https://acleddata.com/report/russias-shadow-fleet-presents-sustained-hybrid-war-threat-sea

How many Russian shadow fleet vessels will be intercepted by EU countries between June 4 and August 12, 2026?

• My most likely prediction for Russian shadow fleet vessels intercepted by EU countries is 2.50, with a 50% chance of falling between 1.50 and 4.50.
• The forecast is anchored in a 2026 base rate where European nations like France, Sweden, and Belgium have averaged approximately 1.5 to 2.0 interceptions per month.
• Enforcement activity is expected to remain steady or slightly increase due to the upcoming 21st EU sanctions package, which targets 20 additional tankers.
• Tactical shifts by Russia, specifically the reflagging of vessels to its own registry, act as a primary constraint on the number of legally defensible physical boardings.
• Regional caution, particularly from Baltic states like Estonia fearing military escalation, suggests that most interceptions will likely occur in the Atlantic or North Sea.
• The 80% highest density interval (0.50 to 6.50) reflects the volatility of these events, which often occur in clusters or depend on specific diplomatic triggers.

Key figures

Historical context

The ‘shadow fleet’ emerged as a significant geopolitical entity following the G7 price cap on Russian oil in December 2022. Historically, enforcement was passive, focusing on financial services and insurance. However, 2026 has marked a shift toward ‘kinetic’ or physical enforcement. In January 2026, France seized the tanker Grinch in the Alboran Sea, marking a new phase of intervention. This was followed by the Belgian seizure of the Ethera in February and multiple Swedish detentions in the Baltic. Between December 2025 and June 2026, approximately 12 vessels were intercepted by European countries, signaling an average rate of 2 vessels per month. The legal framework has shifted toward utilizing UNCLOS Article 92 (exclusive jurisdiction of flag states) and Article 110 (right of visit) by targeting vessels with fraudulent registries.

Tailwinds

• The EU is finalizing a 21st sanctions package for mid-June 2026, which will specifically list ~20 more tankers, making their status as ‘shadow’ vessels easier to verify.
• Countries like the Netherlands are actively developing new legislation to double penalties and expand maritime intervention powers for shadow fleet activities.
• Increased NATO presence through initiatives like ‘Baltic Sentry’ provides the surveillance and intelligence support necessary for successful boarding operations.

Headwinds

• Over 70 shadow fleet vessels have reflagged to the official Russian registry since mid-2025, stripping authorities of ‘false flag’ or ‘statelessness’ pretexts for boarding.
• Estonia and other Baltic states have expressed reluctance to detain vessels due to the risk of military escalation and the presence of Russian naval escorts.
• Sanctions status alone does not provide a legal basis for physical boarding under UNCLOS in international waters, limiting interceptions to specific documentation or safety violations.

Detailed reasoning

My analysis is built upon an empirical baseline of maritime enforcement actions observed in the first half of 2026. Data from ACLED and official maritime records indicate that between January and early June 2026, there were approximately 8-12 documented physical interceptions or detentions of Russian shadow fleet vessels by EU or partner states (including the French seizure of the Grinch and Tagor, the Swedish detention of the Jin Hui, and Belgian seizure of the Ethera). This establishes a historical base rate of roughly 1.5 to 2.0 vessels per month. Given that the forecast window covers 69 days (approximately 2.3 months), a direct extrapolation of this ‘status quo’ rate yields a baseline expectation of 3.5 to 4.6 interceptions.

However, I have adjusted the most likely value slightly lower to 2.50 to account for several critical frictions. First, the definition of the ‘shadow fleet’ is increasingly tied to vessels using ‘false flags’ or ‘stateless’ designations to evade sanctions. In response to recent EU enforcement, Russian operators have increasingly reflagged vessels to the official Russian registry—over 70 such moves were documented in early 2026. This tactical shift removes the primary legal pretext (statelessness under UNCLOS) that European navies have used for physical boardings in international waters. Second, some key regional actors like Estonia have explicitly stated they will refrain from detaining tankers to avoid military escalation with Russian escorts.

Conversely, the forecast accounts for a potential ‘tail’ of higher activity driven by the upcoming 21st EU sanctions package, which is expected to target an additional 20 vessels. While sanctions alone do not grant boarding rights, they increase the pool of ‘verifiable’ shadow vessels and may embolden member states like France and Sweden, which have led enforcement efforts. The 50% HDI of 1.50 to 4.50 reflects the high probability that enforcement continues at its current irregular pace, while the 80% HDI stretching to 6.50 accounts for a potential surge in interventions following new legislative mandates or coordinated naval patrols during the summer months.

Key uncertainties

• A sudden shift in Russian naval behavior, such as providing armed escorts for more tankers, would likely halt EU physical interventions.
• The specific legal grounds for the first physical attempt (e.g., environmental risk vs. false flag) can be contested and may vary by member state jurisdiction.
• Reporting lags mean that a vessel boarded in August might not be publicly labeled as ‘shadow fleet’ by the August 19 deadline.

Conclusion

• The most likely number of interceptions is between 1.50 and 4.50, reflecting a continuation of the 2026 ‘status quo’ where roughly one vessel is detained every 15-20 days.
• Decision-makers should monitor the finalized 21st sanctions package in mid-June; a significant expansion of the list or a new legal mandate for boarding could shift the outcome toward the upper end of the 80% interval.

Forecast and Reasoning

Forecast: Probability distribution:

• 1.00% chance of value below 0.0
• 5.00% chance of value below 0.6
• 10.00% chance of value below 1.2
• 20.00% chance of value below 2.0
• 30.00% chance of value below 2.6
• 40.00% chance of value below 3.4
• 50.00% chance of value below 4.2
• 60.00% chance of value below 5.0
• 70.00% chance of value below 5.8
• 80.00% chance of value below 6.8
• 90.00% chance of value below 8.6
• 95.00% chance of value below 11.2
• 99.00% chance of value below 17.8

Forecast Reasoning

We are forecasting the count of unique Russian “shadow fleet” vessels that experience a first EU-involved physical attempt to stop/board/detain during a 70-day window (2026-06-04 to 2026-08-12), with the vessel’s shadow-fleet status needing to be publicly verifiable by 2026-08-19.

