable, or close. Do not improvise during the release.

15.6 owns the mechanical filters and their calibration. The Polish case study retains a narrower legacy sequence: more than 60 minutes before, 30–60 before, less than 30 before through 15 after, and 15–60 after. Those rows are preserved here to explain the original scenario, but the current 15.6 admission checklist requires no high-impact event within ±2 hours for either currency. That broader entry veto governs throughout the shorter subwindows; use strictly more than two hours at the boundary. The separate 15-minute management trigger concerns an already open position, not permission for a new entry. A different event strategy requires its own predeclared card. Other mentions, including Case 1's calendar and Section 9's automatic block, refer here. Check the calendar before opening the trading platform. Use a consistent source; BLS publishes the Employment Situation schedule and the Fed publishes FOMC dates. Record which events are high-impact for each pair: for EUR/USD, FOMC, NFP, US CPI and ECB decisions; add BoE and UK CPI for GBP/USD and BoJ for JPY pairs.

FOMO: A TRADE WITHOUT A RISK PLAN. Educational illustration; the article explains the detailed rules and assumptions.

Open the image at full size

Illustrative example — synthetic observations, not an account record: unrecognized FOMO trades. A trader labels every losing entry “bad luck” or “irrational market”. Reviewing the hours before entry reveals markers in some losses: a missed opportunity, days without a signal, frustration or “I must do something”. A separate FOMO category and a documented control create something the bad-luck label cannot: a measurable frequency. Whether and how much that frequency falls in your own activity requires your own sample and interval. No general reduction figure is supplied. A vague luck category often means the journal has not identified what needs repair.

5. Case 4 — Random boredom C-trade, −0.72R

This boredom trade has no valid playbook setup or complete management plan, only the urge to do something. Its loss is below 1R, yet its process is worse than Case 2's correctly executed A+ and shares Case 3's unjustified entry. The problem is not the small loss alone: there is no tested setup-based reason to expect an edge.

Illustrative composite, not historical performance or an account record. Times and prices are synthetic. The displayed equity sequence continues from Case 3 at PLN 27,905, retaining the source's whole-PLN rounding convention.

Market context

At 11:30 Polish local time on the scenario's working day, EUR/USD has ranged between 1.0795 and 1.0808, 13 pips over 90 minutes. M15 ATR(14) is 60% of its reference average. Relevant high-impact releases for this example have passed; New York activity has not yet begun. This is the trader's quiet late-morning window, not a claim that the London FX session ends at 11:30.

The supposed setup: no setup at all

The trader has been at the screen since 08:00. An earlier London breakout returned approximately +1R and was journaled, but it is outside the five-case accounting sequence and is not included in any total here. From 10:00 to 11:30, none of eight watched pairs produces a playbook signal.

At 11:32: “I must do something; I am wasting the day.” The trader opens EUR/JPY, which looks interesting because the yen has weakened this week. The H1 candle completed at 11:00 has a body of 0.3 × M15 ATR(20). This is a neutral candle in the scenario's quiet range, not a qualifying setup.

Decision: emotion overrides the rules

The trader buys 0.15 lots EUR/JPY. A stop 25 pips below entry is written in the journal before the click but never submitted to the platform. For a standard JPY-quoted lot, one pip is JPY 1,000. With USD/PLN = 3.80 and USD/JPY = 149, JPY/PLN = 3.80/149, giving approximately PLN 25.50 per pip per lot, PLN 2.55 per 0.10, and PLN 3.8255 per 0.15. The source uses PLN 3.82 as its truncated teaching input, consistently giving R0 = 25 × 3.82 = PLN 95.50, approximately 0.34% of PLN 27,905. This percentage is recognized afterward, not used to select the trade. R0 exists because the stop level was recorded before entry; Case 3 had no such level. Neither a recorded mental stop nor an undefined one provides server-side protection.

The correct class is no setup, no trade. The trader instead invents “a loose B”, which does not meet the playbook definition.

Execution, management and exit

11:33 PL: fill. EUR/JPY moves sideways for 80 minutes.
12:53: price is 18 pips below entry; the trader manually closes rather than waiting for a level that exists only in a note.
Using the source's PLN 3.82 pip-value input: 18 × 3.82 = PLN 68.76, rounded to −PLN 69.
R-multiple = −68.76 / 95.50 = −0.72R. The denominator remains the pre-entry recorded stop distance times the position's fixed pip-value assumption. Higher conversion precision changes both monetary amounts proportionately, leaving −0.72R unchanged.
Holding time: 80 minutes.

Review

The process is seriously deficient despite the modest loss:

No setup: there is no tested market hypothesis supporting an edge. An individual outcome cannot validate one after the fact.
No complete plan: neither an intended profit target nor a condition confirming the thesis was defined.
Reinforcing the wrong habit: a chance profit could teach the trader to repeat unplanned entries and contaminate later behavior.
Time cost: 80 minutes of monitoring consumes attention and energy before the afternoon New York opportunity window.
Missed-opportunity cost: an A+ on another pair could be overlooked while attention is occupied by EUR/JPY.

Error category: boredom trade / opportunity hunger / random execution. These are working labels for the same mechanism: low tolerance for inactivity encourages rationalizations for action.

Corrective playbook proposals

A minimum interval after a profitable trade as an anti-overconfidence control. The source illustrates 90 minutes; the number must be specified and assessed in your own plan, not treated as a market standard.
No playbook signal means no trade. Deliberate non-entry is a complete decision, not a failed trading day.
During the source's 11:30–14:30 Polish-time quiet window, consider only signals generated by the defined playbook scanner rather than manually hunting charts. This is a personal schedule filter, not a universal London/New York market closure; all mandatory checks still apply.
The thought “I must do something” triggers 15 minutes away from the monitor, after securing any existing exposure, rather than a new position.

Illustrative example — synthetic observations, not an account record: an employee copies a US trader's workflow. A person working 09:00–17:00 reads about a US day trader taking four to six New York-session trades daily and tries to replicate that count during London and pre-New York hours, scanning between meetings and emails. Months later the account is down and C-trades dominate the journal. The mechanism is an attention mismatch: the full-time trader can monitor most of the session; the employee has perhaps 30–90 minutes of focused availability. That smaller window may offer only one suitable setup, or none, and does not justify forcing four entries. Match the workflow to available attention, not an admired online trade count. 15.5 develops schedules by trader profile.

