Operator brief · 153

Seven capture sections, and what each one is actually for.

The key idea

The principle

Capture is not record-keeping — it is data entry for a machine that will believe you.

The distinction matters because it changes how carelessness manifests. A sloppy diary produces a sloppy diary. A sloppy capture sheet produces confident, precise, wrong numbers in eleven workbooks — branch EVs computed from misrouted trades, hit probabilities built on half-remembered milestones, fee drag understated because swap was skipped. The downstream system has no mechanism for detecting that its inputs were approximate. It will compute on them and report the results with the same authority it reports everything else.

The dependency map

What each capture section is feeding.

The manual states the purpose of each section in its own terms, and reading them together makes the architecture visible. Identification routes the trade to the correct week, month, and cycle, preventing rollup errors. Classification feeds branch counts, EV probability logic, regime analysis, and trade-type diagnostics. Risk and position data feed risk dollars, fee drag, volatility and stop analysis, and capital dynamics. Outcome and milestone fields feed hit probabilities, EV, trade-management analysis, and branch performance. Execution quality separates good process from lucky outcome. Nothing on the sheet is there for completeness.

FigureCapture section → what it feeds in CP3 and downstream

Capture section	Feeds	Failure if approximate
Identification / reference	Week / month / cycle routing	Rollup errors — trade counted in the wrong period
Screenshot evidence	Later audit of the setup	The setup story gets rewritten after the outcome
Classification	Branch counts, EV logic, regime analysis	Every branch statistic inherits the error
Risk & position data	Risk $, fee drag, stop analysis, capital dynamics	Net expectancy diverges from gross unnoticed
Outcome & milestones	Hit probabilities, EV, branch performance	The probability bridge computes on fiction
Execution quality	Adherence, emotional tags, process review	Luck becomes indistinguishable from skill

Condensed from the Trade Capture Document table. The right column is why the left column cannot be filled in approximately — each of those outputs is consumed by a decision, and none of them can detect that its input was estimated.

The screenshot clause

Images exist to stop the story from being rewritten.

The screenshot fields — higher-timeframe, trigger timeframe, and scanner — are given a stated purpose that is unusually candid: they support later audit and prevent rewriting the setup story after the outcome is known. That is a control against the operator, not against the data, and it targets a genuine cognitive process rather than dishonesty. Memory reconstructs a losing trade's setup as having been weaker than it looked and a winning trade's as stronger. A timestamped image of what was actually on the screen is the only defense, and it costs three seconds at the time.

The fee fields

Commission, spread, swap, and the Wednesday triple.

The risk section asks for commission, spread, swap, and specifically for Wednesday triple-swap exposure — a level of granularity that looks excessive until it reaches the analytics. The QA rules are direct about why: audit fee and swap capture regularly, because commission and swap drag can quietly destroy compounding quality. Friction is the cost that never announces itself in any single trade and compounds relentlessly across a year, and it is invisible unless captured per trade. The Fee_Swap_Analytics tab exists to surface it by branch and timeframe, and it can only surface what the sheet recorded.

Filled when, not just filled

The immediate fields and the later fields are separated deliberately.

The usage instructions split the sheet in two. Immediately after execution: identification, classification, risk, position, and screenshots. When the trade closes: outcome R, hit flags, break-even status, fees, MAE and MFE, duration, pips, and notes. The split follows what is knowable when, but it also protects the first group from the second — recording branch and risk before the outcome exists means those fields cannot be influenced by it. The declaration made at clearance gets committed to paper while it is still a declaration rather than a prediction that turned out well or badly.

The blank-versus-zero rule

An empty field and a zero are different claims, and the workbooks read them differently.

The QA rules include a small instruction with real consequences: use blanks deliberately, and do not use zero placeholders where blank means no data unless the workbook expects zero. The two say opposite things. A blank MAE says this was not measured; a zero MAE says this trade never traded against the operator at all, which is a strong and usually false claim. Averages computed across a column where missing values were filled with zeros are dragged toward zero by exactly the trades that were never observed, and the resulting number carries no marker that anything was estimated.

Connected inside MARS

Evidence →

Compliance Panel 3 →

MAE/MFE Efficiency Lab →

Workflow →

Every brief documents the same shipped system.

The complete MARS package — eleven workbooks, three TradingView indicators, the full manual library — $497.

Get MARS Pro — $497 Explore the Platform