Across the analyst set, three forecasts cluster around a median of ~3–4, primarily extrapolating from ACLED-style reporting of several EU enforcement actions in early 2026 (roughly ~1–1.5 per month) and treating the process as Poisson-like with uncertainty in the rate. One analyst argues for a much higher median (~6–7) based on claimed additional early-June boardings and an anticipated policy-driven surge; however, those specific incremental facts were not corroborated by the other analysts’ research summaries, and (crucially) some cited actions fall outside the window anyway, so I discount that forecast but still take from it the possibility of a fatter upper tail if a coordinated campaign occurs.

Synthesis: a reasonable central tendency is ~4 interceptions in-window (continuation of ~1.5–2/month tempo), with meaningful downside risk (0–1) if escalation concerns or rerouting reduce boardings, and an upper tail (10–18) if multiple EU states run intensified, multi-incident enforcement in chokepoints or ports during the window. I assign negligible probability to exceeding the open upper bound (25) in just 70 days absent an extreme policy shift, so I keep all percentiles in-range.

An amateur forecast would likely (a) straight-line extrapolate a short recent streak into an implausibly high count, or (b) anchor on legal/operational difficulty and predict 0–1, underweighting the demonstrated 2026 enforcement tempo. This forecast improves by: (1) using the pre-window 2026 incident rate as the main base rate, (2) explicitly treating enforcement as policy- and risk-constrained (limiting the mean), and (3) retaining a heavier right tail than a single-rate Poisson to reflect plausible campaign/surge scenarios. Confidence is moderate: the main uncertainty is political/operational choice rather than vessel availability.

Forecast rationale (numeric):

— Iteration 1 — Across the forecasts, the reasoning is broadly similar:

• Base-rate anchoring: Each model starts from recent early-2026 enforcement activity against the shadow fleet and extrapolates that pace across the June 4–August 12 window. That produces a baseline of roughly 3–4 interceptions over the period.
• Modest upward adjustment: All three forecasts then raise that baseline somewhat because of recent EU policy momentum, especially the 20th sanctions package and discussion of a more assertive interpretation of UNCLOS that could enable more boarding or interception attempts.
• Operational support factors: The models also note that summer conditions in the Baltic and broader seasonal activity could make interceptions somewhat more feasible.
• Constraining factors: Despite the policy shift, they emphasize substantial limits from legal ambiguity, escalation risk with Russia, the presence of Russian escorts, and the possibility that the event definition requires a relatively clear physical attempt or stop.
• Consensus on shape of the forecast: The central expectation is modest, not dramatic, with most probability mass around 3–5 vessels.
• Main disagreement is tail risk: The forecasts differ mainly on how much weight to give the extremes. All allow for a low-end outcome of 0–1 if enforcement is cautious or definitions are restrictive, and a right tail where a coordinated enforcement push or policy shift could produce a much larger spike, potentially double digits.

Overall, the collective reasoning points to a small-to-moderate number of interceptions, driven by incremental policy tightening rather than a full-scale enforcement campaign.

— Iteration 2 — Across the forecasts, the core reasoning is that EU interceptions of Russian shadow-fleet vessels are likely to remain modest but nontrivial, with a central expectation around 3–5 vessels over the June 4 to August 12 window and a median near 4.

Main factors driving the estimate

• Recent enforcement pace as the anchor: The forecasts rely heavily on early-2026 observed activity, roughly 1.25 to 1.6 interceptions/actions per month, which scales to a few vessels over a ~2.3-month period.
• Evidence of slight acceleration: Several rationales note that May and early June incidents suggest enforcement may be picking up, nudging the estimate above a straight extrapolation from earlier months.
• Operational and legal constraints: Interceptions are treated as costly, legally constrained, and episodic, limiting the chance of a large sustained campaign.
• Russian countermeasures: Reflagging, escorts, and evasive tactics are expected to suppress the total and help explain why zero remains plausible.
• Potential for cluster events: The upper tail is driven by the possibility of a coordinated EU crackdown, a policy shift, or incident-driven bursts of enforcement.

Areas of consensus

• Most likely outcome: A small single-digit number, centered on about 4 vessels.
• Zero is possible but not dominant: There is meaningful probability of no interceptions if enforcement stalls or is deterred.
• Right-skewed uncertainty: The distributions are broadly right-tailed, reflecting occasional spikes rather than a steady high rate.

Main differences

• How strongly to extrapolate recent acceleration: Some forecasts are more cautious, staying close to the observed rate; others push slightly higher based on recent momentum.
• How wide to make the tail: One view keeps the upper tail moderately broad, while another allows a much wider extreme range if there is a disruptive enforcement surge.

Overall synthesis

The shared logic is that EU countries are likely to continue intermittent, limited interception activity rather than launch a major sustained campaign. The best estimate clusters around 4 vessels, with outcomes mostly in the low single digits, but with a meaningful chance of 0 and a smaller chance of a burst of activity pushing the count notably higher.

— Iteration 3 — The forecasts mostly converge on a low-single-digit expectation, centered around 3–4 interceptions over the June 4 to August 12 window.

Core reasoning pattern

• Historical enforcement pace is the anchor: recent early-2026 activity is treated as the best base rate, roughly 1.25–1.4 actions per month, which translates to about 3 interceptions over the forecast period.
• Slight upward adjustment is applied because:
  • enforcement appears to be gradually accelerating,
  • summer conditions may make maritime operations easier,
  • and there is policy momentum in the EU toward stronger action.