Case 4 · R0 recorded, SL not submitted — PLN 3.82/pip is a fixed teaching assumption. A missing SL does not erase the record.

Open the diagram at full size

6. Case 5 — Winner with a behavioral error: +0.18R versus a conditional +1.6R plan scenario

The fifth trade closes profitably, but discretionary departures from the management plan limit the captured move. Reviewing winners matters as much as reviewing losers: repeatedly truncating large winners can damage a trend-following payoff distribution that relies on them to offset losing streaks. That mechanism does not mean every early profitable exit is wrong; compare it with the actual tested exit rule.

Illustrative composite, not a historical result or account record. The USD/JPY path is synthetic. Equity continues from Case 4's displayed sequence. PL means Polish local time. The 09:18 decision uses an already completed M5 bar; it does not imply a standard M5 candle closes at 09:18.

Market context

USD/JPY's 08:00–09:00 PL pre-London range is 148.82–149.15, or 33 pips, within the required 12–35. Although near the upper boundary, the condition is binary and this setup card defines no borderline range case. Proximity alone does not downgrade A+ to B; only a quality criterion with a permitted borderline case written before the result can do that. Daily ATR(14) is 95 pips and the calendar is clear within Section 4's event window.

Setup: A+

At 09:18 PL, the latest completed M5 breakout bar closes at 149.28, 13 pips above the 149.15 range high. Tick volume is 1.5 times the reference average and body size is 0.7 × M15 ATR(20): A+. The one-percent budget on displayed equity of PLN 27,836 is approximately PLN 278. SL = 148.82 minus five pips = 148.77, 51 pips from planned entry 149.28. One standard lot has JPY 1,000 pip value; using JPY/PLN = 3.80/149 gives about PLN 25.50 per lot or PLN 2.55 per 0.10. PLN 278 / (51 × 2.55 × 10) ≈ 0.2138 lots, rounded down under the scenario's conservative sizing choice to 0.20. The source fixes the position pip-value input at PLN 5.09, giving R0 = 51 × 5.09 = PLN 259.59 ≈ PLN 260. The exact conversion is about PLN 5.1007; retain the chosen teaching input consistently instead of silently changing the R denominator.

Execution within the scenario's tolerance

09:18:39 PL: market buy 0.20 lots USD/JPY, with a modeled 2.1-pip spread. Fill 149.288 is 0.8 pip worse than the planned 149.280 reference. This is entry shortfall against the plan, not necessarily all quote-to-fill slippage; the displayed ask would be needed to separate those components. The scenario accepts this execution.

In-trade: the initial move

09:18:39, entry: 149.288.
09:42 PL: price 149.68, favorable excursion 39.2 pips, approximately +0.77R. The trader thinks it is going well.
09:51: price 149.79, favorable excursion 50.2 pips, approximately +0.98R, near TP1.

The mistake: mood overrides the plan

09:53 PL: price 149.82, MFE 53.2 pips or approximately +1.04R. The plan calls for closing half at the one-R distance, fill plus 51 pips = 149.798, approximately 149.80.

Instead, the trader decides to wait for 149.90 and close everything there rather than half at the planned level. The change is driven by the favorable move, not an authorized rule.

09:54: price 149.79, down three pips: “Just a normal pullback.”
09:57: 149.72, down ten pips: “It will come back.”
10:01: 149.64, current profit about 35 pips; MFE remains 53.2 pips because a running maximum does not shrink. “I should have closed at +50.”
10:04: 149.52, current profit about 23 pips. The trader freezes, reluctant to close below the previously seen gain.
10:18: 149.38, approximately nine pips up. A manual close fills at 149.378. Realized gain is 9.0 pips × PLN 5.09 = PLN 45.81, rounded to PLN 46. Full-position R = 45.81 / 259.59 ≈ +0.18R.

What happened after the exit

10:23: 149.42, a retest near the exit area.
10:48: 149.73, back toward the high seen while the trade was open.
11:24: 150.12, above that high; the closed trade's MFE remains 53.2 pips.
13:08: 150.67, the later path's peak, 138.2 pips or approximately +2.71R from entry. This is not the trade's MFE. MFE is measured only between entry and actual exit, where it was 53.2 pips or +1.04R.

Under an additional hypothetical path assumption, closing half at +1R contributes +0.50R and trailing the other half to around 150.40 contributes about half of 111 pips / 51, or +1.09R. The combined counterfactual is approximately +1.6R for the original position.

This calculation adds an assumption; it does not reconstruct a proven executable plan result. The complete M15 SMA-20 and stop path from 10:18 to 13:08 is unavailable, so we cannot establish that the remaining half would survive the pullback to 149.38 and exit only after the later high. The +1.6R estimate depends on that survival and exit assumption. The later +2.71R peak is potential movement, not profit attributable to the trader; a trailing exit normally gives back some peak excursion.

Actual result: +0.18R. The conditional shortfall against the modeled path is approximately 1.4R of unrealized hypothetical profit, about 1.59R minus 0.18R. It is not an additional booked loss or a demonstrated amount the real plan would necessarily have earned.

Review

Error classification: plan modification under euphoria and premature discretionary closure under fear. Two separate decisions:

09:53, euphoric modification: move the intended target from the one-R partial exit toward 149.90, approximately 61.2 pips or 1.20R, and change the exit allocation without a plan-based reason. Being profitable does not make an unauthorized change valid.
10:18, fear-based closure: exit because the open gain has shrunk instead of following the defined structure/trailing rule. The trader mentally treats a decline from +1R toward +0.2R as a loss despite positive realized P&L.

Repair: use a tested preplaced partial-TP mechanism at entry rather than relying on a manual action at 09:53. Implement the trailing rule through supported automation or an alert tied to its completed-bar condition and a predetermined action. Verify whether automation runs on the broker server or requires an active client terminal. The rule should govern execution rather than the trader's mood.

