What will be the number of deaths that a concensus of credible reporting declares for both Iran and US’s attacks on ships on August 11th.

• My most likely prediction for the number of reported deaths is 0, with a 50% chance of falling between -0.50 and 3.50.
• Maritime traffic through the Strait of Hormuz has collapsed by 93% since March, significantly reducing the opportunities for lethal attacks on crews.
• Recent military engagements in May have seen high interception rates with zero confirmed U.S. or crew casualties reported by credible consensus.
• The resolution criteria prioritize the lowest reported numbers and U.S. government figures, which tends to suppress total counts in disputed ‘gray zone’ incidents.
• A significant 80% interval (up to 26.50) is maintained to account for the persistent risk of a single successful missile strike on a commercial tanker or naval vessel.
• External markets suggest a low probability of formal war (6.5%), supporting a forecast of limited, sporadic casualties rather than mass naval warfare.

Key figures

Historical context

The current conflict escalated sharply on February 28, 2026, with U.S. and Israeli strikes on Iran, followed by a naval blockade of the Strait of Hormuz. The deadliest period occurred in March 2026. On March 4, a U.S. submarine sank the Iranian frigate IRIS Dena, resulting in 87 to 104 deaths. Commercial shipping also suffered: attacks on March 1 and March 6 killed 4 people each. Throughout March and early April, the Iranian Merchant Mariners Syndicate claimed at least 44 seafarers were killed. By May, however, the conflict entered a more defensive phase. While the U.S. launched ‘Operation Project Freedom’ on May 4 to escort vessels, and Iran claimed 5 civilian deaths on May 5, major lethality has decreased as commercial traffic through the region collapsed from 2,700 monthly transits to approximately 200. This pattern suggests that while the environment remains hostile, the lack of targets and improved interceptions are lowering the daily death rate compared to the initial breakout of hostilities.

Tailwinds

• ‘Operation Project Freedom’ involves increased U.S. naval presence, raising the risk of high-casualty direct engagements.
• Iran’s continued use of ‘suicide’ boats and ballistic missiles remains a lethal threat to the remaining 1,500 stranded ships.
• The 59% chance that the U.S. blockade persists through June ensures a prolonged high-friction environment.
• Any single successful strike on a large crew-carrying vessel (like the IRIS Dena precedent) would instantly push the count toward 100.

Headwinds

• The 93% collapse in maritime traffic through the Strait of Hormuz (down to 6 transits/day) drastically reduces the target pool.
• U.S. and Israeli defensive capabilities have shown high success rates in intercepting drone and missile swarms in May.
• Resolution rules favor the ‘lowest reported figure’ in disputes, which structurally minimizes the final count.
• Diplomatic ‘off-ramps,’ including a 61% probability of a nuclear deal, may incentivize restraint as the August deadline approaches.

Detailed reasoning

My most likely prediction for the number of deaths is 0, reflecting a high probability that current diplomatic efforts or the extreme reduction in maritime traffic will prevent further lethal encounters before August 11. My analysis is structured around three primary factors: exposure, current lethality trends, and the specific rules of the resolution criteria. First, exposure has plummeted. According to PortWatch data, the Strait of Hormuz has seen a 93% collapse in vessel transits, from a baseline of 90 per day to fewer than 6. This drastically limits the opportunity for Iranian or U.S. forces to strike crew-carrying vessels compared to the high-activity period in early March.

Second, I analyzed the intensity of recent engagements. While early March 2026 was highly lethal (with events like the IRIS Dena sinking and multiple commercial fatalities), activities in May have been characterized by successful interceptions. AP and CENTCOM reports from May 7-8 indicate that recent Iranian drone and missile attacks on U.S. Navy ships were thwarted without casualties. This suggest a ‘maturation’ of defensive measures and a shift toward ‘gray zone’ tactics that prioritize disruption over mass-casualty escalation. External forecasts support this, placing the probability of a formal declaration of war at a low 6.5%.

Third, the resolution criteria exert a ‘suppression effect’ on the final count. The rules prioritize U.S. government figures and mandate using the lowest reported number in cases of dispute. Historically, Iranian claims of civilian casualties (such as the May 5 claim of 5 deaths) are often unconfirmed or disputed by U.S. officials. Under these rules, such figures would likely resolve to the lower value (often zero) unless a consensus is reached.

