Operator brief · 20

Purple, blue, green: inside the 7-Tier Decision Engine.

The key idea

The tab's question

Everything upstream diagnoses. This tab decides.

By the time information reaches the 7-Tier Decision Engine, the rest of the stack has already done its work: the journal has recorded, the weekly summary has aggregated, the EV engine has scored, the structural diagnostics have classified, the gate engine has fixed the capital state. What none of those layers does is answer the operational question — what exact deployment tier should the next cycle run at, right now? The decision engine exists to convert all of that upstream evidence into one resolved answer. It is the execution translation layer: the last point where the system's knowledge becomes the system's behavior.

The three blocks

One block asks, one block computes, one block governs.

The tab's layout separates three jobs that most trading spreadsheets fatally blend. The purple block — columns A and B, Input & Selection — is the control intake layer: the current gate, drawdown state, daily EV state, weekly permission tier, cycle context, and any operator override selection enter here and only here. The blue block — columns D and E, Calculation & Value — is where resolution logic runs: the inputs are combined against the doctrine to compute a candidate tier. The green block — columns G through L, Structural Overlay — is governance: Category 6 confirmation, calculated diagnostic overlays, and the graded suppression states that can cap or veto what the blue block computed. Nothing in the green block introduces new inputs, and nothing in the purple block computes. Each block can be audited alone.

FigureThe decision engine's three-block anatomy

Block colors identify the workbook's own layout (columns A:B, D:E, G:L). The separation is the design: intake never computes, computation never governs, governance never introduces inputs.

Graded, not binary

The overlay suppresses by degrees — that's rarer than it sounds.

The green block's governance is not an on/off veto. The overlay states form a gradient: a clean read leaves the candidate tier untouched; a caution state discourages aggressive promotion; a suppression state actively restricts deployment posture; a hard veto declares expansion structurally unjustified and restricts the high tiers; multiple hard vetoes signal systemic degradation and force survival-first operation. Most risk systems, retail and institutional alike, implement exactly one veto: don't trade. Graded suppression preserves far more information — it lets the system stay in rhythm at reduced posture rather than lurching between full deployment and full stop, and it makes the decision log vastly more instructive, because the log records how suppressed each cycle was, not merely whether it ran.

The doctrine table

The bottom rows are where resolution becomes instruction.

Beneath the three blocks, the decision table — spanning the full width of the tab — collects the resolved outcome into operational doctrine: the final tier, the cycle pool percent from the active gate row, the per-trade average risk, the operator directive, and the risk posture in words. This is the only part of the tab an operator strictly needs at execution time, and its position is deliberate: you cannot read the answer without scrolling past the machinery that produced it. Every figure in the table is traceable upward — the tier to the blue block's resolution, the pool to the purple block's gate, the caps to the green block's overlay. A decision whose provenance can be walked in thirty seconds is a decision that can be audited, logged, and defended months later.

FigureOne pass through the tab — from intake to doctrine

The key idea

The blocks are a constitution, not a layout choice.

It would have been easy to build this tab as one dense formula column. The three-block separation costs screen space and buys the property that makes the whole system trustworthy: no layer can quietly do another layer's job. Inputs can't smuggle in computation. Computation can't grant itself permissions. Governance can't invent new evidence. When something looks wrong — and in any live system, something eventually looks wrong — the fault isolates to one block in minutes. That auditability, more than any individual formula, is what makes the final tier a decision the operator can stake capital on.

Connected inside MARS

Throttle Control Panel →

Seven-Tier Risk Model →

Structural Diagnostic Engine →

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