W.E.T. Geopolitical Risk Index: methodology

The W.E.T. Geopolitical Risk Index is a daily 0–100 reading of how much geopolitical conflict risk prediction-market traders are pricing.

Components

Each day the index pulls active markets from the politics and elections categories (Kalshi + Polymarket) whose question references a conflict-risk event — war, military action, invasion, a strike, nuclear escalation, sanctions, a coup — plus the live flashpoints (Israel/Iran/Gaza, Ukraine/Russia, Taiwan, North Korea). Risk-reducing questions (ceasefires, peace deals, withdrawals) are not discarded — they are included oriented 1−p (a ceasefire priced at 78% is 22% of unresolved risk), so the index reads the whole board while every component still points the same way: higher = more risk.

Value

For each component we read the implied YES probability (0–1) using the site's standard price extraction. The index is the volume-weighted mean across events, expressed 0–100 — each real-world event contributes once, through its leading qualifying market, so one conflict priced across several contracts is not double-counted. No single market may exceed 25% of the weight (the excess is redistributed), so one whale contract can't be the index. Larger markets otherwise carry more weight. When fewer than three qualifying events are live, the index reports insufficient data rather than a number.

Publication & freshness

The value is computed and published once daily to a durable, dated snapshot. The "as of" date advances every run; the "last moved" date advances only when the value changes by at least 0.5 points, so an unchanged reading is never dressed up as fresh. Every daily point is appended to a permanent history.

What this index is not

This is a benchmark reading of market prices, not a W.E.T. forecast and not advice. A high value does not predict that the event happens — it reports what prediction-market traders are collectively pricing right now. Prices move continuously and can be wrong.

An accuracy record for a benchmark requires resolved outcomes accumulated over time. W.E.T. publishes this index daily with a dated, immutable history from first publication — that published series is the audit trail. Calibration analysis will ship strictly when the resolved-outcome corpus supports it, never before. See the site methodology for the same discipline applied to the W.E.T. Score.