However, I maintain a significant upper tail in the forecast to account for the risk of a single catastrophic incident. The March 4 sinking of the IRIS Dena, which killed roughly 100 people, serves as a recent precedent for how quickly a death toll can spike if a major naval or merchant vessel is successfully hit. My 80% interval extends to 26.50 to capture the possibility of a few such successful strikes occurring over the 95-day window. Finally, a 61% chance of a nuclear deal by year-end suggests a possible diplomatic de-escalation by late summer, which would further suppress lethal activity.

Key uncertainties

• Whether Iranian military casualties (Navy/IRGC) will be counted by a consensus of western media is a major source of volatility.
• The possibility of a ‘black swan’ event, such as a large tanker sinking accidentally due to a mine, could cause a massive spike in deaths.
• Potential diplomatic breakthroughs could suddenly drop the incident rate to zero before the August resolution.
• The degree to which ‘consensus reporting’ will agree on specific numbers given the propaganda-heavy nature of the conflict.

Conclusion

• The current environment is characterized by a high-tension blockade but a 93% collapse in maritime traffic, which significantly reduces the pool of potential targets for ship crew fatalities.
• Decision-makers should expect the reported death toll on August 11 to be low (likely under 5), though they must remain prepared for a high-impact ‘tail risk’ event involving a single large vessel sinking.
• The ‘lowest consensus’ reporting rule effectively creates a high bar for counting fatalities, favoring lower official resolutions even if higher numbers are claimed.

Question: how many people will be reported dead as crew of ships attacked by the U.S. or Iran between May 8 and Aug 11, 2026. Forecasters agree the maritime theater is active but recent high-signal incidents (CENTCOM/AP, Reuters, IMO, UKMTO/JMIC) have mostly produced damage, injuries, seizures and near-misses rather than confirmed post‑May‑8 crew deaths, and the resolution rules exclude pre‑window March–early‑May fatalities. Historical analogues (1987–88 Tanker War/Operation Praying Mantis, 2019 Gulf of Oman sabotage, 2021 Mercer Street, 2023–24 Houthi campaign) anchor the view that months of harassment often yield zero or low single‑digit deaths, while a rare direct naval battle can rapidly produce double‑digit or larger tolls. Most teams place substantial probability mass on zero (roughly 35–45%), meaningful mass on 1–5 deaths from one or two lethal merchant strikes, and a long but thin tail for 10–50+ deaths if a warship is hit or sunk; numerical tail weight varies across models (some assign ~15–30% to moderate-to-high casualties, others much less). Forecasters emphasize tracking confirmed attribution and official U.S./Iran counts (IMO/AP/Reuters favored for resolution), watching missing‑crew cases that can later convert to fatalities, and treating diplomacy/Project Freedom as a partial but unproven downward force on casualties.

Forecast and Reasoning

Forecast: Probability distribution:

• 5.00% chance of value below 0.0
• 10.00% chance of value below 1.0
• 20.00% chance of value below 2.0
• 30.00% chance of value below 3.0
• 40.00% chance of value below 4.0
• 50.00% chance of value below 6.0
• 60.00% chance of value below 8.0
• 70.00% chance of value below 12.0
• 80.00% chance of value below 20.0
• 90.00% chance of value below 45.0
• 95.00% chance of value below 110.0

Forecast Reasoning

We are forecasting the total number of ship-crew deaths that (as of the Aug 11, 2026 UTC snapshot) are explicitly attributed by a consensus of major outlets to either Iranian attacks on ships or US attacks on ships, counting only deaths from May 8, 2026 onward.

Key drivers:

• Time window is short (~95 days) and the count starts at 0 on May 8; background March deaths do not count.
• The resolution rule strongly down-weights contested casualty claims (US government numbers override Iran’s; otherwise use the lowest credible figure). This makes many plausible real-world fatalities “not count” unless clearly confirmed.
• Recent incidents described by multiple analysts appear to involve frequent exchanges/interdictions but few clearly confirmed crew deaths; that supports a low median.
• There is still meaningful right-tail risk from a single rare high-fatality event (e.g., a merchant ship struck with multiple crew killed, or a naval vessel suffering major loss of life), but I judge the probability of totals near the 199 open upper bound to be well under 5% in this ~3 month window.

Analyst synthesis:

• Forecaster 1’s median (~85) and heavy >200 tail looks driven by assuming sustained high-intensity maritime losses and/or relying on dubious/irrelevant incident lists; it is also inconsistent with the consensus-filter / “use the lowest / use US numbers” rule, which should suppress totals.
• Forecasters 2 and 3 both emphasize the strict consensus criteria and recent low confirmed fatalities, giving low medians (1–5) with moderate tails.
• Forecaster 4 keeps the same median order-of-magnitude but assigns a heavier escalation tail (p95 ~115). I partially agree: the right tail should be fat, but not as central as in Forecaster 1.