Main upward drivers

• A coordinated EU crackdown or expanded boarding mandate could raise the count quickly.
• A triggering incident such as cable damage, sabotage, or a spill could prompt multiple interceptions in a short period.
• The distribution is therefore treated as right-skewed, with a meaningful chance of a larger-than-normal spike.

Main downward constraints

• Legal and operational limits on physically intercepting vessels remain significant.
• Russian evasive behavior, including escorts, route changes, or deterrence, could suppress activity.
• Some forecasts also allow for enforcement pauses or a broader collapse in operational tempo, keeping the lower end near zero.

Overall consensus and uncertainty

• The clearest consensus is on a base case near 3–4 vessels.
• The main disagreement is not over the center, but over the size of the tail:
  • some forecasts keep the upper range modest,
  • while others assign substantial probability to a surge scenario with many more interceptions if circumstances change.

In short, the collective reasoning expects a small number of interceptions, but with asymmetric upside risk if EU policy or a triggering event leads to a crackdown.

Summary of Forecasting Rationales

Historical Baseline Forecasters ground their reasoning in the established run rate of physical interceptions observed in early 2026. Led primarily by France, Sweden, and Belgium, EU coastal states have demonstrated a steady, consistent rhythm of targeted enforcement actions against the Russian shadow fleet.

Factors Driving Interceptions (Upward Pressures) Several catalysts are expected to sustain or accelerate this baseline during the summer window:

• Favorable Weather: Calmer summer waters and extended daylight in the Baltic and North Seas significantly reduce the operational risks of complex naval boardings via helicopters or interceptor boats.
• Expanded Legal Frameworks: The rollout of the EU’s 21st sanctions package expands the pool of blacklisted targets. Furthermore, EU states are increasingly willing to leverage UNCLOS provisions—citing statelessness, lack of legitimate insurance, and severe environmental hazards—to justify interdictions.
• Enhanced Surveillance: Ongoing multinational military exercises (such as BALTOPS) and joint maritime operations provide high-density patrols and robust intelligence-sharing networks that make tracking non-compliant vessels highly efficient.

Factors Constraining Interceptions (Downward Pressures) Despite high political will, forecasters universally note severe headwinds that prevent mass detentions:

• Russian Deterrence and Tactics: Russia has escalated its posture by deploying armed naval escorts alongside valuable shadow fleet convoys, heavily deterring EU forces from risking armed confrontation. Additionally, shadow operators are actively rerouting away from European chokepoints and increasingly reverting to official Russian flags to claim sovereign immunity, neutralizing the “stateless” loophole.
• Severe Logistical Bottlenecks: EU nations lack the deep-water port capacity, specialized crews, and financial appetite to indefinitely impound dozens of aging, uninsured, and structurally deficient supertankers. This imposes a hard ceiling on the sheer number of vessels that can be safely processed.

Synthesis Ultimately, the rationales agree that opposing forces are at play. Heightened EU political resolve and optimal summer boarding conditions are counterbalanced by severe logistical limits and the growing risk of military escalation with Russian escorts. Consequently, forecasters anticipate a continuation of the established, targeted enforcement rhythm rather than a dramatic surge in mass detentions.

SUMMARY

Question: How many Russian shadow fleet vessels will be intercepted by EU countries between June 4 and August 12, 2026? Final Prediction: Probability distribution:

• 10.00% chance of value below 0.89
• 20.00% chance of value below 1.54
• 40.00% chance of value below 2.593333
• 60.00% chance of value below 3.55
• 80.00% chance of value below 4.95
• 90.00% chance of value below 6.25

Total Cost: extra_metadata_in_explanation is disabled Time Spent: extra_metadata_in_explanation is disabled LLMs: extra_metadata_in_explanation is disabled Bot Name: extra_metadata_in_explanation is disabled

Report 1 Summary

Forecasts

Forecaster 1: Probability distribution:

• 10.00% chance of value below 1.25
• 20.00% chance of value below 2.05
• 40.00% chance of value below 3.35
• 60.00% chance of value below 4.45
• 80.00% chance of value below 6.05
• 90.00% chance of value below 7.65

Forecaster 2: Probability distribution:

• 10.00% chance of value below 0.8
• 20.00% chance of value below 1.35
• 40.00% chance of value below 2.35
• 60.00% chance of value below 3.35
• 80.00% chance of value below 4.65
• 90.00% chance of value below 5.8

Forecaster 3: Probability distribution:

• 10.00% chance of value below 0.62
• 20.00% chance of value below 1.22
• 40.00% chance of value below 2.08
• 60.00% chance of value below 2.85
• 80.00% chance of value below 4.15
• 90.00% chance of value below 5.3

Research Summary

The research compiles news and tracking data through early June 2026 showing multiple EU-involved interceptions of suspected Russian “shadow fleet” tankers in 2026 (examples cited: France’s interception of Grinch in January, Belgium with French support seizing Ethera on 28 Feb, Sweden detaining Jin Hui in May, and France/UK intercepting Tagor on 1 June). It identifies an approximate 2026 observed rate of about 1.0–1.6 EU interception actions per month (roughly 5–8 documented EU actions across ~5 months), and uses that as a base rate to project a baseline of about 2–4 interceptions for the 69-day period 4 June–12 August 2026. The research also notes contextual developments that could raise interception probability (UK authorization to board vessels in late March 2026, France increasing penalties in April 2026, appointment of an EU “shadow fleet” coordinator, and the BALTOPS naval exercise in the Baltic on 4–20 June) and constraints that could limit actions (UNCLOS/legal limits on boarding and detention, frequent release of vessels after fines, Russian countermeasures such as escorts and reflagging). The research further reports that roughly 564–632 vessels are under EU sanctions and that the Baltic, Mediterranean and Atlantic are key geographic hotspots (the Baltic accounting for a large share of Russian maritime oil flows). It also notes no specific, liquid prediction-market markets were found tracking this exact interception question.

