Will the ceasefire between Iran and the United States continue to hold 7 days from today?

TL;DR

I estimate a 67.2% chance that the U.S.-Iran ceasefire is still treated as in effect on June 9, because the ceasefire label has survived serious direct fire but the next week is a live diplomatic and military danger window.

Context

The forecast date is Tuesday, June 2, 2026, so the resolution date is Tuesday, June 9, 2026. As of June 2, major reporting still treats a ceasefire as existing: Reuters reported a “shaky ceasefire in place” while Iran reviewed a proposed agreement and Trump said talks were continuing, and AP reported that Iran-linked outlets said Tehran had stopped communicating with mediators while Trump disputed that claim (Reuters via Investing.com, June 2, 2026; AP, June 2, 2026).

The ceasefire is not calm. AP reported on June 1 that the U.S. bombed Iranian radar and drone sites after Iran shot down an American drone, and that Iran then targeted U.S. soldiers in Kuwait with missiles that the U.S. said it shot down; AP still called this a nominal ceasefire being tested, not a ceasefire that had ended (AP, June 1, 2026). CENTCOM’s own June 2 statement says U.S. forces disabled another tanker while “the ceasefire with Iran continues,” which is central for this resolution standard (CENTCOM, June 2, 2026).

Evidence

The historical backbone points to a high one-week survival rate once the threshold is not zero violence. The ETH/PRIO Civil Conflict CeaseFire dataset covers 2,202 civil-conflict ceasefires across 66 countries and 109 conflicts from 1989 to 2020; its authors report that almost all ceasefires suffer some violations, with failed nationwide ceasefires having median durations of 10 days at a one-fatality threshold, 65 days at a 25-fatality threshold, and 193 days at a 100-fatality threshold (Clayton et al., Journal of Conflict Resolution, 2023; PRIO dataset summary). A simple exponential model using the 65-day median gives a one-week failure risk of $1-e^{-7\ln(2)/65}=7.2\%$, or 92.8% survival. I cut far below that anchor because this case is interstate, has no clear neutral monitoring mechanism, and is already seeing direct U.S.-Iran exchanges.

The same-case record matters more. The ceasefire began around April 7-8 and has lasted about 56 days by the forecast date under the source framing used by AP, Reuters, and CENTCOM (Reuters via Investing.com, April 8, 2026; Reuters via Investing.com, June 2, 2026). During that span, it absorbed incidents that would break a stricter ceasefire coding rule. On May 7, CENTCOM said Iran launched missiles, drones, and small boats at three U.S. destroyers in the Strait of Hormuz, and that U.S. forces then struck Iranian military facilities, while CENTCOM said it did not seek escalation (CENTCOM, May 7, 2026). On May 27, CENTCOM said Iran fired a ballistic missile toward Kuwait and launched five one-way drones near Hormuz, calling it an egregious ceasefire violation rather than a formal end to the ceasefire (CENTCOM, May 28, 2026). On May 31, CENTCOM said U.S. forces struck Iranian radar and drone command sites in Goruk and Qeshm Island after the shootdown of a U.S. MQ-1, but also said no U.S. service members were harmed and that it would protect U.S. interests during the ongoing ceasefire (CENTCOM, May 31, 2026). AP then summarized on June 2 that none of the warring parties had officially exited the ceasefires, even as the term was rapidly losing meaning (AP, June 2, 2026).

The maritime data shows why the crisis is still unstable. My PortWatch pull used daily Strait of Hormuz vessel-transit counts and capacity, with 1,000 daily observations in the extract from September 5, 2023 through May 31, 2026; the latest observation was 10 vessels on May 31, 2026 (IMF PortWatch). The full comparable May history in that extract is:

This cuts both ways. The near-closure of Hormuz creates constant contact between U.S. blockade enforcement and Iranian coercion. It also gives both sides a large payoff from keeping a ceasefire frame alive long enough to reopen traffic.

The diplomatic signal is mixed. On May 28, Axios reported that U.S. and Iranian negotiators had reached a 60-day memorandum of understanding to extend the ceasefire and begin nuclear talks, but that Trump still had to approve it and Iran had not confirmed acceptance (Axios, May 28, 2026). AP reported on May 29 that Trump had not decided whether to move forward with the deal, which would extend the fragile ceasefire by 60 days and reopen Hormuz (AP, May 29, 2026). By June 2, Reuters reported that Iran was reviewing the proposed text but had not communicated with Washington for several days, while Trump said talks had been continuous and that he expected a deal over the next week (Reuters via Investing.com, June 2, 2026). Lebanon is the biggest spoiler: AP reported on June 2 that Israeli strikes in southern Lebanon killed 11 people and that Lebanon fighting had become a significant sticking point because Iran wants any U.S.-Iran extension to also end the fighting there (AP, June 2, 2026).

My forecast is a scenario-weighted estimate:

$$P(YES)=\sum_s P(s)P(YES\mid s)$$

where $s$ is the path the crisis follows through June 9.

That sums to 67.2%. I put more weight on YES than the surface violence suggests because official and major-source language has repeatedly separated ceasefire violations from ceasefire collapse. I keep the probability well below 75% because the next week includes a live decision on the proposed MOU, fresh direct strikes, the June 2 tanker-disabling incident, and a Lebanon linkage that could let either side say the bargain has failed.

What's non-obvious

The key fact is not that both sides are shooting. The key fact is that the ceasefire label has survived the shooting. CENTCOM called May 27 an egregious violation, then said on May 31 and June 2 that the ceasefire was still ongoing; AP and Reuters are using words like nominal, shaky, and still in place, not ended (CENTCOM, May 28, 2026; CENTCOM, May 31, 2026; CENTCOM, June 2, 2026; AP, June 2, 2026; Reuters via Investing.com, June 2, 2026). Under this question’s resolution rule, that label is not cosmetic. It is part of the event.

The other hidden crux is Lebanon. Lebanon fighting does not itself resolve this question NO, because the question is about direct U.S.-Iran hostilities. But Iran is trying to make Lebanon part of the U.S.-Iran bargain, while the U.S. and Israel treat it as more separable; that makes Lebanon a real trigger for a U.S.-Iran ceasefire breakdown even though it is not the resolving event by itself (AP, June 2, 2026; AP, June 2, 2026).

Limitations

The biggest limitation is definitional. The resolution criteria list airstrikes, missile strikes, and naval attacks as examples of renewed hostilities, but they also exclude isolated incidents unless major sources characterize them as ending the ceasefire. If the next week looks like June 1 repeated several times, a resolver could call that sustained hostilities even without a formal exit. My estimate assumes the resolver gives strong weight to AP/Reuters/BBC-style characterization.

The proposed 60-day MOU is not public, and neither Trump nor Iran had clearly given final approval as of the June 2 reports (Axios, May 28, 2026; AP, May 29, 2026; Reuters via Investing.com, June 2, 2026). The historical ceasefire data is civil-conflict data, not a clean interstate analogue, so I use it only as a weak base-rate anchor (PRIO dataset summary). The shipping data is AIS-based and the latest observation in my extract is May 31, not June 2, so it can miss dark or delayed vessels during conflict (IMF PortWatch). Annulment looks unlikely because AP, Reuters, and CENTCOM are reporting the ceasefire status frequently enough that a source-based resolution should be possible.