Result: a low median with a pronounced right tail—most probability mass in 0–20, but with a non-trivial chance of 20–100+ if a major, clearly reported fatal incident occurs.

An amateur forecast is likely to (a) mistakenly include the March deaths mentioned in the background despite the explicit “count starting today, 2026-05-08”, (b) treat unverified/contested casualty claims as countable, ignoring the rule that US figures (or the lowest reported number) dominate in disputes, and/or (c) extrapolate linearly from earlier spikes without accounting for the strong undercounting effect of the resolution criteria. This forecast instead anchors on the post–May 8 window, explicitly incorporates the consensus/lowest-number rules (pushing the median down), while still reserving a realistic fat-tail for rare high-casualty maritime events (p90–p95).

Forecast rationale (numeric):

— Iteration 1 — Across the forecasts, the reasoning converges on a low central estimate with a heavy right tail.

Main reasoning patterns

• Most ship-attack incidents so far have produced few or no confirmed deaths.
  The dominant baseline assumption is that August 11 will most likely reflect the usual pattern of single-digit or low double-digit crew fatalities, not a mass-casualty event.

• Resolution rules push the estimate downward.
  Because the question is about what credible reporting will clearly and consensually declare, the forecasts assume that:

  • only confirmed deaths count,
  • ambiguous, disputed, or unconfirmed claims may be excluded,
  • and some incidents may end up with zero confirmed deaths if reporting remains incomplete or de-escalation occurs.

• The distribution is fat-tailed due to rare catastrophic events.
  Both rationales emphasize that although the central case is small, there is a meaningful possibility of a large naval loss or sinking that could sharply raise the total. A prior example of a major crew loss is used to justify this tail risk.

Areas of consensus

• The most likely outcome is around 10–11 deaths.
• The expected range is skewed toward low-to-moderate fatalities.
• A rare but substantial escalation could produce a much higher total.
• The answer is highly sensitive to what credible reporting confirms by the cutoff date.

Main difference in emphasis

• One forecast leans a bit more on a lower median and explicitly notes that disputed deaths may be minimized by the resolution rules.
• The other gives a slightly higher central estimate and more explicitly models a baseline monthly casualty rate plus a shock scenario.

Bottom line

The combined reasoning is that the most plausible result is a low double-digit death toll, but with a non-trivial chance of a much higher figure if a major ship loss occurs and is credibly confirmed by August 11.

— Iteration 2 — The forecasts share a common structure:

• Baseline expectation is low. Both rationales anchor on recent 2026 maritime incidents, which have mostly involved few or no deaths. This pushes the central estimate into the single digits to low teens, rather than anything large.

• Counting rules matter a lot. The estimates are dampened by the fact that only deaths that are credibly and consensually reported by Aug. 11 count, and some rationales further emphasize crew-only eligibility and the need for clear attribution. These filters reduce the number relative to raw incident counts.

• The distribution is highly skewed. While the most likely outcome is a small death toll, both forecasts keep a heavy right tail to allow for rare escalations like a ship sinking, major fire, or broader naval clash that could produce many fatalities.

• Recent pattern vs. tail risk. One rationale leans a bit higher, around the mid-teens, while the other centers lower, around the mid-single digits. Despite this difference, both agree that extreme outcomes are possible but unlikely, and both assign meaningful probability to very low totals.

Overall, the shared reasoning is: small expected death count, strong dependence on attribution/reporting rules, and a wide uncertainty range driven by rare catastrophic scenarios.

— Iteration 3 — Overall, the forecasts converge on a low confirmed death toll—roughly in the high single digits to low double digits by Aug. 11. The main reasoning is that, over a short window in an active maritime conflict, some fatal incidents are plausible but not assured, and many attacks are expected to produce no deaths or only small numbers of confirmed crew fatalities.

Key factors driving the estimates

• Short time horizon: Limited time for multiple large fatal incidents to accumulate.
• Typical incident size: Most Iran- or U.S.-linked ship attacks are expected to involve 1–3 confirmed deaths, not mass-casualty events.
• Reporting/attribution constraints: Only deaths that are credibly reported and clearly attributable are likely to count, which pushes estimates downward.
• Crew vs. passenger distinctions: Some incidents may involve people not counted under the question’s rules, reducing the final number.
• Tail risk from rare catastrophes: A major ship loss or sinking could quickly add dozens or even 100+ deaths, but this is treated as unlikely.