Sources used (websites/organizations referenced in the research):

• ACLED (Armed Conflict Location & Event Data Project): https://acleddata.com/
• NATO / BALTOPS information: https://www.nato.int/ and BALTOPS reporting pages (NATO / participating navies)
• BALTOPS / U.S. Navy and exercise pages (reports about the June 4–20, 2026 exercise)
• Reuters (international news reporting): https://www.reuters.com/
• Major national government sites referenced for policy/announcements (examples): UK government https://www.gov.uk/ ; French government https://www.gouvernement.fr/
• EU institutions (Council / Commission) reporting on sanctions and coordination: https://www.consilium.europa.eu/ ; https://ec.europa.eu/
• Polymarket (prediction-market platform) — general site searched for markets: https://polymarket.com/
• Metaculus (forecasting platform) — general site searched: https://www.metaculus.com/
• Manifold Markets (prediction-market/forecasting site): https://manifold.markets/
• UNCLOS text / United Nations Division for Ocean Affairs and the Law of the Sea: https://www.un.org/Depts/los/convention_agreements/texts/unclos/unclos_e.pdf

(Where the research cited specific news incidents, it drew on multiple contemporary news reports and event-tracking summaries from the outlets and institutions listed above.)

RESEARCH

Report 1 Research

I’ll help you gather comprehensive information to assist with this forecasting question. Let me conduct several searches to find relevant news, prediction markets, base rates, and historical context about Russian shadow fleet interceptions by EU countries.

Forecast — Post 421

Modal: 3-6 (45.0%) • partial_shift • 8m3s

Interpretation

Reading: strict

The resolution criteria are highly specific and narrowly defined: they set an exact UTC time window, require a precise type of ‘first physical attempt’ (excluding passive monitoring or ignored verbal orders), demand EU member state participation, and require publicly verifiable shadow fleet status from named qualifying sources by a hard deadline. This narrow, literal specification favors a strict reading with little room for charitable inclusion of borderline cases.

Time bounds

• observation_window: Events with first physical interception attempt between 2026-06-04T00:00:00Z and 2026-08-12T23:59:59Z inclusive; shadow fleet status must be publicly verifiable by 2026-08-19T23:59:59Z
• resolution_date: 2026-08-12

Edge cases identified

1. The June 1, 2026 interception of the tanker Tagor falls just BEFORE the June 4 start date and should be excluded — only interceptions on or after June 4 qualify.
2. Joint operations: if a non-EU country (e.g., UK, Norway) leads an interception but an EU state participates, it counts; if no EU member state participates, it does not count.
3. A vessel intercepted in the window but not confirmed as shadow fleet until after 2026-08-12 (but before 2026-08-19 23:59:59 UTC) still qualifies due to the reporting lag provision.
4. A vessel on an EU/US/UK sanctions list for Russian oil transport automatically satisfies the identification standard even without explicit ‘shadow fleet’ labeling by a tracker.
5. A vessel previously shadow fleet but explicitly noted as having ceased those operations prior to the interception does not count.
6. Multiple interception attempts on the same vessel count only once (unique vessels, not unique interception events).
7. Heave-to orders that are ignored without further physical follow-up do not constitute a qualifying interception; only obeyed heave-to orders or active physical maneuvers qualify.
8. The word ‘intercepted’ in the question title is slightly narrower than the resolution criteria which also include ‘detained’ — the criteria govern, so detention ashore or in port counts if the physical stop occurred in the window.
9. Vessels intercepted in international waters vs. EU territorial waters: the criteria do not restrict by location, so both qualify as long as an EU state participates.
10. If a tracker labels a vessel as ‘suspected shadow fleet’ post-interception (before the Aug 19 deadline), it qualifies — the labeling need not be contemporaneous with the interception.

Research (3/3 variants, shared evidence pool)

Total evidence registered (shared pool): ?

Research Brief

Evidence confidence: medium

Scenario 1: Low count (0–2 vessels): enforcement suppressed by Russian military escort and legal/political barriers [medium evidence]

Conditions favoring

Russia’s deployment of the Admiral Grigorovich and naval escorts deters EU boarding attempts; absence of EU-level legal authority and UNCLOS constraints limit member states to rare opportunistic interceptions; shadow fleet operators reroute through less-patrolled corridors; political caution about escalation prevails through summer 2026.

Conditions against

The accelerating 2026 pace (5 in Jan–April) and multiple member states independently pursuing enforcement makes complete suppression unlikely; France’s track record of quarterly interceptions suggests at least some actions; the ‘stateless vessel’ legal opening for 16 specific ships has already been established.

Scenario 2: Moderate count (3–5 vessels): continuation of 2026 pace, consistent with arrival projection mean [high evidence]

Conditions favoring

The computed arrival projection (source 12) centers on mean ~3.45 vessels over 2.3 months at ~1.5/month; the established 2026 enforcement momentum continues; multiple EU member states (France, Sweden, Belgium, potentially others) each contribute 1–2 interceptions; Russian naval escorts only deter a fraction of potential interceptions.

Conditions against

Russian military countermeasures intensify; legal challenges following Tagor precedent slow subsequent operations; political negotiations reduce enforcement appetite in summer 2026.

Scenario 3: Higher count (6–10 vessels): accelerating enforcement beyond historical pace [medium evidence]

Conditions favoring

Sweden’s alone having 5 actions by May suggests a higher underlying rate than 1.5/month; multiple EU states each conducting 2–3 interceptions over a 10-week window; new legal instruments or political catalysts (e.g. Russia-Ukraine developments) accelerate enforcement; EU coordination formalizes, closing the ‘episodic’ gap.

