Operator brief · 376

Where the line between discretion and governance actually falls.

The key idea

The placement rule

Automate where emotion is strongest and repetition is highest.

The boundary is not drawn by difficulty or by importance — it is drawn by two properties considered together: how reliably emotion corrupts a decision, and how often the decision repeats. Sizing scores high on both, which is why it is fully mechanical. Reading whether a level is holding scores low on the first and high on the second, and stays human because no formula does it as well. The rule produces some counter-intuitive placements, and every one of them is defensible on those two axes rather than on a preference for automation. Applying the rule honestly occasionally produces placements the designer dislikes, and those are the ones worth trusting most — a boundary drawn by two stated properties survives disagreement better than one drawn by preference.

The human side

Selection, timing, and structural reading stay with the operator.

MARS makes no attempt to decide which trades to take. Whether a setup qualifies, whether the structure supports the thesis, whether this hour is the right hour — these remain human, and not as a concession. Discretionary reading of live market structure is genuinely difficult to formalise, and systems that attempt it tend to encode a narrow slice of a trader's judgement while discarding the rest. Keeping the final click human is a design decision about where formalisation adds value, not a hedge about what the software could theoretically do.

FigureWhich side of the line, and why

Decision	Emotion risk	Repetition	Owner
Trade selection	Moderate	High	Human
Entry timing	Moderate	High	Human
Position sizing	Severe	Every trade	System
Drawdown response	Severe	Rare, decisive	System
Rule changes	Severe	Occasional	System

Placement is set by two properties: how reliably emotion corrupts the decision, and how often it repeats.

The system side

Anything that determines how much capital is exposed.

Everything governing exposure is mechanical without exception: per-trade risk, tier authority, gate state, open-exposure accounting, and the response to drawdown. These share a property that selection does not — they are the decisions where being wrong is unrecoverable rather than merely costly. A poorly chosen trade costs one R. A poorly chosen size during a drawdown can cost the account, and the recovery arithmetic means it cannot simply be traded back. Irreversibility, more than anything else, is what pulls a decision across the line. Irreversibility also explains why the boundary does not move with experience: a highly skilled trader's sizing mistake during a drawdown is exactly as unrecoverable as a novice's, because the recovery arithmetic does not consult the résumé.

Why not automate everything

A fully mechanical system fails differently, not less.

The obvious question is why stop at deployment. The answer is that removing the human removes the human's advantage along with their failure modes: a fully systematic approach must specify its edge in advance and completely, and is then blind to everything it did not anticipate. Discretionary traders adapt to novel structure in ways rule sets do not. MARS is built on the position that this adaptability is real and worth keeping — and that it becomes an asset only once it is prevented from touching the sizing decision.

The asymmetry that makes it work

The human reads the market. The system reads the human.

Each party is assigned the task the other performs badly, and the second half of that arrangement is the part traders underestimate. No formula reads live structure as well as an experienced discretionary trader; equally, no trader assesses their own expectancy, adherence, or drift honestly, because the assessment is performed by the same mind that produced the behaviour. The system is not supervising the market on the trader's behalf. It is supervising the trader, which is the harder problem and the one nobody solves for themselves.

Living with the line

The boundary only holds if crossing it is visible.

A line that can be quietly stepped over is decorative. What makes this one operational is that every crossing leaves a record: an override is logged, a size outside tier is flagged, a rule change goes through change control with a written note. The trader retains the physical ability to do all of these — MARS is a workbook system, not a broker lock — so the enforcement is evidentiary rather than physical. The design bets that most discipline failures happen in the dark, and that a decision one has to write down first is a decision one mostly does not make.

Connected inside MARS

The Core Idea →

Authority Hierarchy & Governance →

Throttle Control Panel →

The Rails →

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