Corrective playbook proposals

Implement TP1 as the supported, tested limit/partial-close arrangement at entry rather than a discretionary market exit. Verify account netting/hedging behavior so an opposite order does not accidentally open a new position.
Write the reason before any permitted in-trade plan change. “It is going well” is not a sufficient reason to override the rule.
The source proposes closing half after a give-back exceeding 50% of in-trade MFE and trailing the rest. This changes the exit rule and requires a versioned test; it is not an immediate repair justified by this one case.
Track in-trade MFE minus realized result as a diagnostic give-back measure. A reduction can be consistent with better execution but may also reflect smaller moves or a changed setup mix. Interpret it alongside the R distribution and uncertainty.

Case 5 · MFE stops at closure — Fixed PLN 5.09/pip; a hypothetical outcome is not a guaranteed missed profit.

Open the diagram at full size

Illustrative example — synthetic observations, not an account record: the missing captured-profit measures. Alongside R and win rate, record maximum favorable excursion during the actual holding period and realized R. The article calls their ratio capture rate, a working journal metric rather than an industry standard. For Case 5, its denominator is in-trade MFE of about +1.04R, so 0.18 / 1.04 ≈ 0.17. Using the later +2.71R peak would measure a hypothetical continued holding period rather than actual execution. No universal improvement after adding TP1/trailing automation is claimed; your own sample must establish that. These fields help separate the market hypothesis from execution, but a low capture rate alone does not identify the cause. Compare it with the same exit rule's appropriate benchmark.

7. Six stages: pre-trade, entry, in-trade, modification, exit and post-trade

The five cases support a general six-stage model. Each stage has distinct tasks, metrics and traps.

Open the diagram at full size

Stage 1: pre-trade — illustrative 3–10 minutes, or 1–3 with a prepared employee workflow

Scan, identify the setup, classify A+/B/C, decide to enter or pass, calculate size, establish SL/TP, check calendar and correlations, and complete the mental-readiness check. Follow the predefined algorithm: all necessary conditions pass, then execution is permitted; a missing mandatory condition means no trade. Intuition does not replace a required check, and the illustrative durations do not authorize rushing an incomplete checklist.

The number of setups you can assess depends on available attention: full-session access differs from 30–90 minutes alongside employment. 15.5 covers profile-specific scan windows and weekly schedules. This article's short-attention workflow uses preconfigured scanner alerts rather than open-ended manual chart hunting, while retaining every required confirmation.

Pending orders when you cannot watch the market

If you will be away during the setup window, make and record the decision beforehand. A supported pending-entry arrangement can include server-side SL/TP and an expiry so that an unfilled instruction does not remain active indefinitely. Verify whether the broker provides native OCO and what happens to unfilled or triggered orders. Without native OCO, two independent orders require cancellation of the other after the first fills. Confirm that protective levels accompanying entry are actually accepted on the server and clarify how a partial TP is implemented; availability and account mode matter. Confirmed server-side orders do not depend on your local power or connection, but still depend on broker operation and executable liquidity.

For a pending-order London-breakout variant, place Buy Stop above the completed 08:00–09:00 PL range and Sell Stop below it, with supported protection, partial-exit arrangement and expiry. Unfilled orders expire as specified. This does not reproduce the completed-candle setup automatically: a stop entry activates at its price trigger, bypassing body ≥0.3 × M15 ATR(20) and tick volume ≥1.2 times the reference average assessed after candle completion. An intrabar false break can therefore enter. Without native OCO, the opposite pending order remains live until cancelled, requiring working supervision. Treat this as a separate setup variant with its own card, trigger, confirmation rules and test, not the same strategy remotely executed.

Illustrative example — synthetic observations, not an account record: a whipsaw activates both intended OCO legs. A trader leaves Buy Stop and Sell Stop around a breakout shortly before the Employment Situation release, illustrated at 14:30 Polish time, and walks away. Price spikes several dozen pips upward and immediately reverses; both entries activate and both stops suffer poor fills as spreads widen. This example requires cancellation that is absent, delayed or not guaranteed atomically; it must not be presented as the behavior of a verified atomic server-side OCO implementation. The trader pays for two losing legs instead of capturing one directional move. Pending orders do not protect against disappearing liquidity. Cancel them before the applicable news-exclusion window and reassess after stabilization according to Section 4.

Stage 2: entry — illustrative 5–60 seconds

Select market, limit or stop execution; inspect spread and liquidity; submit; verify the fill and actual entry price. If slippage exceeds the predefined tolerance, illustrated as one pip on a liquid pair, record the exception and assess execution viability under the plan. One fill alone does not establish that the strategy has lost its edge.

Spread and short positions

A short FX position normally closes against Ask while a default MetaTrader FX chart displays Bid. Its stop can therefore trigger without the Bid candle reaching the same numerical level. When designing a short stop from Bid-chart structure, account for the relevant Ask/Bid difference in the pre-entry distance and sizing; do not blindly widen an existing stop after entry. Enable the Ask line where available, using the source's MT4 route through chart Properties → Common → Show Ask line, and verify the equivalent control in your platform version. Check the actual order specification and execution record rather than concluding from a Bid candle alone that the stop triggered incorrectly.

Submit protection, not merely a mental stop. Prefer a supported entry with the predefined server-side SL; where a separate post-fill submission is required, verify protection immediately and follow the emergency procedure if it fails. Confirm the accepted order parameter in the position/order record, not just a chart line that may be local. Separate planned risk, R0 fixed before entry, from protection, the broker-confirmed stop. Without recorded R0, the trade cannot enter defined-R statistics, as in Case 3, but still belongs in currency, equity-percentage and error records. With R0 but no submitted protection, as in Case 4, the denominator exists without the safeguard. Supported bracket/OCO arrangements should be tested for the actual account rather than assumed from a product label.

Execution quality: measure rather than assume

Market-making, ECN/STP and internalization arrangements affect execution and costs. Record entry and exit slippage instead of presuming intent from the model name. Persistent adverse asymmetry, especially around stops or releases, is an operational issue to investigate with the broker and potentially address through account type, execution method or provider changes. Demo is not evidence of live execution quality. Its treatment of slippage, freeze levels, requotes and event spreads may differ materially; demo metrics describe that simulation, not the fills a funded account will receive.