Conditions against

Russian military escorts and escalation concerns constrain operations; legal basis remains narrow (stateless vessel doctrine limited to ~16 ships); summer diplomatic activity could moderate enforcement; historical data does not support sustained rates above 3/month across all EU states combined.

Scenario 4: Zero interceptions: complete enforcement halt [low evidence]

Conditions favoring

Russia’s naval escort strategy fully deters EU boarding attempts; EU states face a diplomatic or legal moratorium on shadow fleet interceptions (e.g., negotiated settlement with Russia, legal challenge to Tagor-type operations); all planned operations are stood down due to escalation concerns.

Conditions against

No evidence of any diplomatic process that would halt enforcement; France and Sweden have demonstrated willingness to act independently; the 2026 enforcement trend shows no signs of reversal as of early June; even 2025’s lower pace still yielded 3 interceptions.

Key uncertainties

• Whether Russia’s Admiral Grigorovich and naval escorts will physically deter EU boarding attempts or merely create political complications
• Whether additional EU member states beyond France, Sweden, and Belgium will conduct interceptions in this window
• Whether any new EU-level coordination framework or legal instrument emerges to systematize enforcement during summer 2026
• How shadow fleet operators adapt routing in response to the Tagor interception (e.g., shifting from Atlantic/English Channel to alternative routes that avoid EU patrol zones)
• Whether the pace established by Sweden (5 interventions alone by May 2026) continues into summer or was a concentrated burst driven by specific vessels
• Whether Russian ceasefire/diplomatic dynamics materially affect EU willingness to enforce maritime pressure through summer 2026

Forecast variants (3/3)

Variant 0: inside_view (inside_view_v1)

Inside View: Causal Chain Analysis

Current state → mechanism → outcome:

The causal chain is: accelerating EU enforcement momentum (5 ACLED actions Jan-Apr 2026, Sweden alone 5 by May) → sustained operational capability across France/Sweden/Belgium + legal precedents established (Tagor, Ethera, Grinch/Caffa) → expected ~1.5-2 interceptions/month over the 2.3-month window → projected 3-5 qualifying interceptions.

Weighting by evidence strength:

• Scenario “Moderate count (3–5 vessels)” tagged relative_evidence_strength: HIGH — this is the arrival-projection-backed scenario. Mean 3.45, median 3, p5=0, p25=1, p75=5, p90=8 from the computed model. This scenario gets dominant mass.

• Scenario “Low count (0–2 vessels)” tagged relative_evidence_strength: MEDIUM — Russian military escort (Admiral Grigorovich escorted 12+ tankers by late May), legal fragility, “episodic and politically fragile” enforcement. This is a real suppressor but not enough to override the trend.

• Scenario “Higher count (6–10 vessels)” tagged relative_evidence_strength: MEDIUM — Sweden alone had 5 by May suggesting ACLED undercounts; p75=5, p90=8 from arrival model. Possible but not the center.

• Scenario “Zero interceptions” tagged relative_evidence_strength: LOW — essentially impossible given France’s quarterly pace and Sweden’s 5-interception 2026 track record.

Key structural factors:

• France: ~quarterly pace (~1 per 3 months) → 0-1 expected in 2.3 months
• Sweden: 5 in ~4 months = ~1.25/month → ~2-3 expected over 2.3 months
• Belgium: 1 confirmed 2026 action → sporadic
• Combined rate plausibly 1.5-3/month, but Russian escort deterrence moderates
• “Stateless vessel” legal basis established for 16 specific ships → concrete targets exist
• No EU-level coordination framework = episodic, not systematic

Distribution construction:

• Floor at 0 (confirmed current state June 5)
• Arrival model: mean ~3.45, median ~3, p5=0, p25=1, p75=5, p90=8, p95=9
• Given HIGH-evidence scenario centers on 3-5, I put substantial mass there
• Medium-evidence low scenario (0-2) gets meaningful mass from military deterrence
• Medium-evidence high scenario (6-9) gets moderate tail mass
• Very little mass above 9 (p95 level)

Bin allocations:

• 0-3 (bin_0): ~35% — current floor is 0, low scenario (0-2) plus some of the 3-count from moderate scenario
• 3-6 (bin_1): ~35% — core moderate scenario mass (3-5 range dominates)
• 6-9 (bin_2): ~18% — higher scenario supported by Sweden’s pace, p75-p90 range
• 9-13 (bin_3): ~7% — above p90, tail mass
• bin_4+ (13+): ~5% combined — very thin tail

I’ll allocate: bin_0=0.35, bin_1=0.35, bin_2=0.18, bin_3=0.07, bin_4=0.03, bin_5=0.01, bin_6=0.005, bin_7=0.005, bin_8=0.00 → sum = 1.00

Variant 1: outside_view (outside_view_v1)

Reference class: counts of EU-member-state interceptions of Russian ‘shadow fleet’ vessels over multi-month windows. Base rate: using ACLED and country reports the brief documents ~5 enforcement actions in Jan–Apr 2026 (source: ACLED) → ~1.25/month, which implies ~2.9–3.5 interceptions over the ~2.3-month (June 4–Aug 12) window (the brief’s arrival-model gives mean ≈3.45, median 3; p25=1, p75=5) (brief source: arrival_forecast tool / ACLED). Adjustment: I weigh that base rate upward in probability mass for the 3–5 outcome-range because (a) multiple member states (France, Sweden, Belgium) have shown willingness to board/seize (France’s Tagor and prior seizures; Sweden recorded 5 interventions by early May) and (b) the brief flags a computed arrival projection tagged as strong quantitative evidence (scenario labelled “Moderate count (3–5)” marked relative_evidence_strength = high). I also widen tails because the evidence quality is medium and because Russia’s deployment of naval escorts (Admiral Grigorovich escorting a dozen+ tankers, brief sources 10/14/15) and legal/political frictions create a meaningful downside risk to enforcement. Weighting by scenario evidence strength (the brief’s moderate/’high’ scenario should carry materially more mass than the medium/low scenarios) yields the distribution below.

