Skip to content
Straits

Methodology — archived

Hormuz Index v0.1 (archived).

This page is the frozen v0.1.2 methodology, preserved for citation. The single 0–100 Hormuz Index it describes was superseded on 2026-05-21 by v0.2, which split the brand into two parallel indices — Crisis Pressure (state) and Escalation Probability (30-day forecast). Live methodology lives at /methodology#hormuz-index; this archive is unchanged from launch.

Hormuz Index · v0.1.2

How the Hormuz Index was computed (v0.1).

The v0.1 Hormuz Index was a single 0–100 number summarizing crisis pressure on the Strait of Hormuz across five measured signals: prediction-market closure odds (25%), Brent options implied volatility (25%), GDELT geopolitical event flow (20%), AIS transit deviation (15%), and war-risk insurance pricing (15%). It recomputed every 5 minutes, with weights, rails, and per-input contribution published on the audit endpoint.

Why v0.1 was superseded. Four of the five v0.1 inputs measured the current severity of the crisis — what just happened, what was being priced, what underwriters were charging today. Only the prediction-market component was genuinely forward-looking. The v0.1.2 patch (2026-05-20) softened the band language to describe state rather than forecast; v0.2.0 went further, splitting the index structurally so each number is backed only by inputs that match its claim.

v0.1 inputs and weights.

  1. 25%

    Closure probability

    Implied probability from Polymarket and Kalshi contracts on Hormuz closure and Iran-conflict escalation. Gated to a residual weight when contract volume was thin.

    Polymarket + Kalshi public APIs

  2. 25%

    Oil-market dread

    Composite of 30-day Brent at-the-money implied volatility and 25-delta put/call skew. Measured what the institutional oil-options market was pricing for a near-term shock.

    CBOE OVX via FRED + Stooq + Yahoo fallback

  3. 20%

    Event pressure

    Goldstein-scale-weighted count of escalation events from GDELT over a rolling 72-hour window, with exponential decay.

    GDELT 2.0 events database

  4. 15%

    Physical reality

    IMF PortWatch daily transit count deviation from the pre-crisis baseline. The unique physical-strait signal.

    IMF PortWatch chokepoint6 + proprietary AIS aggregation

  5. 15%

    Insurance

    War-risk insurance multiple vs peace-baseline rates for VLCC voyages through Hormuz. The slowest input to react but the most consequential when it moved.

    Trade press + proprietary aggregation

v0.1 combination rule.

The v0.1 base composite was a weighted average of the five normalized component scores. An override fired when any single component scored above 75: the final value was then at least 0.9× that component’s score, even if the weighted average was lower.

Polymarket contracts were gated by liquidity: when 7-day dollar-volume on relevant Hormuz contracts fell below $25,000, the market spine collapsed to a residual 5% weight and the remainder was redistributed pro-rata across the other inputs. This binary gate was the load-bearing defense against a single whale moving the headline; v0.2 replaced it with a continuous low-confidence reporting floor on the dedicated Escalation Probability Index.

v0.1 history.

Documents written under v0.1.x methodology versions live in the legacy hormuzIndex/ Firestore collection. Their liveValue field is preserved as published (immutable); no further writes from v0.2 onward go to that collection. v0.2 indices write to crisisPressureIndex/ and escalationProbabilityIndex/.