14 analytical agents predict calendar weeks for different conflict resolution milestones. Predictions are updated every 12 hours. The curves show where agents expect each milestone to land. The higher they peak, the stronger their consensus around the week.

When will the United States cease offensive military strikes against Iranian targets?

Resolution criteria: 7 consecutive days without a US offensive strike against Iranian military or nuclear infrastructure.

Consensus cumulative probability of occurring within

Assessment: 3 July 2026, 6:00 PM CET

36%
1 week
by 10 Jul
52%
2 weeks
by 17 Jul
68%
4 weeks
by 31 Jul
90%
2 months
by 1 Sept
94%
3 months
by 1 Oct
Cons.W28(6 Jul)36%
Select agents to highlight
Change assessment 3 Jul 2026, 6:00 PM CET

Note: Assessments before 13 March 2026 were backfilled using only evidence available at each assessment date. These earlier estimates may still reflect hindsight bias and are included for illustration.

Agent Predictions

Agent ↑1 week2 weeks4 weeks2 months3 months
Consensus 36% (-3)52% (=)68% (+2)90% (+1)94% (+1)
Competing Hypotheses85% (+5)89% (+6)92% (+7)98% (=)98% (=)
Regional Dynamics46% (+2)66% (+4)79% (-3)96% (-2)99% (=)
Misperception Analyst38% (+3)60% (+5)84% (+9)99% (+5)99% (=)
Escalation Analyst46% (-2)64% (-2)82% (=)99% (+2)99% (=)
Nuclear Posture1% (=)3% (=)14% (+3)61% (+8)80% (+10)
IRGC Specialist22% (=)37% (+4)57% (+12)89% (+5)92% (=)
US Military Analyst32% (-1)60% (+7)83% (+12)97% (+3)97% (=)
Inadvertent Escalation21% (-1)39% (-1)60% (=)85% (-4)93% (-4)
Adversarial Red Team10% 22% 36% 79% 93%
Proliferation Analyst38% (-4)60% (+8)78% (+17)98% (+4)98% (+1)
Tail Risk Analyst64% (-1)76% (+2)81% (=)87% (=)87% (=)
Strategic Scenario Planner31% (=)51% (+12)65% (+11)94% (+7)98% (+4)

Technology team exercise. Calendar weeks are model-generated estimates, not forecasts. Curve height reflects consensus among analytical agents, not calibrated probabilities. Dates represent the Monday of each calendar week (ISO 8601 convention). Agents predict at weekly granularity — differences within a week are not meaningful. Assessments before 13 March 2026 were backfilled using only evidence available at each assessment date. These earlier estimates may still reflect hindsight bias and are included for illustration. Do not use for operational planning.