How the probability mass maps to bins (brief evidence cited inline):

• 3–6 (bin_1): 60% — centers the high-evidence arrival-projection (mean ≈3.45, median 3) and reflects the brief’s high-strength scenario (“Moderate count (3–5)”) supported by ACLED + country-level actions (France, Sweden, Belgium) (brief sources: arrival_forecast tool; ACLED; country reports).
• 0–3 (bin_0): 25% — captures the credible suppression case (brief’s “Low count (0–2)” medium-evidence scenario) driven by Russian naval escorts and legal/political barriers (Admiral Grigorovich escorts, UNCLOS/limited EU interdiction authority) (brief sources: Atlantic Council / iNews / PRIF / European Commission briefing).
• 6–9 (bin_2): 13% and 9–13 (bin_3): 2% — longer right tail for faster-than-expected acceleration (brief’s “Higher count (6–10)” medium-evidence scenario, supported by Sweden’s concentrated early-2026 activity and the arrival-model p75–p95 tail: p75=5, p90~8, p95~9).
• bins 4–8: 0% — outcomes above ~13 are extremely unlikely given current run-rates, legal constraints, and the brief’s p95 ≈9.

Calibration notes: I treated the brief’s computed arrival-process projection as strong quantitative evidence (per the instructions) and therefore concentrated most mass near its mean/median while respecting structural downside risks cited in the brief (naval escorts, legal limits). Evidence confidence = medium in the brief, so I preserved nontrivial probability in both lower and higher tails rather than overconcentrating on a single bin. This distribution reflects an outside-view anchoring on the historical multi-month interception rate (~3 per 2.3-month window) with modest adjustments for political/legal deterrents and the brief’s scenario evidence strengths (I gave the brief’s single ‘high’ scenario the largest share of mass, and weighted the two medium scenarios materially less).

Variant 2: contrarian_v2 (contrarian_v2)

Adversarial reading

The bulk of the evidence (ACLED counts, Sweden’s 5 interventions by May, France’s quarterly pace, and the arrival-process projection) pulls toward a moderate outcome centered on 3–5 vessels intercepted in the ~2.3-month window. The computed mean of 3.45 and the high-evidence-strength scenario both reinforce this cluster.

Counter-evidence found (or absence)

Russia’s deployment of the Admiral Grigorovich and sustained naval escorts (12+ tankers by late May) plus the absence of any EU-level interdiction authority and the explicitly “episodic and politically fragile” character of enforcement constitute substantive counter-evidence that the 2026 pace may not be sustained once Russian countermeasures are fully operational. The zero-interception floor as of 5 June and the lack of any new legal instrument post-Tagor further support the possibility that the arrival model over-weights the pre-June momentum.

Distribution implications

The moderate-count bin (roughly 3–6) still receives the largest single mass because the quantitative arrival projection and multiple member-state track records remain the strongest signals, but the Russian-escort and episodic-enforcement counter-evidence shifts additional probability into 0–3 (bin_0) relative to a naive centering on the 3.45 mean. Higher bins (6+) receive only modest mass because the structural barriers documented in the brief make sustained acceleration less likely than the arrival model’s upper tail implies.

Consolidate

Strategy: logit_mean • Drafts: 3/3

Reviewers (3/3)

anchoring → shift_lower (low magnitude, medium confidence) anchoring_v1

The draft rationale explicitly uses the arrival-process projection’s mean of 3.45 as the central anchor, placing ~44% of mass in 3–6 (bin_1). However, the projection’s own median is 3 and p25 is 1, meaning the distribution is right-skewed — the mean is pulled above the median by the fat upper tail. Centering on the mean rather than the median (or mode=1) constitutes an optimistic anchor: the modal outcome from the arrival model is 1 interception, not 3–4. Additionally, the brief explicitly flags two suppressive factors that are directionally new relative to the historical base rate: (1) Russia deployed the Admiral Grigorovich to escort 12+ tankers by late May 2026, a military countermeasure that did not exist at the same scale during January–April 2026 when the 5 ACLED actions occurred; and (2) the draft only “qualitatively discounts” this with language like “Russian escorts only deter a fraction of potential interceptions” without shifting the p50 downward. The result is that 3–6 (bin_1) receives far more mass (44%) than 0–3, 30% (bin_0), even though the arrival model’s own p25=1 and mode=1 suggest the most common single outcome is in the lower range.

Flagged concerns

• Mean vs. median anchoring: The draft places p50 near the arrival model’s MEAN (3.45) rather than its MEDIAN (3) or MODE (1). For right-skewed Poisson/NB distributions, the mean exceeds the median, so centering on the mean biases mass too far right. The arrival model itself reports p25=1, indicating substantial probability below 3.
• Insufficient discount for new suppressive factor: Russia’s naval escort campaign (Admiral Grigorovich escorting 12+ tankers by late May) was NOT present at the same scale during January–April 2026 when the historical base rate was established. The draft acknowledges this only qualitatively but does not structurally shift the central estimate downward from the pre-escort baseline, effectively treating a new deterrent as negligible.
• Optimistic rate assumption: The brief notes ACLED records ~5 actions in Jan–Apr 2026 (~1.25/month), yet the arrival model uses 1.5/month (a 20% upward adjustment). The upward adjustment is rationalized by Sweden’s higher count, but the draft does not adequately account for the possibility that Sweden’s burst was a concentrated early-2026 campaign rather than a sustained rate, as explicitly flagged in the key uncertainties.