Areas of consensus

• The central estimate is modest, around 9–10 deaths.
• There is meaningful probability of zero or near-zero qualifying deaths.
• The distribution is right-skewed: most outcomes are small, but rare extreme events could produce much higher totals.

Main difference in emphasis

• The forecasts differ mainly in how much weight they assign to extreme tail outcomes:
  • one is more cautious in the middle with a slightly heavier tail,
  • the other centers a bit more tightly around single-digit totals.
• They do not materially disagree on the overall picture: a small confirmed total is most likely, with a low-probability but very large upside risk.

Summary of Rationales

The rationales uniformly anticipate that the final recorded death toll will be heavily suppressed by a combination of strict resolution rules, shifting military tactics, and ongoing diplomatic efforts, though they all acknowledge a distinct tail risk of a mass-casualty event.

Strict Resolution Criteria and Reporting Biases The most heavily weighted factor across all rationales is the stringent reporting criteria, which exerts massive downward pressure on the countable tally:

• Timeframe and Proxy Exclusions: The count strictly begins on May 8, 2026, excluding any prior casualties. Additionally, deaths must be explicitly attributed to the US or Iranian state militaries. This semantic barrier almost certainly disqualifies deaths caused by proxy groups like the Houthis, which media typically label as “Iran-backed” rather than direct Iranian state actions.
• Dispute Mechanics: In the event of a reporting dispute, US figures take absolute priority, followed by Iran’s, defaulting to the lowest reported number. Because the US military is highly conservative with unverified enemy body counts (often reporting “boats destroyed” instead of specific casualties) and Iran historically suppresses its own military deaths to avoid projecting weakness, many actual fatalities will be filtered out.
• Vagueness Penalty: Media or state reports using vague terms like “dozens” or “several” will add zero to the final count.

Geopolitical Trajectory and Tactical Shifts The current state of the conflict points toward reduced lethality in the maritime domain:

• Diplomatic Pauses: Both sides are currently engaged in back-channel ceasefire negotiations, leading to pauses in major offensive operations and US escort missions.
• Depleted Targets: Iran’s large surface fleet was decimated earlier in the conflict. Remaining naval assets are primarily small fast-attack crafts carrying very few crew members, inherently limiting the casualties of any single skirmish.
• Strategic Pivot: Commercial shipping in the region has plummeted. Instead of lethal kinetic strikes, Iran is shifting toward a “bureaucratic” approach of maritime extortion and bloodless boardings to avoid triggering devastating US retaliatory strikes.
• Low Base Rates: Historically, commercial maritime strikes rarely result in fatalities due to the massive size and compartmentalization of modern vessels, low crew densities, and advanced US air defenses.

Tail Risks and Escalation Scenarios Despite the strong structural drivers keeping the baseline expectation low, all rationales emphasize the need to account for a sudden spike in casualties. If the fragile ceasefire talks collapse and high-intensity blockade enforcement resumes, a severe escalation could occur. The primary tail risks include the accidental, catastrophic sinking of a fully crewed commercial tanker before the crew can evacuate, or a direct, lethal exchange resulting in the destruction of a larger US or Iranian naval vessel.

Agent Forecast (discrete, confidence: low)

• 0: 12.0%
• 2: 9.3%
• 3: 9.3%
• 1: 9.0%
• 4: 8.0%

Base Rate: IMO confirmed ~10 commercial vessel crew killed in ship attacks during the first 69 days of the 2026 Iran war (Feb 28–May 8), concentrated in early-March intense-warfare phase. Zero confirmed crew deaths from attacks during the 22-day April 8–30 ceasefire window. The May 8 starting point is under an extended ceasefire regime, making the conditional forward rate substantially lower than the cumulative-period rate.

Initial Reasoning:

• As of May 8, 2026, the 2026 Iran war is active with an extremely fragile ceasefire.
• Iran controls the Strait of Hormuz, the US is running a naval blockade, and both sides exchanged fire on May 7-8 (no confirmed crew deaths per US).
• Active peace negotiations via Pakistan on a 14-point memorandum are ongoing but repeatedly nearly collapse.
• The count starts May 8, 2026 and resolves August 11, 2026 (~95 days).
• Before May 8: ~10+ commercial vessel crew died in attacks (Mussafah 2 tug: 4 dead March 6; Mayuree Naree: 3 missing March 10-11; others: ~4 more).
• The rate was highest in early March (intense active warfare) and dropped significantly after the April 8 ceasefire.
• The resolution applies US-priority counting, which would reduce Iranian civilian death claims (US disputes the 5 civilians claimed from May 4-5 boats; US says IRGC vessels).
• For May 8-August 11: Scenario A (~25%): Peace deal reached quickly (~3-5 deaths); Scenario B (~35%): Delayed deal with ongoing incidents (~6-20 deaths); Scenario C (~25%): No deal, fragile ceasefire continues (~15-40 deaths); Scenario D (~15%): Full escalation with major incidents (40-150+ deaths).
• The modal estimate is 2-4 deaths, with a heavy right tail due to potential escalation.
• Mean estimate ~12-15.
• The distribution is right-skewed.
• The 44 Iranian seafarers killed by US attacks were largely on Iranian ships/in Iranian ports before May 8 and the US would count those at 0.
• Consensus reporting will only count specific, confirmed, mutually-acknowledged deaths.

