NexusWatch · Yemen · 2026-10-27
🎯 Today's Call
The Ledger — 67 resolved today, 14 hit · OONI Brier 0.058 over 263, -2% vs base rate · FX Brier 0.234 over 3427, -5% vs base rate · 938 open, next resolves 2026-10-28
67 calls settled against external ground truth: 14 hit, 53 miss. Uganda currency depreciation resolved HIT — we said 36% (base rate 25%) by 2026-10-27. The most informative miss: Morocco, where we called 35% (base rate 24%) for currency depreciation by today — the dirham held. Four Uganda calls still open through November 10th, each 37-39% against a 25-26% base rate.
📊 Top Signal
Russia and Iran are running the highest censorship levels we can measure, and they're running them simultaneously. OONI recorded 53,999 confirmed blocking measurements in Russia over the last seven days, and 27,981 in Iran through October 26th. That's not a crackdown — that's infrastructure. Both states are now operating persistent, nationwide filtering at a scale that dwarfs every other monitored country: Turkey is third at 4,455 measurements, less than a tenth of Russia's load. Iran's live censorship component jumped +3 points today to 12, lifting its total CII to 100 (structural 85 + live 15). This matters because network interference is a leading indicator — it precedes arrests, troop movements, and wire coverage by days or weeks. When two nuclear-armed states with active conflict equities (Ukraine, Syria, Yemen) are simultaneously operating max-scale filtering, the question isn't whether something is being suppressed. It's what.
🌍 The Board
Movers
Brazil 48 (▲+11) — driver not identified in today's data. Sudan 96 (▲+6) — driver not identified in today's data. Venezuela 78 (▲+6) — driver not identified in today's data. Syria 89 (▼-8) — disaster component decay; no new events in 24h. India 46 (▼-6) — OONI recorded 4,080 confirmed blocking measurements in the last 7 days, down from prior week. Myanmar 90 (▲+5) — OONI recorded 175 confirmed blocking measurements through October 26th.
Crises
Iran (level 85 + deviation 15) → Live censorship at 12, the highest component reading today. OONI data through October 26th shows sustained nationwide blocking. OFAC designated four vessels (PRINCE, HARMONY, FLAMINGO 1, DESTINY) under Iran-EO13902 on October 21st — sanctions pressure intensifying as filtering peaks. Watch for further designations or currency pressure.
Sudan (level 90 + deviation 6) → Volatile 7-day trajectory (6 → 14 → 11.5 today), no external driver identified. Second-highest structural risk globally, but no confirmed trigger in today's data.
Markets
Crude oil ETF (USO) dropped 1.95% to $166.23 — the largest single-day move in the energy basket. Energy sector ETF (XLE) rose 0.45% to $69.64, decoupling slightly. Natural gas ETF (UNG) flat at $9.56 (-0.21%). Gold ETF (GLD) up 0.33% to $367.82. Dollar index ETF (UUP) firmed 0.17% to $29.01. Treasury bond ETF (TLT) gained 0.42% to $81.68. Iran's CII at 100 and the Russia/Iran censorship spike aren't reflected in energy prices yet — Hormuz risk premium remains muted.
What would change our mind: A OFAC designation targeting Iranian crude export infrastructure, or OONI measurements dropping below 15,000 weekly in Russia for two consecutive weeks.
🙊 What We're Not Saying
- Brazil's +11 deviation jump — no driver identified in today's data. No censorship spike, no sanctions event, no disaster. Unexplained move.
- Venezuela's +6 rise to 78 — OONI logged only 6 measurements through October 25th, down from prior weeks. Not a censorship story.
- Sudan's volatility (6 → 14 → 11.5) — no headline, no OONI data, no quake. The 7-day chart shows a spike to 14 on one day; we don't know what drove it.
- BBC World wire items dated to Israeli elections and Brazil — both references appear misaligned with the 2026 calendar. Cannot verify and did not use.
🧨 The Long Fuse
Watch Uganda's currency through mid-November. We have four open calls stacked between November 7-10, each forecasting 37-39% chance of ≥2% depreciation against a 25-26% base rate. One hit today; if the next three resolve the same direction, it's a model edge. If they scatter, it's noise. The shilling's 14-day path will tell us whether our FX model has explanatory power or whether we're reading shadows. Checkable, dated, falsifiable — exactly how forecasting should work.