M = [(Participant × Institutional × Sovereign) · ν]²
BMR is the market-side signal layer for the DFTE (Dual Field Trading Engine). It reads the price field the same way KindEarth reads the world field — as a coherence system, not a prediction machine.
feeds/ Raw data ingestors (market data, COT, options, macro)
core/
normaliser.py Scale signal normalisation → [-1, +1]
nu_engine.py ν coherence computation across scales + timeframes
lsii_price.py Late-Move Inversion Index (translated from KindPath Q)
field_state.py ZPB/DRIFT/IN-Loading/SIC classifier
curvature.py Market Curvature Index (tokenisation gap)
bmr_profile.py Full BMR field reading synthesiser
mfs.py Market Field Score output
tests/
bmr_server.py FastAPI signal server (same pattern as q_server.py)
| State | ν | Meaning |
|---|---|---|
| ZPB — Coherent Trend | > 0.75 | All scales aligned, M amplifying |
| DRIFT — Transition | 0.40–0.75 | Partial alignment, weakening |
| IN-Loading — Compression | 0.15–0.40 | Scales diverging, pressure building |
| SIC — Event | < 0.15 | Coherence collapse, forced movement |
BMR (Market Field Score) + KEPE (World Field Score) → DFTE trade selection