Macro Fragility Posture Matrix

Step 1: Read Four Tool States

Each tool independently computes its own signal level (NORMAL / ELEVATED / HIGH / CRITICAL) using its own methodology. The matrix does not modify or override individual tool calculations.

Step 2: Simplify to Three-Tier State

State 0 (Normal):   NORMAL (score 0–30)
State 1 (Elevated): ELEVATED (score 31–60)
State 2 (High+):    HIGH or CRITICAL (score 61–100)

State 1 (Elevated) affects the escalation watch list but does not contribute to S-level calculation. Only State 2 (HIGH or CRITICAL) counts toward the S-level.

Step 3: Compute S-Level

S-Level = count(tools at State 2)
S0 = 0 tools at HIGH+  → All Clear
S1 = 1 tool at HIGH+   → Single Fragility
S2 = 2 tools at HIGH+  → Dual Fragility
S3 = 3 tools at HIGH+  → Triple Fragility
S4 = 4 tools at HIGH+  → Systemic Consensus Collapse

Step 3b: Composite Score

The composite score displayed in the hero section is derived directly from the S-level:

Composite Score = S-Level × 25
S0 → 0/100, S1 → 25/100, S2 → 50/100, S3 → 75/100, S4 → 100/100

This is a discrete mapping, not a weighted average of individual tool scores. It reflects the count of tools at HIGH+, not the magnitude of individual readings.

Step 4: Match CONFIG Pattern

The specific combination of which tools are HIGH+ determines the CONFIG pattern (e.g., CONFIG-13 = GCDI + TYFI + GODI all HIGH, TOCI below HIGH). There are 16 possible non-trivial combinations. Each has a named signal and historical episode library.

Step 5: Map to Posture Labels

Posture labels describe historical tendencies observed in past periods with similar signal configurations. They use environment-state language (e.g., "rates readability: low", "growth-confidence support: weak") rather than portfolio-action language. They are not predictions or recommendations.

Known Limitations

• Historical sample sizes are small (2–5 episodes per CONFIG pattern since 2000)
• S-level classifications are applied retrospectively — they were not computed in real-time
• The matrix assumes tool independence, but shared variables (gold, 10Y yield) create partial correlation
• Posture labels are qualitative and cannot be mechanically translated to portfolio actions
• The framework does not account for individual circumstances, risk tolerance, or investment objectives
• Component tools update at different frequencies (daily, weekly, quarterly), creating mixed-vintage composites
• The composite score (S-level × 25) is a discrete step function, not a continuous risk measure

Data Source Classification

This matrix aggregates tool states produced from a mix of public-domain, public-use, and separately reviewed external data sources. Source and rights status are disclosed at the individual tool level. The matrix page displays tool scores and signal states (AhaSignals' own analytical outputs) rather than raw market data.