ceiling → shift_lower (low magnitude, medium confidence) ceiling_v1

The key structural ceiling constraint that the draft partially integrates but does not sufficiently weight is the UNCLOS “stateless vessel” legal basis, which as of late March 2026 was applicable to only 16 specific vessels (source 23). This is a hard discrete pool: qualifying interceptions within the window are bounded by (a) how many of those 16 vessels transit EU-adjacent waters in the ~2.3-month window AND (b) are not already escorted by the Admiral Grigorovich. Source 14 confirms Grigorovich had already escorted 12+ tankers by late May 2026, meaning a substantial fraction of the most-boardable targets are now under active naval protection. Additionally, France’s quarterly pace implies at most 0–1 French interceptions over the 2.3-month window, and Sweden’s 5-interception burst may have been concentrated against a finite set of identifiable stateless vessels (source 23’s 16-vessel pool). The draft puts ~24% of mass in bins 6+ (bin_2 through bin_8), implying plausible outcomes of 6–25+ vessels — but if the boardable pool is ~16 vessels and Grigorovich already escorts 12+, the structural ceiling for the window is closer to low single digits (3–6 at the outside), not 6+. The draft’s ~18% in 6–9 (bin_2) appears to over-weight outcomes that would require boarding rates exceeding the available pool of unescorted, legally-boardable vessels.

Flagged concerns

• Hard pool ceiling on boardable targets: Source 23 states the ‘stateless vessel’ legal basis applied to only 16 specific vessels as of late March 2026. Source 14 documents that the Admiral Grigorovich had already escorted 12+ of those tankers by late May 2026. This implies the unescorted, legally-boardable pool is likely ≤4–6 vessels at any given time, imposing a structural ceiling on the window count well below the 6–9 range (bin_2) receiving ~18% probability in the draft.
• France quarterly pace ceiling: Source 5 confirms France’s Tagor was its 4th seizure since September 2024 — roughly 1 per quarter. Over the 2.3-month window, France contributes at most 0–1 qualifying interceptions. The draft does not reflect this single-country ceiling, which combined with Sweden and Belgium implies a realistic combined maximum of ~4–6 for the window, not 9+.
• Swedish burst vs. sustained rate ambiguity: Sweden’s 5 interventions by early May (source 9) likely drew from the finite pool of 16 identifiable stateless vessels (source 23). Once those specific ships have been detained/released or rerouted, Sweden’s forward rate may drop sharply. The arrival model’s 1.5/month rate extrapolates this burst forward, but if it was target-pool-limited, the true forward rate is lower, compressing mass toward 0–3 (bin_0).

math → shift_higher (medium magnitude, high confidence) math_v1

The draft probabilities do NOT sum to 1.0. The published values sum to 0.9999999999999999 when converted to the required bin_id keys, which is within ±0.01 tolerance; however, there is a critical shape inconsistency: the rationale cites an arrival-process projection with mean ≈3.45, median ≈3, p75=5, p90=8, p95=9, yet the distribution assigns only 44.4% to 3–6 (bin_1) when the mean is 3.45. For a distribution with mean 3.45, the bulk of probability mass should lie in or near bin_1 and the lower end of bin_2, not split nearly evenly between bin_0 and bin_1. A mean of 3.45 with the stated percentiles implies that 0–3 (bin_0) should receive approximately 25% (p25=1), not 30%, and 3–6 (bin_1) should receive more than 44% to accommodate the mean. The current split overshoots bin_0 and undershoots bin_1 relative to the stated quantile structure.

Flagged concerns

• Mean–Mass Inconsistency: Rationale states arrival projection mean ≈3.45, median ≈3, with p25=1 and p75=5. For a distribution with this mean and median centered near 3, the modal bin should be 3–6 (bin_1), receiving majority mass (>50%). Published bin_0=30.1% and bin_1=44.4% imply a mean closer to 2.5–3.0, not 3.45. The assigned mass is too evenly split and underweights bin_1.
• Percentile Plausibility: The stated p5=0, p25=1, p75=5, p90=8, p95=9 are reasonable for an arrival-process model, but the published distribution does not produce these percentiles. With bin_0=30%, bin_1=44%, bin_2=18%, p25 would fall in bin_1 around 1–1.5 (plausible), but the heavy bin_0 mass pulls p5 and the lower quantiles too far down relative to the stated parameters.
• Scenario Weighting Mismatch: The brief tags the ‘Moderate count (3–5)’ scenario as ‘relative_evidence_strength: HIGH’ and the ‘Low count (0–2)’ scenario as ‘relative_evidence_strength: MEDIUM’. The distribution should reflect this by assigning substantially more mass to the moderate scenario’s range (3–5, contained in bin_1) than to the low scenario (0–2, contained in bin_0). Published 30% (bin_0) and 44% (bin_1) are nearly equal, contradicting the evidence weighting.

Reconcile: partial_shift

Reviews consumed: 3

The draft places ~30% in bin_0 and ~44% in bin_1, anchored on the arrival projection mean (3.45). The anchoring reviewer argued for shift_lower (mean vs. median; Russian escort suppression), and the ceiling reviewer also argued for shift_lower (16-vessel stateless pool, 12+ already escorted). However, the math reviewer correctly flagged a shape inconsistency: with stated mean 3.45 and median 3, 3-6 (bin_1) should carry more mass than 0-3 (bin_0), not less. These pull in opposite directions on bin_0 vs bin_1. I integrate the math reviewer’s structural point (bin_1 should dominate, consistent with HIGH-evidence moderate scenario per the brief) while partially honoring the suppression arguments (escort deterrence, finite boardable pool) by keeping meaningful but not dominant mass in bin_0 and trimming the upper tail (bin_3+) relative to the draft. Net effect: bin_1 receives the plurality of mass (consistent with median-3 arrival projection and HIGH-evidence scenario), bin_0 retains ~28% reflecting suppression risk, upper tail compressed because ceiling reviewer’s 16-vessel pool argument credibly caps higher outcomes. This is a partial shift — shape now matches the cited arrival-model percentiles better while honoring both lower-tail suppression and upper-tail ceiling concerns.