Key Uncertainties:

• Whether the fragile ceasefire holds through August 11 or collapses into renewed naval combat
• Whether the 14-point peace memorandum is signed (would lock in low-death scenario)
• How many incidents will reach ‘consensus of credible reporting’ given US-Iran factual disputes — many real deaths may not count
• Whether any major navy vessel is sunk with mass crew casualties (low-probability, high-impact tail)
• How outlets handle the May 4-5 IRGC-vessel sinking precedent — it falls before the count window but sets a template for future US/Iran fact disputes
• Whether Iranian crew on attacked Iranian vessels get attributed to ‘US attacks on ships’ under the consensus rule

Evidence Sources (9):

• Wikipedia - List of ships attacked during the 2026 Iran war
• Eagle Intelligence - Seven Seafarers Killed as Hormuz Attacks Persist
• Al Jazeera - More than 40 Iranian seafarers killed during US-Israeli war
• Wikipedia - 2026 Iran war ceasefire
• Al Jazeera - Iran says US military killed five civilians in attacks on passenger boats
• OPB/AP - US military says it intercepted Iranian attacks on 3 Navy ships
• CNN - US and Iran closing in on agreement aimed at ending war
• BBC News - Trump’s hopes for an Iran peace deal come with caveats
• ABC Australia - What we know about renewed fighting between US and Iran

Grounding Facts (passed to reviewers):

1. As of May 8, 2026, renewed hostilities are reported between Iran and the US, with Iran attacking three US Navy destroyers, though the US claimed no assets were struck.
2. The US and Iran agreed to a 2-week ceasefire on April 8, 2026, which US President Trump extended on April 21, but it has been violated by both sides and is considered extremely fragile.
3. As of May 6-7, 2026, the US and Iran are nearing agreement on a 1-page, 14-point memorandum to end the Iran war, which includes nuclear enrichment suspension, sanctions relief, and Hormuz reopening within 30 days, with Iran currently reviewing the proposal.
4. US President Trump paused ‘Project Freedom,’ an initiative launched May 4-5, 2026, to escort ships through Hormuz, on May 6, citing “great progress” in peace talks.
5. From May 4-5, 2026, the US CENTCOM reported sinking 6 IRGC vessels; Iran claims the US attacked two civilian passenger boats, killing 5 civilian passengers, a claim the US has not confirmed.
6. From February 28 to April 1, 2026, 44 Iranian seafarers were killed, including 22 civilian sailors, 16 fishermen, and 6 dock workers, according to the IMMS.
7. The International Maritime Organization (IMO) states Iran’s attacks on vessels have killed at least 10 seafarers since the war started.
8. Pakistan’s Foreign Minister is mediating between the US and Iran, while Iran’s parliament speaker dismissed the proposed peace deal as an “American wish-list,” and the Iranian Foreign Minister stated Iran would only accept a “fair and comprehensive” agreement.
9. As of March 9, 2026, at least 7 seafarers had been killed in ship attacks, and the Joint Maritime Information Centre (JMIC) threat level was “CRITICAL.”

Draft: 3: 8.5%, 2: 8.0%, 4: 7.5%, 1: 7.0%, 0: 7.0%

Reviewer Panel:

Reviewer (reasoning)

The draft’s core weakness is its ungrounded scenario probabilities—each scenario (A–D) lacks calibration to actual incident rates under the fragile ceasefire and ignores credible recent data showing near-zero crew deaths since mid-April. It also misapplies the consensus rule by dismissing Iran’s civilian-boat casualty claims outright, despite their reporting by major outlets, and overweights pre-May incident rates without accounting for ceasefire compliance trends.