If platform access fails, an already submitted server-side stop remains independent of your local power and internet. Distinguish it from client-terminal trailing: when the terminal stops, the last submitted stop remains but further client-calculated updates stop. 15.1 covers backup power, connection, devices and configuration copies. 15.8 supplies the incident procedure once prevention fails, including broker contact and confirmation of current state before any further trade instruction.

Stage 3: in-trade — five minutes to several hours

Monitor without negotiating the rules. Check whether predefined management triggers have been reached. The source illustrates one chart check every 10–20 minutes between alerts; the appropriate cadence depends on the tested strategy and control setup. Do not modify SL/TP unless a defined trigger calls for break-even, partial TP or trailing. The ordinary exit categories here are TP1, TP2, trailing, SL and time; a prewritten emergency exit remains legitimate. Other discretionary exits such as “it looks weak” are interventions to record as departures from this plan.

Stage 4: modification at predefined triggers

The example plan permits three management changes:

Move to break-even: after +1R, move the SL to entry. This is price break-even; commissions, financing and slippage can still make realized cash P&L negative.
Partial TP: close the planned fraction, illustrated as 50%, at its specified level.
Trailing: define condition, observation and action. For the long examples, activation occurs when completed-bar M15 SMA-20 exceeds entry plus 1R. Set SL to SMA-20 minus five pips, ratcheting only upward and only after M15 closes. Short variants require their own mirrored specification. A vague intention to trail is not an executable rule.

Other changes, such as tightening from fear, adding size because the trade is winning or moving TP farther for more profit, are breaches unless explicitly authorized beforehand. Record them as such.

Stage 5: exit — illustrative 5–30 seconds

Close through the planned TP, SL, trailing, time or emergency route. Record time, fill, slippage and defined R result. Cancel orphaned pending orders, including the untriggered opposing entry where applicable, then reconcile state: compare platform history with the broker report for every tranche's time, price, size, commission and swap. Investigate discrepancies rather than automatically dismissing them as rounding. Document an emergency exit and its circumstances promptly. 15.9 develops whole-journal reconciliation.

Stage 6: post-trade — illustrative 10–25 minutes after closure

Take a short mental break, illustrated as 10–20 minutes, before reflective review rather than grading the trade 30 seconds after the fill. Complete the journal template, classify errors, update consistently defined capture rate and identify reproducible or avoidable behaviors. Apply any existing anti-tilt or anti-overconfidence sizing rule to the next eligible setup; do not invent a new size merely because the last trade won or lost.

8. What to journal after each stage

The journal is a diagnostic tool. Its purpose is to answer where money is being lost with estimates and uncertainty, rather than a bare verdict. “Good trade” or “bad trade” provides no useful aggregation fields. Each phase has required numerical or categorical entries. Subjective one-to-ten ratings may supplement psychological notes, but they are starting observations rather than a performance metric to optimize.

15.9 covers journal applications, spreadsheets and Notion databases, their dated features/prices and the minimum overall schema. Here the focus is which fields each stage of one trade completes so that later aggregation is possible.

Stage 1: pre-trade record, before entry

Illustrative values, not historical results: the examples below come from synthetic Case 1.

Field	Format	Example
Setup identification date/time	YYYY-MM-DD HH:MM:SS with zone/offset	YYYY-MM-DD 09:14:08, Europe/Warsaw
Pair	Text	EUR/USD
Strategy/setup and version	Playbook name plus version ID	London Open Breakout v1.2
Classification	A+ / B / C	A+
Conditions passed	Individual checklist entries	Range, ATR, calendar, correlations and time of day pass; retain body and volume checks too.
Planned entry: R0 reference	Price	1.0863, the completed breakout-bar reference.
Planned SL	Price and risk distance	1.0827; 36 pips = 1 risk-distance unit.
Planned TP1/TP2	Rule, price and R	TP1 at fill plus 36 pips, which becomes 1.08998 after the fill; remaining exit follows M15 SMA-20 trailing.
Size	Lots and equity-risk percentage	0.20 lots; 0.96% planned equity risk.
Readiness: binary checks	YES/NO per condition	At least 6 hours of sleep: YES; alcohol within the last 12 hours: NO; weekly drawdown magnitude exceeds 2R: NO; daily-loss stop already triggered: NO. These are example plan filters, not universal medical thresholds.
Optional subjective readiness note	1–10 and 1 sentence	Supplementary context, not the admission metric.

Stage 2: entry record, after the fill

Click timestamp, with milliseconds if needed for latency.
Confirmation timestamp in the same clock/zone; the difference is latency.
Actual entry fill price.
Actual broker-confirmed stop price.
Spread at entry.
Entry shortfall against the plan, separately from slippage against the executable quote where recorded.
Order type: market, limit or stop.
No fill or rejection: record the platform's stated reason, such as requote, freeze-level restriction, rejected stop or GTD expiry, with time and the plan-based decision to retry or abandon. A rejected order is a valid journal event. Without it, opportunity count is understated and execution quality cannot be assessed properly.

Stage 3: in-trade record

Chart-check count, collected in the source's illustrative 15-minute intervals.
Running MFE: largest favorable excursion in pips.
Running MAE magnitude: largest adverse excursion in pips.
Thoughts without actions, such as wanting to move the stop, add size or close early.

Stage 4: record every modification

Modification timestamp.
Type: break-even, TP1, trailing or other.
Within the plan: YES/NO.
If NO, explanation and error category.

Stage 5: exit record

Close timestamp.
Each tranche's exit price separately.
Exit reason: TP, SL, trailing, time or emergency.
Exit slippage against the appropriate reference.
Total P&L in consistently volume-weighted pips, net R where defined and PLN or the identified account currency.
Entry-to-exit holding time.
Orphaned pending orders removed: YES/NO.
Reconciliation: each tranche compared with the broker report for time, price, volume, commission and swap; record agreement or the discrepancy and its amount.

