The Probability Desk

How the Probability Desk works: a worked scenario

A worked, illustrative example of the Probability Desk report standard — provenance box, weak-signal card, consensus gap, tripwire ladder and transmission map, on one scenario.

A worked example showing the standard structure of a Probability Desk flagship report. The figures below are illustrative, included to demonstrate the format and the discipline — not a live forecast or a recommendation.

The forecast question

Will commercial shipping through the Strait of Hormuz experience a continuous disruption of more than 72 hours before 31 July 2026, confirmed by vessel-tracking data or at least two credible maritime sources?

Current probability6%
Prior4%
▲ +2 pts since last edition
Confidence3/5
Model legEstimateConf.Direction & note
Historical base rate3%4↔ reference class: short, reversible Gulf disruptions are rare and brief
OSINT weak-signal model9%2▲ inspection notices + naval activity deteriorating
Market-implied signal qualitative~2%2↔ tanker rates and regional risk premia calm
Expert prior7%3▲ maritime-risk houses raising watch levels
Simulation (multi-agent)unavailablenot run for this illustrative example
Editorial judgment−1 ptdown-weight thin, single-origin local signals
Biggest drivers: the weak-signal and expert legs. Biggest uncertainty: whether the local inspection notices corroborate. What would change our mind: vessel-tracking confirmation of rerouting, or an official all-clear with normal throughput.

Executive summary

The base case remains an orderly strait. We raise the probability of a short disruption modestly, to an illustrative 6%, because early, local indicators are deteriorating faster than market pricing suggests. The extreme-tail path — an extended closure — stays well below 1%, but it earns standing monitoring on impact and detectability, not on likelihood: a universal owner is structurally exposed to the energy, freight and inflation transmission, and the cost of monitoring is trivial against the cost of surprise. What to watch next: vessel-tracking throughput and whether the local notices reach independent corroboration.

Probability snapshot

ScenarioProbabilityImpactVelocityConfidencePriority
Benign — tensions ease62%LowSlow3Watch
Base — orderly strait, episodic friction30%Low–Med3Standard
Adverse — brief (>72h) disruption6%HighFast3Elevated
Severe — multi-day disruption1.3%SevereFast2Elevated
Extreme tail — extended closure0.7%SystemicFast2Priority

We mapped a wider set of paths internally; the five above are the set that matters for long-duration capital. Weights are illustrative and sum to 100%.

Low probability, not low importance

The extreme-tail path — an extended closure of the strait — sits at an illustrative 0.7%. It is unlikely because the economic damage would be mutual and the historical precedent is for short, reversible episodes. It matters anyway because the impact would be nonlinear and broad: an oil and freight shock transmits to inflation, rates and currencies, reaching every asset class a whole-market book holds. This is a tail we watch, not a base case we expect.

Weak signalStrength 3/5Noise 3/5Affects: likelihood

Inspection notices and unusual naval activity reported near a key approach to the strait

Source
Two local-language notices + regional maritime reporting (illustrative)
Why it matters
Inspection regimes have, in past episodes, preceded transit slowdowns
Why it may be noise
Seasonal exercises occur without escalation
What would confirm
Vessel-tracking rerouting plus an independent third source
What would invalidate
An official all-clear and normal throughput within 48 hours
Asset-owner relevance
Sovereign and insurer energy and marine exposure; pension inflation and duration

The consensus gap

Desk view
6%
Consensus (qualitative)
~2%

Consensus-gap score: 6 / 10 — some underpricing. The gap exists because the earliest indicators are local-language and have not yet reached English-language coverage or option pricing. What closes it: vessel-tracking confirmation, or a single maritime-incident headline.

We do not state a precise market-implied probability here because none was calculated; the consensus figure is a qualitative estimate.

Tripwire ladder

~10%moves higher if
Vessel-tracking shows rerouting; a second independent source corroborates the inspection notices; regional risk premia begin to firm.
~18%into mainstream concern if
A vessel is struck, seized, or detained in the strait; a transit halt exceeds 24 hours; tanker rates spike.
~3%moves lower / retires if
An official all-clear with normal throughput; de-escalation signalling; independent sources confirm routine activity.

Monitoring frequency: intraday on vessel-tracking and maritime newswire; daily on regional risk premia.

Asset-owner transmission

  • TriggerTransit disruption in the strait
  • ImmediateTanker rerouting; freight and insurance costs rise
  • MacroOil-price pressure feeds goods inflation
  • MarketCentral-bank reaction; duration and curve repricing
  • Owner exposureSWF energy; insurer marine/cat; pension inflation & duration; EM debt

What would change our mind

Lower the probability: an official all-clear, de-escalation talks, sustained normal throughput. Raise it: corroborated rerouting, a maritime incident, firming risk premia. Change the impact: evidence of spare rerouting capacity (lower) or a concurrent supply shock elsewhere (higher). Invalidate the scenario: a durable security arrangement that removes the disruption pathway.

Data ingested & what is missing

This illustrative example draws on the macro and energy spine and on open maritime and regional reporting. Honestly noted as not used here: a multi-agent simulation leg (not run for this sample), commercial vessel-tracking and option-implied pricing (not licensed for this example), and corroborating local-language sources (single-origin at the time of writing). A live report discloses the same gaps rather than papering over them.

Forecast accountability

A live version of this forecast would be logged with its prior, tracked as the probability moves, and scored against the outcome on the resolution date. Publishing and grading our own calls — not just making them — is the discipline that separates decision-grade research from commentary.

Illustrative scenario analysis prepared to demonstrate the Probability Desk report standard. Probabilities are weighted across base-rate, expert-prior and (where available) simulation inputs; methodology available on request. Editorial scenario analysis only — not investment, actuarial, or geopolitical advice.

The Daily Brief

The morning briefing for the people who allocate long-horizon capital.

Research, charts, video and podcast analysis for the institutions investing at the scale of the world.