How the scores are built

Methodology

A hybrid of curated geopolitical intelligence and live market-derived signals, combined into one Score of Scores.

01Overview

M3I Research is a hybrid geopolitical-risk + market-stress scoring platform. It blends curated qualitative intelligence (refreshed every few hours) with live market-derived signals to produce a composite Score of Scores (SOS) for investors tracking macro risk during active conflicts. The system was built for the 2026 US–Iran conflict but is designed to generalize to any resource-war scenario.

02Score of Scores

The SOS is a weighted average of five sub-inputs:

25%War Riskqualitative read on military-escalation probability from diplomatic signals, military posture and historical conflict patterns.
15%Market Turbulencecomposite of independent market-stress signals (see below).
20%Ceasefire Riskprobability of ceasefire collapse, informed by diplomatic developments and market-spread analysis.
20%Hormuz ShippingStrait of Hormuz disruption severity from vessel traffic, AIS monitoring and news sentiment.
20%7-Model Scanner Averageaverage of all seven analytical models in the system.
0–29 LOW 30–49 MODERATE 50–69 ELEVATED 70–100 HIGH
03The seven models
MODEL 01
War Resolution
Probability of diplomatic resolution within defined timeframes. Uses historical war-pattern analysis, diplomatic engagement level and resource-war duration benchmarks.
MODEL 02
Market Turbulence
Multi-signal composite of real-time market stress: VIX term structure, credit spreads, oil volatility, equity drawdowns, safe-haven flows and sector rotation.
MODEL 03
Recession Risk
Probability of a US recession driven by energy-price shocks, consumer-spending compression and monetary-policy constraints.
MODEL 04
Escalation Risk
Probability of conflict intensification beyond current parameters. Tracks military posture, proxy activations and nuclear-threshold indicators.
MODEL 05
V-Shape Recovery
Probability of a rapid market recovery on conflict resolution, based on historical post-war recovery patterns and current structural-damage assessment.
MODEL 06
Historic Dip Zone
Whether current market levels represent a historically significant buying opportunity relative to the prevailing geopolitical risk.
MODEL 07
Hormuz Shipping
Strait of Hormuz disruption severity, combining AIS vessel tracking, news-derived headline sentiment and structural closure indicators.
04Historical war pattern engine

The Pattern Engine analyses 13 US conflicts since 1861 to identify resolution patterns. Key insight: resource wars (oil, shipping lanes) resolve roughly 21× faster than ideological wars (~103 days vs ~2,340 days on average). The engine classifies the current conflict as a resource war and applies compressed-timeline probabilities. Dataset: 4 resource wars, 7 ideological, 2 occupation; 6 of 12 completed wars had failed ceasefires before resolution.

Known limitation: a small historical dataset, and the current conflict involves a nuclear-threshold state with no prior US resource-war analog. The engine explicitly flags this as uncharted territory in its bear case.

05Live engine vs curated scores

The site runs in hybrid mode. Curated scores are refreshed every few hours by the M3I analytical engine, capturing qualitative intelligence that pure market signals can't (diplomatic developments, military posture). The live market engine recomputes proxy scores continuously from real-time data (VIX, Brent–WTI spread, gold, BTC, credit). When curated data is fresh it controls the SOS display; when stale, the live engine takes over so the scores never freeze.

06Data sources

Market data: Yahoo Finance (real-time). Hormuz vessel tracking: AIS data and news-derived estimates. Geopolitical intelligence: multi-source news aggregation (Reuters, CNBC, Al Jazeera, CNN, Bloomberg and others). Historical conflict data: Congressional Research Service and academic databases.

07Limitations

Small dataset: 13 US conflicts; non-US and proxy conflicts not yet incorporated. Nuclear-threshold gap: no historical analog for a US resource war against a nuclear-threshold state. Proprietary calibration: individual signal weights and constants are not published. Single-conflict optimization: built for the 2026 US–Iran conflict; generalization is untested. Not financial advice: M3I scores are analytical outputs, not investment recommendations.

08API access

An open-source API is available on GitHub — MIT licensed, no authentication required.

Repository: github.com/m3iresearch8/m3i-api  ·  Endpoint: m3iresearch.com/api/scores.json

M3I Research is for informational and educational purposes only. Nothing on this site constitutes financial, investment, legal, or tax advice, an offer or solicitation to buy or sell any security, or a recommendation of any investment strategy. The scores, signals, gauges and interpretations are outputs of a proprietary research framework and should not be relied upon as the sole basis for any investment decision. References to historical patterns, valuations and market levels describe past observations and do not guarantee future results. All investing involves risk, including the possible loss of principal. M3I Research is not a registered investment adviser or broker-dealer. Always do your own research and consult a qualified, licensed financial professional before making any investment decision.

← Back to the dashboard

M3I scores are analytical outputs, not investment recommendations.