Stage 6: post-trade record, approximately 10–25 minutes later

Plan matches playbook: YES/NO.
Entry shortfall/slippage within tolerance and server-side stop confirmed: YES/NO plus values.
Modifications plan-compliant: YES/NO plus deviations.
Behavioral-discipline breach: YES/NO plus description.
In-trade MFE and MAE in R, realized R and consistently defined capture rate.
Error from the fixed category list: none, FOMO, boredom, plan modification, missing SL, sizing error or other.
Operational conclusion in no more than three sentences, naming concrete actions.
Proposed playbook change: none; immediate operational repair; or setup-structure change entered as a candidate new version.

Section 10's mini-template fills these fields with Case 1 values; it is not repeated here.

Over 30 fields may look demanding, but the source budgets approximately five to eight minutes per trade once the workflow is familiar. That is the cost of creating usable records, not a measured completion-time guarantee. Two hundred “good trade” comments still cannot answer a diagnostic question without sortable fields.

MAE and MFE for this one trade

MAE, maximum adverse excursion, and MFE, maximum favorable excursion, measure extreme movement from the fill only during entry-to-exit. Record pips and defined R. Their magnitudes are running maxima and never shrink; a signed negative MAE becomes more negative as its adverse magnitude grows. Later price action does not change either, as Case 5 demonstrates. Case 1 has MAE −5.4 pips or −0.15R and MFE +74.2 pips or +2.06R. Use recorded price data or a tool that actually calculates excursions; an ordinary order-history tab need not contain the intratrade path. These fields support later stop/target diagnosis. If a position closes at its original stop, its fill-relative adverse excursion includes the original stop distance adjusted for entry shortfall and exit slippage; it is approximately at least 1R under these cases' adverse-fill assumptions, not a universal identity if entry improves or risk references differ. That statement no longer applies after a stop moves to break-even or trails into profit. A TP exit establishes at least the favorable excursion to its actual exit level; favorable execution or a gap may exceed the nominal target. 15.9 covers definitions, whole-journal excursion plots, diagnostic tables and automated tools.

What to retain in one trade record — If R0 is unknown, retain cash PnL and the error; leave R undefined.

Open the diagram at full size

9. Common errors at each stage

A phase-based error map changes the question from “Why did I lose?” to “Did I breach the process, and at which stage?” Pre-trade and modification are important places to investigate. The source supplies no population sample to establish that they account for most retail losses, and a correctly executed losing trade may have no breach at all.

Stage	Error pattern	Journal evidence	Corrective action
1. Pre-trade	Inflated class, B labeled A+; no recorded stop; skipped event window; mental size calculation.	A+ without complete checks; empty R0 field.	Required checklist form; no valid setup without its record; define the stop before entry; block entries under Section 4's event rules.
2. Entry	Market order against stale quotes; 1.0 instead of 0.1 lots; wrong instrument; accepting a requote without checking.	Slippage beyond tolerance or size unlike the calculation.	Tested position-sizing/order tool with supported SL submission; verify the ticket before confirmation.
3. In-trade	Checking more often than required; consuming news/opinions while holding; managing the position alongside competing work.	Increasing chart checks followed by more modifications.	Alerts at defined management triggers instead of unnecessary continuous intervention.
4. Modification	Tighten SL from fear, widen it hoping for recovery, extend TP, add size or close before an authorized trigger.	Off-plan modifications; low capture rate; losses both below and above 1R.	Only the plan's break-even, partial TP and trailing changes; record other actions as deviations unless explicitly preauthorized.
5. Exit	Unclosed tranches, orphaned pending orders, missing fill record or an unplanned overnight hold.	Insufficient data for R; discrepancy against the broker report.	Complete intended closures, remove orphaned orders and reconcile the account state.
6. Post-trade	No review; grading 30 seconds after exit while emotional; rewriting the setup after one result.	Empty conclusions and unversioned setup variants.	Reflect after a 10–20-minute break; register structural changes as candidate versions rather than automatic next-day rules.

Use the map to locate a possible breach rather than assuming that the loss itself proves one. 15.6 owns pre-entry filters; 15.8 covers wrong size, wrong instrument and liquidity incidents; 15.10 covers learning errors and setup decisions. One exit-specific example belongs here because it directly changes realized R:

Illustrative example — synthetic observations, not an account record: rollover defeats the planned tight-stop loss. A trader holds thinly traded pairs overnight with tight stops and wakes to a loss much greater than the intended −1R. Around the broker's trading-day rollover, commonly 17:00 New York and usually 23:00 Polish time outside daylight-saving mismatch weeks, spreads widen and the stop executes into thin liquidity with substantial slippage. Verify the actual broker/instrument schedule. There is no universal spread-expansion number here: measure your account's quotes over several nights before relying on overnight execution assumptions. A tight stop still exists as an order but cannot guarantee a tight cash loss. Decide beforehand whether to close before rollover, use a tested wider-distance/lower-volume variant within the same account-risk budget, or explicitly accept the modeled stop-market execution risk. Widening an existing stop at unchanged size simply increases exposure; it must not retrospectively redefine that trade's original R0. Crosses such as AUD/NZD and NZD/CAD and exotics such as EUR/TRY and USD/ZAR warrant particular execution checks.

Trap: overfitting the playbook to your own journal

Repeatedly tailoring rules to historical journal patterns can produce an apparently improving backtest without better new trades. Warning signs include a new filter after every losing streak, fewer signals without improved outcomes, and rules justified only by “it worked in my data”. At the individual-trade level, record every new rule as a version and assess its subsequent execution separately rather than crediting it with the data used to invent it. Broader validation on genuinely held-out data, the fitting mechanism and the trial register belong in 15.4.

10. Trade-analysis template for the journal

Section 8 provides the full phase-by-phase fields. Here are the sizing calculation and a compact one-trade record to copy.

Position-sizing formula
lots = risk_PLN / (stop_distance_pips × pip_value_per_0.1_lot_PLN × 10)

Case 1: EUR/USD, PLN 28,400 equity and one-percent risk budget.
risk_PLN = 28,400 × 0.01 = PLN 284.
Stop distance = 36 pips, from planned entry 1.0863 to SL 1.0827.
Pip value per 0.10 lots = PLN 3.80 under the article's FX assumption.
lots = 284 / (36 × 3.80 × 10) ≈ 0.208.
Round down to 0.20 lots at the applicable supported step, never above the risk budget. R0 = 36 × PLN 7.60 = PLN 273.60, approximately 0.96% of equity. That denominator remains fixed for this trade.

