Reading the monthly diagnostic panel as one picture.

The reading order

Four questions, asked in sequence.

Start with load: what did the month impose? Pressure answers it, and answering first prevents every subsequent reading from being unconsciously graded against an assumption of normal conditions. Then response: how well did the system handle that load? Structural health answers it, and only means something now that the load is known. Then freedom: where did capital actually stand? Average gate rank and compression answer together. Finally shape: was any of this steady? Persistence, volatility, and transitions close the read. The order runs from context to conduct to position to texture, which is the sequence in which the answers actually depend on each other.

The composite states

The combinations that carry the most information.

Certain pairings are worth naming because they recur and because each implies a different next action. High pressure with high health and preserved freedom is a tested-and-held month — document what worked; this is the profile the system is trying to reproduce. High pressure with low health is the full diagnostic trigger: the month was hard and the response degraded, so the review works through compliance, then execution, then structure. Low pressure with low health is the quiet warning — discipline slipping without the excuse of difficulty, and usually the earliest catchable form of drift. And any reading paired with low persistence carries a caveat: an unstable month's other metrics are less trustworthy, because the averages behind them describe a state the system never actually settled into.

The standing prohibitions

Two rules the whole diagnostic layer enforces here.

The first is the SDE's interpretation discipline, and it applies verbatim to this panel: never read one metric alone — the operator's job is to reconcile the whole stack, not cherry-pick the strongest reading. A single excellent health score does not redeem a month whose pressure was low and whose persistence collapsed. The second is the authority boundary that governs every diagnostic in MARS: when the gate map sits in Floor or worse, governance dominates and positive diagnostic readings are context only. No composition of these six metrics authorizes deployment, lifts a cap, or argues with a brake state. They describe the month; the ladder governs the account.

• Reconcile, don't cherry-pick — a mixed panel is a mixed month, and the review should say so.
• Low persistence discounts every average in the panel. Note it before trusting the rest.
• Diagnostic strength never outranks capital state. This is the same boundary the EV lab, VIP, and benchmark all live under.

Where the read goes

Into the regime label, and into the review record.

The six metrics' immediate destination is the regime engine, which scores each on the −2 to +2 ladder and compresses them into one state with one driver sub-label. But the panel read shouldn't be skipped in favor of the label: the regime is a summary, and summaries are for communication and sequencing, while the panel is where the diagnostic detail lives that a summary necessarily discards. The disciplined monthly routine reads the panel in order, records the composite state in the review, and then notes the regime label alongside it — so that a year later the record shows both what the month was and what it was classified as, which is exactly the evidence needed to eventually calibrate the classifier itself.

The key idea

Six metrics were designed to be read as one sentence.

'A heavy month, handled with discipline intact, capital freedom moderately compressed, states holding steadily' is a complete monthly diagnosis, and it uses all six inputs in their proper relationship. Read separately, the same six produce a scatter of numbers that invite selective attention. The order — load, response, freedom, shape — is what turns the panel into a sentence, and the sentence is what a review can actually act on.