Objections:

• Ignores IMO and JMIC base rates showing zero confirmed crew deaths in attacks since the April 8 ceasefire, contradicting the assumed ongoing casualty rate.
• Dismisses Iran’s claim of 5 civilian boat passengers killed (May 4–5) despite coverage by Reuters, AP, and Al Jazeera, misapplying the “dispute” hierarchy.
• Lacks justification for heavy right tail (40–150+ deaths) post-May 8 given the two-week ceasefire extension and low operational tempo of ship attacks.

Reviewer (calibration)

The draft forecast overweights the likelihood of successful peace talks given recent statements from Iranian officials dismissing the proposed deal and the reality of ‘renewed hostilities’ reported on May 8. It also may underestimate the practical difficulty for deaths to meet the strict ‘consensus of credible reporting’ and ‘US-priority/lowest number’ resolution criteria, which will significantly dampen the final countable death toll. While actual deaths may occur, the stringent reporting requirements make the number that will ultimately resolve this question likely lower than the raw casualties.

Objections:

• The strict ‘consensus of credible reporting’ and ‘US government priority/lowest number’ rules are very likely to suppress the official death count, even if actual casualties are higher. The forecast acknowledges this but may not fully incorporate its dampening effect across the probability distribution.
• The forecast’s weighting towards quick or delayed peace deals seems optimistic given Iran’s parliament speaker dismissing the deal as an ‘American wish-list’ and the foreign minister’s conditional acceptance, alongside ongoing ‘renewed hostilities’. This suggests a higher likelihood of prolonged conflict or escalation.
• The recent event of the US sinking 6 IRGC vessels (May 4-5) where Iran claimed 5 civilian deaths and the US has not confirmed these highlights the challenge of obtaining attributed and consensus-agreed numbers under the specified resolution criteria. This suggests the probability of 0 or very low countable deaths from specific incidents, even if actual deaths occur, is higher.

Reviewer (fact-checker)

3 factual error(s). 5 unsupported claim(s). Verdict: significant_errors

Unsupported claims:

• Active peace negotiations via Pakistan on a 14-point memorandum are ongoing but repeatedly nearly collapse.
• The rate was highest in early March (intense active warfare) and dropped significantly after the April 8 ceasefire.
• The resolution applies US-priority counting, which would reduce Iranian civilian death claims
• the US would count those [44 Iranian seafarers] at 0
• Consensus reporting will only count specific, confirmed, mutually-acknowledged deaths.

Objections:

• [moderate] Iran controls the Strait of Hormuz — Corpus says: No evidence Iran controls the Strait. US launched Project Freedom May 4-5 to escort ships through Hormuz; US Navy destroyers operate there May 7-8; Iran requires ships to submit data but US counters.
• [moderate] the US is running a naval blockade — Corpus says: No evidence of US naval blockade. US launched ‘Project Freedom’ to escort commercial ships through Strait of Hormuz against Iranian attacks; intercepted Iranian attacks on US Navy ships.
• [minor] Mayuree Naree: 3 missing March 10-11 — Corpus says: Mayuree Naree attacked March 10, 2026, 3 crew members missing/believed dead.

Adjustment Rationale:

Adjusted from draft by shifting ~7 percentage points of probability mass from the 11–200+ tail into the 0–5 bin, with the largest single increase at 0 deaths (0.070 → 0.120). Two reviewer points drove this: (1) Reviewer 1 correctly noted that no crew deaths from ship attacks were confirmed during the April 8 ceasefire period, which is the regime in effect on May 8 when the count starts — my draft had over-anchored on the early-March intense-warfare base rate; (2) Reviewer 2 correctly emphasized that the strict consensus + US-priority + lowest-number resolution mechanism compounds across every incident, not just edge cases, meaning real deaths often won’t translate into countable deaths (e.g., US’s 0-civilian assertion would prevail over Iran’s 5-civilian claim in the May 4-5 IRGC vessel incident even if it had occurred after May 8). Both effects point the same direction: lower modal count. Reviewer 3’s flagged characterization errors (‘Iran controls Hormuz’, ‘US blockade’) were not load-bearing for the forecast — corrected framing here is that US Project Freedom is escorting commercial ships through Hormuz against Iranian interdiction attempts, while Iran demands data submissions; this doesn’t change the underlying rate analysis. New modal estimate: 0–3 deaths (combined ~40%); right tail preserved for escalation/ceasefire-collapse scenario but trimmed. Confidence remains low — outcome still spans orders of magnitude depending on whether the 14-point peace memo holds, whether escort/interdiction incidents produce countable crew deaths, and whether outlets converge on shared numbers by August 11.

forecast_sdk | 525.5s | research $1.0329 + reviewers $0.0139 + reconcile $0.2665 = $1.3132