Operational rule: use a checked spreadsheet or position calculator instead of mental arithmetic under time pressure. No completed calculation means no entry.

Repeatedly transcribing pips and lots under a moving market creates fat-finger opportunities. The Polish source gives EarnForex's Position Size Calculator as an example of a tool to investigate, not a newly verified recommendation or guarantee of current features. Check the actual platform/version, code and permissions. Some tools only calculate size; a version with a trading module may also submit an order and SL/TP. Verify and test which behavior you have before relying on one click to replace several manual fields.

Mini-template: one journal record

Copy the fields into Excel, Google Sheets, Notion or a journal application. One record represents one trade, with tranche details retained where needed. Complete it the same day, using mostly binary or numerical fields. The source budgets five to eight minutes once practiced and 15–25 minutes during the first 30–50 trades while the habit forms; these are workflow assumptions rather than measured requirements.

Illustrative values, not historical results: the example column uses synthetic Case 1.

Field	Format	Case 1 example
ID	Text	T-YYYY-067
Identification date/time	YYYY-MM-DD HH:MM with zone	YYYY-MM-DD 09:14, Europe/Warsaw
Pair	Text	EUR/USD
Setup and version	Name and version ID	London Open Breakout v1.2
Class	A+ / B / C	A+
Plan-compliant at entry?	YES / NO	YES
Risk percentage, lots and R0	%, lots, currency amount	0.96%; 0.20 lots; R0 PLN 273.60
Entry planned / actual / shortfall	Price / price / pips	1.0863 / 1.08638 / +0.8 pip, including 0.3 pip versus displayed ask 1.08635.
Server-side SL confirmed?	YES / NO	YES
Modifications and compliance	List plus YES/NO	Trailing 15 minutes late: NO; manual TP1 instead of planned limit: NO.
MAE	R	−0.15R
MFE	R	+2.06R
Realized result	R with cost basis	+1.21R from fills on the no-commission intraday example; separate commission sensitivity +1.19R.
Capture rate: article's working measure	Realized R / in-trade MFE R	0.59
Holding time	Minutes	227 at minute-level timestamps, 09:14 to 13:01; the second-level record gives 226 minutes 29 seconds.
Error category	Fixed list	Delayed trailing; manual TP1 deviation.
Corrective action	At most 1 sentence	Sound alert for completed SMA-20 on M15 activation, plus a tested limit TP1 instead of a manual market exit.
Playbook setup update?	YES / NO / CANDIDATE	NO structural change; repair the operational implementation.

This compact set becomes sortable across records: class, version, compliance, R0, execution shortfall, MAE/MFE and error category. An unfilled column cannot inform later diagnosis. Section 8 gives the full 30-plus-field scheme. The source suggests building toward it over 50–100 trades as journaling becomes routine, but essential pre-entry risk and actual-fill records must be captured from the beginning; those counts are habit-building examples, not permission to omit critical evidence.

Calculating results for these five trades

15.9 centralizes expectancy, profit factor, win rate, drawdown, SQN, Sortino, execution drag and their interpretation. Here we calculate only what follows from these five teaching records.

Step 1: identify which trades have defined R
Case 3 has no R because no stop level was recorded before entry. Do not invent R0 afterward or place it in an R average. Its result remains approximately −PLN 538, or −1.9% of entry equity. Four trades retain the same denominator convention:
Case 1 +1.21R; Case 2 −1.06R; Case 4 −0.72R; Case 5 +0.18R. Their displayed rounded R values sum to −0.39R.

Step 2: mean R for those four
Two winners out of four; average displayed winner (1.21 + 0.18)/2 = 0.695R, about +0.70R; average loss magnitude (1.06 + 0.72)/2 = 0.89R.
E = 0.50 × 0.695 − 0.50 × 0.89 = −0.0975R, about −0.10R.
Profit factor = 1.39 / 1.78 ≈ 0.78.

Step 3: the selected plan/setup-compliant mean, E_adj
Select Cases 1 and 2 because setup classification and entry plan match the playbook. The label matters: Case 1 still has delayed trailing and a manual-TP deviation, so this is not a subset with flawless execution.
E_adj = (1.21 − 1.06)/2 = 0.075R, approximately +0.08R.
This does not establish an edge. Two observations provide extremely weak precision under ordinary sampling assumptions. The value is demonstrating the selection and calculation. The difference between this mean and the four-trade mean compares different subsets; it does not isolate execution's causal effect. That requires an appropriately matched comparison using the same rule and sample with uncertainty, rather than attributing every difference to operator behavior. Use your own data and intervals under 15.9, then apply 15.10.

Step 4: capture rate, this article's working measure
CR = realized R / in-trade MFE in R. Use a consistent definition within the same journal rather than a universal “good” threshold. Case 1: 1.21 / 2.06 ≈ 0.59. Case 5: 0.18 / 1.04 ≈ 0.17. The later post-exit +2.71R path does not belong in Case 5's denominator. If MFE is zero, the ratio is undefined; do not manufacture a value.

Costs beyond a bare entry-to-exit R-multiple

A theoretical gross edge must survive execution and account costs. Keep the following four categories separate:

Spread: already embedded in actual fills when buying at Ask and selling at Bid. Do not subtract it again from fill-based P&L. Its behavior varies by broker, instrument and event/overnight conditions; use your account's evidence.
Commission: account-specific charges commonly stated per lot or per side/round trip. Execution-model labels alone do not determine the fee schedule. These intraday examples have no rollover, so the separate commission sensitivity is the only extra trading charge modeled. The assumption is USD 6.50 per standard-lot round trip, not an industry norm. Other account fees require separate treatment.
Swap/financing: holding across the trading-day rollover can produce a debit or credit based on rates and broker terms. The source discusses the common Wednesday triple-swap convention for some T+2 FX instruments, but the actual day and holiday treatment are instrument-specific. Check the current contract specification before holding overnight; no fixed financing number is assumed here.
Tax: excluded from this article's trade-R metric. The journal supplies dates, amounts and conversion evidence for reporting, but is not itself a tax computation. The Polish source's PIT-38 context is retained as Polish jurisdictional material; Section 14 covers tax/recordkeeping and international scope. Apply the actual account holder's jurisdiction rather than replacing Polish rules with a supposed universal equivalent.

