NexusWatch · Yemen · 2026-10-21
🎯 Today's Call
The Ledger — 68 resolved today, 14 hit · OONI Brier 0.058 over 263, -2% vs base rate · FX Brier 0.245 over 3024, -5% vs base rate · 939 open, next resolves 2026-10-22
68 calls settled against external ground truth today: 14 hit, 54 miss. We called Uganda at 34% for currency depreciation by today (base rate 26%) — it depreciated ≥1.5%. We called Morocco at 35% (base rate 26%) — it didn't move. Top open call: Morocco again at 38% for depreciation by October 30th, twelve points above its 25% base rate.
📊 Top Signal
OFAC designated twelve entities under Iran sanctions on October 15th — four banks, six individuals, and two energy infrastructure firms. Every one targets the financial and energy sectors that keep Iran's economy functioning: Chabahar International Bank Overseas, Farzanegan Energy Pars, and named compliance officers at state institutions. Iran's censorship deviation sits at +12 today on OONI's measurement, marking sustained network interference through the week. Russia logged 53,291 confirmed blocking measurements in the last seven days — nearly double Iran's 29,797 and the highest weekly count in the OONI feed. Two states running record interference levels while one absorbs the largest single-day sanctions package we've seen this month. What it means: Iran's external financing channels are narrowing fast, and the blocking surge suggests internal communication control ahead of something — protests, policy shifts, or external pressure the state anticipates. Energy markets haven't moved much (crude oil ETF down 0.12%), so this isn't priced in yet.
🌍 The Board
Movers
- Sudan 90 (▲+6) — driver not identified in today's data.
- Turkey 43 (▼-8) — driver not identified in today's data.
- Taiwan 44 (▼-7) — driver not identified in today's data.
- Haiti 75 (▼-4) — driver not identified in today's data.
- Iran 85 (▲+12) — OONI recorded 29,797 confirmed blocking measurements in the last 7d.
- North Korea 81 (▲+8) — driver not identified in today's data.
Crises
- Sudan — level 90, deviation +11.5. Seven-day trajectory shows a jump from 6 to 11.5 today. No driver identified in the data, but it's the second-highest structural risk on the board and the deviation is climbing. Watch for displacement reports or sanctions news.
- Iran — level 85, deviation +12. OFAC hit twelve entities on October 15th, and OONI logged nearly 30,000 blocking measurements this week. Sustained censorship at this scale typically precedes internal crackdowns or external escalation.
Markets S&P 500 ETF (SPY) up 0.34%, crude oil ETF (USO) down 0.12%, Treasury bond ETF (TLT) up 0.84%. The energy sector ETF (XLE) gained 0.25% despite flat crude — positioning, not supply shock. Natural gas ETF (UNG) up 0.38%. Dollar index ETF (UUP) down 0.34%, which usually loosens pressure on emerging market currencies — but Morocco's call sits at 38% and Uganda just depreciated. Iran's sanctions package and Russia's censorship surge aren't moving energy prices yet, which means Hormuz risk isn't priced in. If blocking continues and OFAC adds more designations, that changes.
What would change our mind: Iran's OONI measurement count dropping below 5,000 confirmed blocks over seven days, or OFAC removing any of the October 15th designations.
🙊 What We're Not Saying
- Sudan's +6 move to deviation 11.5 — no driver identified. It's rising on a seven-day trajectory, but we have no sanctions, no quake, no OONI surge to explain it.
- Turkey's -8 drop and Taiwan's -7 drop — both large moves, both unexplained by anything in the data.
- North Korea's +8 — volatile over seven days (0 → 0 → 0 → 0 → 0 → 8 → 0 → 8), but no event in the context ties to it.
- Russia's 53,291 blocking measurements are the highest weekly count in the feed, but we have no external corroboration of what's being blocked or why the volume spiked.
🧨 The Long Fuse
Morocco's currency call has been the top divergence from base rate for a week. We're at 38% for depreciation by October 30th (base rate 25%), and today's resolution batch shows we missed it for October 21st at 35%. If the currency holds through month-end without moving, the model's Morocco signal will have produced a week of high-confidence calls that all missed — a calibration problem worth watching. If it does depreciate in the next nine days, we'll have been early but directionally correct. Either way, it's a falsifiable test of whether the model's FX divergence signals are reliable or noisy.