Extracted Percentile Values: {0.1: 2, 1.0: 2, 5.0: 2, 10.0: 3, 15.0: 3, 20.0: 4, 25.0: 4, 30.0: 5, 35.0: 6, 40.0: 7, 45.0: 8, 46.0: 8, 47.0: 8, 48.0: 9, 49.0: 9, 50.0: 9, 51.0: 10, 52.0: 10, 53.0: 11, 54.0: 11, 55.0: 12, 60.0: 15, 65.0: 18, 70.0: 23, 75.0: 30, 80.0: 38, 85.0: 48, 90.0: 62, 95.0: 95, 99.0: 155, 99.9: 220}

Model’s Answer:

Forecast Analysis: Deaths from Iran-US Ship Attacks by August 11, 2026

Situation Assessment

Current Status (as of May 8, 2026):

• Count starting today = 0 deaths
• April ceasefire appears to be holding tenuously
• Peace talks ongoing via Pakistan mediation
• ~1,500 ships stranded, ~20,000 crew in limbo
• Recent disputed May 4 incident (Iran claims 5 deaths, US denies) - but this occurred BEFORE count start date
• Shipping transits at historic lows, reducing exposure
• No consensus deaths from ship attacks since count start

Historical Reference (not included in count):

• March 2026: ~7-10 commercial crew deaths from Iranian attacks
• These occurred BEFORE the May 8 count start date

Key Drivers and Scenarios

Low Outcome (0-5 deaths): The ceasefire continues to hold through August 11, peace talks progress, and minimal lethal incidents occur. Disputes over any incidents favor US government figures (which tends to report lower casualties). The reduction in shipping transits naturally reduces exposure. Likelihood: ~40-45%

Medium Outcome (6-25 deaths): Ceasefire holds imperfectly with some incidents. A few smaller vessels are attacked with confirmed casualties. Sporadic escalation but no full-scale resumption of hostilities. Likelihood: ~30-35%

High Outcome (26-100+ deaths): Ceasefire breaks down significantly. Multiple coordinated attacks occur, or a single large vessel (tanker with 20-40 crew) is sunk accidentally as warned in the background. Full escalation of maritime hostilities. Likelihood: ~15-20%

Very High Outcome (100-200+ deaths): Complete collapse of peace process, massive escalation, multiple large vessels sunk, or catastrophic accident involving large tanker with full crew loss. Likelihood: ~5-10%

Critical Assumptions That Could Alter Forecast

Assumption 1: Ceasefire Stability (~35% chance of failure) If ceasefire collapses entirely before August 11, casualties could spike dramatically. Current indicators suggest fragility but ongoing talks.

Assumption 2: No Accidental Large Vessel Sinking (~85% this holds) Background explicitly warns about “sudden spikes such as large tankers sunk accidentally.” A single large tanker could have 20-40+ crew.

Assumption 3: Dispute Resolution Favors Lower Counts (~80% confidence) Resolution criteria prioritize US government figures, which historically report fewer casualties than Iranian claims. This biases toward lower reported totals.

Distribution Rationale

Given:

• Current 0 deaths baseline
• Ceasefire in place (but fragile)
• 3+ months remaining until resolution
• Explicit lower bound requiring values > 1
• Significant right-tail risk from escalation/accidents
• US priority in disputes biases toward lower counts
• Default range extends to 199.5

I weight heavily toward lower values reflecting status quo continuation while maintaining a substantial right tail for escalation scenarios.

Median estimate: ~8-9 deaths - Reflects expectation of some incidents over 3 months but continued restraint 90th percentile: ~55 deaths - Accounts for moderate escalation or single large incident 99th percentile: ~150 deaths - Captures major escalation scenarios

Percentile .1: 2 Percentile 1: 2 Percentile 5: 2 Percentile 10: 3 Percentile 15: 3 Percentile 20: 4 Percentile 25: 4 Percentile 30: 5 Percentile 35: 6 Percentile 40: 7 Percentile 45: 8 Percentile 46: 8 Percentile 47: 8 Percentile 48: 9 Percentile 49: 9 Percentile 50: 9 Percentile 51: 10 Percentile 52: 10 Percentile 53: 11 Percentile 54: 11 Percentile 55: 12 Percentile 60: 15 Percentile 65: 18 Percentile 70: 23 Percentile 75: 30 Percentile 80: 38 Percentile 85: 48 Percentile 90: 62 Percentile 95: 95 Percentile 99: 155 Percentile 99.9: 220

Chokepoint geometry and reduced commercial traffic constrain ordinary exposure to lethal incidents.