Commission sensitivity for Case 1: EUR/USD, PLN 28,400 equity and 0.20 lots. Keep the recorded fills unchanged to isolate this extra cost; this is not proof that a different account type would offer identical fills.

Spread is already in the fills. The 0.3-pip entry slippage versus Ask, 0.3 × PLN 7.60 = PLN 2.28, is also already reflected in entry 1.08638; do not deduct it twice.
Commission = USD 6.50 per lot round trip × 0.20 = USD 1.30; at 3.80, PLN 4.94.
Additional drag = 4.94 / 273.60 ≈ 0.018R, reducing the displayed +1.21R to approximately +1.19R.
No swap is assumed because this case closes before the financing rollover, not merely because entry and exit share a calendar date.

The article's baseline case R values use actual fills, including spread, before separate commission, financing and tax. For the no-commission/no-rollover teaching accounts, there is no further modeled trading charge. Where “net” is used for the ECN sensitivity, it means after the stated commission, not after every possible account cost or tax. Label that basis explicitly in the journal so comparisons remain meaningful.

Operational implication: calculate your actual typical-volume cost and compare it with expectancy on a matching basis. A theoretical-price backtest that omits real execution costs can overstate results, particularly with short holding periods and narrow targets. A backtest already using executable bid/ask fills must not subtract spread again. 15.4 develops cost-aware testing.

Illustrative example — synthetic observations, not an account record: M1 scalping that mainly pays fees. A trader takes a dozen to several dozen M1 trades daily between meetings and celebrates a high win rate. Once costs are reconciled, frequent spread and commission charges consume the gross advantage. Positive gross expectancy can become negative net expectancy; win rate alone cannot settle the result. More trades repeat both the edge and its costs, rather than making a weak per-trade margin adequate by volume alone.

What to do with the numbers next

15.9 groups overall/per-setup win rate, mean R, capture rate, execution-error share and slippage with illustrative reference/alarm thresholds. Those thresholds are scenario assumptions, not market standards. The system-versus-operator decision belongs in the audit: first reliable data and execution, broker reconciliation, playbook version, rule breaches and costs in original R0; then market regime; finally evidence about the edge. 15.10 gives the symptom, diagnosis and action sequence.

Trade anatomy supplies evidence that cannot reliably be recreated later: what was known when the decision was made. Trades with rule breaches or unreconciled costs cannot cleanly validate or invalidate the strategy's intended performance.

Open the diagram at full size

The vertical axis is PLN, not R, because Case 3 lacks R0. Including it on an R curve would require inventing a denominator after the loss. From the stated unrounded teaching amounts: +331.36 − 288.80 − 538.08 − 68.76 + 45.81 = −PLN 518.47. Summing whole-PLN displayed outcomes gives −PLN 519, a rounding difference. Exact ending equity is PLN 27,881.53 from PLN 28,400. The four defined-R displayed results sum to −0.39R. The entry-plan-compliant Cases 1 and 2 contribute +0.15R, although Case 1 has management defects. Cases 3–5 total −PLN 561.03, approximately −PLN 561; only Cases 4 and 5 have defined R, summing to −0.54R. Case 3's PLN 538.08 loss must not be silently assigned zero R in the actual journal.

Counterfactual: a five-case period without the highlighted errors

This is a separate analytical exercise, not an alternative account history. Record it in a hypothetical plan-followed field rather than realized P&L. The assumptions are:

Case 3, FOMO: the calendar fails, class C, so no entry and zero currency cash flow. Zero is the hypothetical no-trade contribution, not an R value for the actual unprotected trade.
Case 4, boredom: no playbook setup, so no entry and zero contribution.
Case 5: half at +1R plus trailing the other half gives the assumed +1.59R, or approximately PLN 412.75 using R0 PLN 259.59. This requires the unverified assumption that the remaining half survives the pullback to 149.38; the necessary SMA/stop path is unavailable.

The hypothetical currency total is +331.36 − 288.80 + 0 + 0 + 412.75 = +PLN 455.31, versus actual modeled −PLN 518.47: approximately PLN 974 difference, precisely PLN 973.78 with those amounts. Case 3's avoided PLN 538.08 is the largest component. That does not prove the missing stop alone caused the whole gap: avoiding the invalid entry also removes its direction and size decisions. Cases 1–2 are held unchanged, including Case 1's small management defects, so the title describes the highlighted-error exercise rather than literally perfect execution. Five synthetic trades show a method, not annual improvement or a representative sample. Estimate real differences on your own documented comparisons with uncertainty.

Operational implication: removing FOMO, boredom entries, emotional TP changes and missing server-side stops may matter more than adding another setup. The size of any benefit requires your own journal and an appropriately defined comparison. Differences between E and E_adj need sample definitions and intervals and do not by themselves identify causality. No universal expectancy, improvement rate or speed of edge deterioration is asserted.

From one trade to a distribution

Five reconstructed trades are not a meaningful estimate of a strategy distribution. With more consistently recorded observations, the same fields support an R histogram, showing the mode and right tail, and an MAE/MFE plot, showing stop clusters and how far winners retrace before recovery. 15.9 contains the plots and diagnostic guidance. A few dozen records can support exploration, not automatically reliable inference.

The prerequisite is a complete record behind every point: setup version, class, pre-entry R0, execution shortfall, excursions, error category and broker reconciliation. A trade without pre-entry R0 cannot become a valid R-coordinate merely because the chart looks tidy. A recorded-but-unsubmitted stop can define R, as in Case 4, but its missing protection must remain flagged; exclude neither its currency loss nor its error from the account record.

FAQ — Frequently asked questions

Do I need this much detail if I trade only once or twice a day?

