Skip to content
Montex Alpha Rail System
$497

Regime Classification · The Ladder

Six metrics, one score, ten named months.

Input metrics

6

Composite range

−12 to +12

Main regimes

10

Sub-labels each

4

The key idea

Further illustration

GREEN ≥ +0.25RYELLOWREDROLLING EVW1W2W3W4W5W6W7W8W9W10W11W12

Weekly EV tagged GREEN / YELLOW / RED against expectancy thresholds, with the rolling-EV line separating persistent edge from one lucky week.

The six inputs

Pressure, health, volatility, transitions, persistence, compression.

Each metric carries its own direction. Monthly gate pressure and gate volatility score better when lower. Structural health and gate persistence score better when higher. Gate transition count scores better when lower, because churn between capital states is instability regardless of the average depth. Gate compression is the sixth, and it is the one the manual explicitly flags for audit before live reliance.

  • Pressure measures how heavy the month was from drawdown, spike clustering, and compression.
  • Health measures how well the system handled the month, not whether the month was easy.
  • Volatility measures instability relative to the size of the month, not raw movement.
  • Persistence measures how long gate states held, which is durability rather than direction.

No weights, on purpose

Every metric contributes through the same scale.

The first production version is deliberately unweighted, uses fixed thresholds rather than rolling percentiles, and refuses to grow past six inputs. The reasoning is that weights invented before enough history exists are guesses dressed as calibration, and a twenty-metric classifier that nobody can audit produces labels nobody can challenge. Empirical threshold calibration is a named future upgrade, deferred until enough live months exist to calibrate against.

Sub-labels are descriptors

03

Four labels per regime, and the pairing is validated.

A sub-label explains why or how the regime is expressing itself: Instability with High Transition is a different month from Instability with Low Persistence, and the difference changes the review focus without changing the main state. The rule holding the ontology together is that the selected sub-label must belong to the selected regime's own four-label set, and a quality check verifies that the pairing exists in the rules table rather than trusting the formula that produced it.

One month, one label

No stacking, no second opinion inside the same row.

The engine produces one main regime and one sub-label per month. This is a design constraint rather than a limitation: a classifier permitted to return two states will return two whenever the evidence is mixed, and mixed evidence is exactly the condition where the operator most needs a single decision to act against. The Transitional and Balanced regimes exist precisely so that ambiguity has somewhere to be recorded honestly.

Reference

The composite score ladder

GroupRegimeScore bandInterpretation focus
PositiveExpansion+10 to +12Strongest aligned state, broad structural strength
PositiveExpansion (Developing)+7 to +9Improving but not yet mature
NeutralBalanced+4 to +6Mixed but stable enough to avoid a stress classification
NeutralTransitional+2 to +3Rotation, handoff, or changing internal state
NeutralCompression0 to +1Neutral-to-defensive squeeze building under the surface
WarningFragile Recovery−1 to −2Improving from stress but still unstable or incomplete
WarningInstability−3 to −5Volatility, transition, or persistence stress rising
NegativeDegradation−6 to −8Structural weakening with broader erosion signs
NegativeDrawdown Control−9 to −10Capital-defence posture dominating the regime
NegativeCritical State−11 to −12Structural breakdown or crisis behaviour

Before you go deeper

Connected inside MARS

This module doesn't work alone.

Connected — 01

Regime Classification (full page)

Open system page →

Connected — 02

Why It Exists

Open system page →

Go deeper

Operator briefs on this territory.

Deep dive — 01

Uneven rungs: why the regime ladder is finest near the middle.

Ten regimes, unequal bands. Resolution is concentrated where the state is about to change, not where it is settled.

Read the full brief →

Deep dive — 02

Ten regimes, forty sub-labels: naming a month precisely.

Main regime = the state. Sub-label = the driver. Ten × four, and never stacked.

Read the full brief →

Deep dive — 03

Regime classification: the system's weather report on itself.

Six metrics, one composite score, ten regimes — how MARS classifies its own internal state monthly.

Read the full brief →

Every module ships in the complete MARS package.

One price. Eleven workbooks, three TradingView indicators, and the full manual library — $497.

Get MARS Pro — $497Explore the Platform