Sustained naval postures and interdiction leave few unprotected high-value targets but raise risk of state-to-state escalation.

Ongoing missile, drone, and small-boat operations produce frequent incidents but few confirmed crew deaths to date.

Defensive interceptions, convoying, and avoidance behavior materially depress the central death count.

A single successful strike on a large tanker, a minefield encounter, or a misidentified naval attack would shift outcomes toward mass fatalities.

Diplomatic freezes, routings, and enhanced escorting would keep additional deaths at or near zero.

Key uncertainties include opaque casualty reporting, missing crew versus confirmed dead, and divergent state narratives.

The distribution is skewed: low modal outcomes coexist with a thin but consequential tail sensitive to one-off catastrophic events.

Forecast: Median deaths = 6.4624

• Diplomatic Outcome: The potential for a peace deal is the most significant factor influencing the number of deaths. If a deal is reached, it could drastically reduce attacks and fatalities.

• Tactical Choices: Iran’s current strategy focuses on low-lethality tactics like boarding and seizing vessels. A shift to sinking ships would significantly increase casualties.

• Conflict Intensity: The frequency of attacks during the forecast window is critical. Continued high levels of aggression could lead to higher death counts, while a decrease in incidents would lower them.

• Accidental Incidents: The presence of trapped vessels and mines poses a risk for accidental strikes, which could lead to mass casualties even without direct attacks.

• Media Attribution: The requirement for consensus among credible media sources to attribute deaths to US or Iranian actions may limit the reported fatalities, especially in disputed incidents.

• Historical Context: Previous maritime conflicts provide some insight, but the unique dynamics of this situation make precise predictions challenging. The potential for a catastrophic event remains a significant risk.

Forecast (percentiles, aggregated):

• p10: 0.0
• p25: 1.0
• p50: 4.0
• p75: 20.0
• p90: 85.0

Per-model forecasts:

• 1: p25=1, p50=5, p75=20, p90=60
• 2: p25=2, p50=7, p75=25, p90=90
• 3: p25=1, p50=3, p75=15, p90=85

Rationale: Summary The estimate for the total number of crew deaths resulting from US and Iranian ship attacks between May 8 and August 11, 2026, reflects a heavily right-skewed distribution. The baseline expectation—anchored by a median of 4 deaths—assumes that the current, albeit fragile, ceasefire largely holds and that any skirmishes primarily involve merchant vessels. Recent dynamics, including a “dual blockade” and the US Operation Project Freedom, have resulted in tense encounters but very few confirmed fatalities since early April. In this environment, intermittent and low-level incidents might claim a small number of casualties, as individual merchant vessel attacks typically result in zero to three deaths. However, there remains a substantial tail risk, pushing the 90th percentile up to 85. This accounts for the possibility of a major escalation or the sinking of a large vessel, echoing the March 2026 sinking of the IRIS Dena, which killed over 80 sailors. If the ceasefire collapses or a high-casualty strike occurs, the death toll could rapidly spike.

Strongest Arguments for Higher Values

• The “dual blockade” and ongoing Operation Project Freedom create a highly volatile environment where miscalculation is likely, raising the probability of a major naval engagement.
• Historical precedent demonstrates extreme tail risks; the March sinking of the IRIS Dena alone resulted in 84 to 104 fatalities, proving that a single catastrophic event could heavily skew the final count.
• Recent skirmishes in early May, despite causing no confirmed deaths so far, indicate that active hostilities and projectile attacks are continuing despite the ceasefire.

Strongest Arguments for Lower Values

• A tenuous ceasefire has been in place since April 8, with ongoing diplomatic efforts and a one-week deadline for a peace proposal that may lead to broader de-escalation.
• US defensive interceptions of recent Iranian attacks have proven highly successful, effectively preventing crew casualties during recent maritime clashes.
• Attacks on commercial and merchant shipping typically result in seizures or material damage rather than mass casualties, inherently limiting the death toll unless a heavily crewed naval vessel is destroyed.

Key Uncertainties

• The stability of the current ceasefire and the outcome of the impending peace proposal deadline, which will dictate whether the region transitions to peace or returns to active warfare.
• The specific nature of future targets; attacks concentrated on merchant ships will likely keep numbers low, whereas strikes on populated military vessels could cause sudden, massive casualty spikes.
• The strict resolution criteria, which require consensus from credible media outlets and prioritize US government figures over Iranian claims, potentially resulting in recognized figures that are significantly lower than broader estimates or unconfirmed reports.