Yes. With fewer observations, losing the information behind one trade removes a larger share of your usable evidence. At 200 trades a year, gathering several hundred takes years, and even then count alone settles nothing: interval width relative to the decision threshold matters. Incomplete records can leave that period diagnostically unproductive. The source's five-to-eight-minute journaling budget is an investment in interpretable evidence, not paperwork for its own sake.

How do I distinguish FOMO from a valid setup that appeared quickly?

Check three markers: (1) classification without individually completing the required conditions, rather than merely a short decision time; Cases 1 and 2 took 23 and 18 seconds with complete checks; (2) an extended no-signal period, illustrated as three or more days, that creates pressure to act; and (3) pre-entry thoughts such as “must do something”, “must recover” or “finally something is happening”. Under this article's stricter immediate filter, a flagged entry is rejected and the reason recorded. Smaller size cannot purchase a missing mandatory condition. Section 4's proposed anti-FOMO size cap applies only to otherwise admissible future entries under a reviewed plan; it is not permission for a flagged C-trade. Compare flagged and unflagged behavior with uncertainty rather than declaring an effect after a fixed elapsed period. A few cases per quarter can leave an interval wider than the apparent difference.

What if capture rate remains below my own benchmark?

Low capture rate identifies limited capture of favorable movement during the actual holding period, not its cause. Compare with the same exit rule on comparable journal observations before attributing the gap to execution. Investigate (1) manual TP rather than the intended limit, with the source's illustrative five-to-30-second delays; (2) absent or overly aggressive trailing; and (3) fear-driven early exits. Implementing the existing rule reliably may mean a supported server-side TP1 arrangement, tested SMA-20 M15 trailing automation and per-setup monitoring. Changes to the exit rule itself require versioned testing. At the same MFE, a higher realized/MFE ratio means higher realized R, but a rising average capture rate across changing trades does not by itself prove higher expectancy or improved execution. Measure both distributions with uncertainty.

Can I use a mental stop instead of one submitted to the platform?

Not as the protective control required by this procedure. A mental level can be renegotiated when price moves against you and does not act during equipment, power or connection failure. 15.8 covers that emergency. Keep the distinction precise: a stop level recorded before entry defines R0 even if not submitted, as in Case 4, which actually lost −0.72R rather than more than 1R. Case 3 recorded no level before entry and therefore has undefined R. Both lack confirmed server-side protection. Do not erase a valid denominator or invent a missing one; submit protection and record its confirmed state. A server-side stop still does not guarantee execution price.

Should A+/B/C classification change over time?

It can evolve through controlled versions. An initial card might list five to seven conditions. Review which conditions distinguish outcomes and with what uncertainty. Finding two apparently unhelpful conditions or a new DXY filter associated with better results creates candidate changes, not automatic proof that they should be removed or added. Use scheduled, illustrated quarterly, evidence reviews and held-out testing rather than intuition or repeatedly fitting the same journal. Every definition change needs a new version and a new measurement period; otherwise comparisons mix incompatible labels. 15.10 defines the review and setup-status criteria.

Can a profitable boredom trade count as correct?

No: by definition it lacks the required playbook setup. A profit does not retroactively validate the decision or establish an edge. Record it as a process error with a favorable outcome, not a model to reproduce. The article's “lucky boredom” label keeps it out of the playbook-compliant performance subset because it was not generated by that playbook. It must still remain in the complete account journal, currency P&L and error statistics; do not hide the money or cherry-pick the account history. The same separation applies to losing off-playbook trades.

Trade anatomy — case studies from setup to exit

Reconstruct the fills before judging the decision

Weight the entry price by the executed quantity

Keep cash costs and the R denominator explicit

Attach the arithmetic to a sequence of decisions

Three exercises with worked answers

From a reconciled execution ledger to the full trade anatomy

1. What trade anatomy means and what it teaches

How stage-by-stage analysis differs from a trade review

What this article will teach you

2. Case 1 — Profitable A+ trade: London breakout, EUR/USD, +1.21R

Market context

Setup: binary A+ classification

Decision

Sizing

Execution

Trade timeline: information available at each moment

Exit and calculation

Review: practical defects inside a winner

3. Case 2 — Losing A+ trade with correct execution, −1.06R

Market context

Setup: structurally the same A+ conditions

Decision, sizing and execution

In-trade: what developed differently

Result

Review

4. Case 3 — Losing FOMO C-trade without a recorded stop: R undefined

Market context

The actual setup class is C, not A+

Decision: the process breach

Execution

In-trade: escalation

Result

Review

News protocol: the calendar filter

5. Case 4 — Random boredom C-trade, −0.72R

Market context

The supposed setup: no setup at all

Decision: emotion overrides the rules

Execution, management and exit

Review

Corrective playbook proposals

6. Case 5 — Winner with a behavioral error: +0.18R versus a conditional +1.6R plan scenario

Market context

Setup: A+

Execution within the scenario's tolerance

In-trade: the initial move

The mistake: mood overrides the plan

What happened after the exit

Review

Corrective playbook proposals

7. Six stages: pre-trade, entry, in-trade, modification, exit and post-trade

Stage 1: pre-trade — illustrative 3–10 minutes, or 1–3 with a prepared employee workflow

Pending orders when you cannot watch the market

Stage 2: entry — illustrative 5–60 seconds

Spread and short positions

Execution quality: measure rather than assume

Stage 3: in-trade — five minutes to several hours

Stage 4: modification at predefined triggers

Stage 5: exit — illustrative 5–30 seconds

Stage 6: post-trade — illustrative 10–25 minutes after closure

8. What to journal after each stage

Stage 1: pre-trade record, before entry

Stage 2: entry record, after the fill

Stage 3: in-trade record

Stage 4: record every modification

Stage 5: exit record

Stage 6: post-trade record, approximately 10–25 minutes later

MAE and MFE for this one trade

9. Common errors at each stage

Trap: overfitting the playbook to your own journal

10. Trade-analysis template for the journal

Mini-template: one journal record

Calculating results for these five trades

Costs beyond a bare entry-to-exit R-multiple

What to do with the numbers next

Counterfactual: a five-case period without the highlighted errors

From one trade to a distribution

FAQ — Frequently asked questions

Your learning progress