Rejected critiques

• [anchoring] Shift mass substantially lower because mode of arrival model is 1 and Russian escorts suppress rate

  • Why rejected: Partially accepted (trimmed upper tail, kept ~28% in bin_0), but a stronger downward shift was rejected because the brief explicitly tags the moderate (3-5) scenario as HIGH evidence strength and the math reviewer correctly noted bin_1 should dominate given the stated mean/median. Centering on mode=1 would underweight the high-evidence scenario.

• [ceiling] Hard pool ceiling of 16 stateless vessels with 12+ already escorted caps the upper tail near 3-6

  • Why rejected: Partially accepted by compressing bins 3+ relative to the draft, but fully accepting would ignore that the 16-vessel figure is from late March 2026 (3 months before window) and new vessels can be added to sanctions/tracker lists; also Sweden’s pace and France/Belgium’s independent capacity suggest the ceiling is not as hard as suggested.

• [math] Shift upward — bin_1 should exceed bin_0 given mean 3.45/median 3

  • Why rejected: Shape correction accepted (bin_1 now 45% > bin_0 28%), but the math reviewer’s full upward shift to 22% bin_0 / 52% bin_1 was rejected because two other reviewers independently flagged credible downside suppressors (Russian escorts, finite boardable pool) that justify keeping more lower-tail mass than a pure arrival-model fit would suggest.

Final Distribution

Evidence Sources (25)

1. Russia’s shadow fleet presents a sustained hybrid war threat at sea — ACLED records 8 total European enforcement actions against shadow fleet vessels: 3 in 2025 and 5 in the first four months of 2026 (January–April 2026), mainly on the basis of suspicion of being stateless or using false flags under UNCLOS.
2. Russia’s shadow fleet presents a sustained hybrid war threat at sea — The 2026 interception rate has been approximately 5 enforcement actions in the first 4 months (January–April 2026), equating to roughly 1.25 interceptions per month, or ~0.3 per week.
3. Why Are So Many Countries Now Seizing Shadow Fleet Ships — In 2025, there were 3 European enforcement actions against shadow fleet vessels (France and Estonia interceptions, Germany 2024), representing approximately 0.25 per month in 2025.
4. France intercepts another ‘shadow fleet’ tanker linked to Russian oil — The Tagor interception occurred on May 31, 2026 (Sunday), with public announcement on June 1, 2026. This falls BEFORE the June 4 start date of the question window and therefore does NOT count toward resolution.
5. France detains tanker linked to Iranian network accused of moving Russian oil — The Tagor interception was France’s fourth seizure of a Russian shadow fleet vessel since September 2024, confirming an accelerating pace of French enforcement actions.
6. Multiple sources checked June 4-5, 2026 — No new shadow fleet interceptions by EU countries have been confirmed since June 4, 2026 (as of June 5, 2026). All reporting on June 4 is about the aftermath of the Tagor (May 31 interception). The current floor for interceptions since June 4 is 0.
7. Europe’s coastal states tighten enforcement on Russia’s shadow fleet — Key confirmed EU interceptions/detentions of shadow fleet vessels: Germany (Eventin, Jan 2025), France (Grinch, Jan 2026), Belgium+France (Ethera, Feb 28 2026), Sweden (Caffa March 6 2026 for seaworthiness, Jin Hui May 2026), France (Tagor, May 31 2026) — totaling 5 ACLED-counted actions in first 4 months of 2026 (excluding May/June).
8. More than 90% of shadow fleet ships are producing insurance certificates when challenged — As of July 2025, well over 90% of shadow fleet tankers passing EU/Baltic coastal states simply comply with insurance document requests without being intercepted. Estonia alone challenged 1,033 ships for documentation since June 2024, but physical interceptions are rare by comparison.
9. Sweden detains suspected Russian shadow fleet tanker in Baltic Sea — Sweden’s Jin Hui boarding (May 3, 2026) was described as Sweden’s fifth intervention in 2026 alone. Sweden also detained the Caffa and Sea Owl in March 2026. Sweden alone had 5 interventions in 2026 by early May, demonstrating accelerating enforcement in just one EU state.
10. The shadow fleet is undermining the maritime order more brazenly than ever — As of April 2026, Russia deployed military warships including the frigate Admiral Grigorovich to escort shadow fleet oil tankers through the Baltic Sea and English Channel, a countermeasure that could deter or complicate EU interceptions during the question window.
11. Nordic Business Journal, Kyiv Independent, ACLED combined — The ACLED count of 5 actions in Jan-April 2026 likely undercounts the total, since Sweden alone had 5 interventions in 2026 by May, suggesting the overall rate may be higher than 1.25/month — possibly 2-3 per month when counting all EU member states including Sweden’s multiple March/April interventions not yet in ACLED data.
12. Arrival forecast model based on ACLED/Kyiv Independent base rates — Arrival forecast for EU shadow fleet interceptions in the June 4 – August 12, 2026 window (~2.3 months): Using arrival rate of 1.5/month (conservative estimate) with overdispersion=0.5, projected mean=3.45, median=3, most likely=1, p5=0, p25=1, p75=5, p90=8, p95=9. Current floor is 0 (no interceptions confirmed since June 4).
13. ACLED - ‘Russia’s shadow fleet presents a sustained hybrid war threat at sea’ — ACLED report (May 2026) states that ACLED recorded eight European enforcement actions against shadow-fleet vessels, of which three occurred in 2025 and five occurred in the first four months of 2026, documenting growing enforcement activity.
14. Putin’s cruise missiles are operating 30 miles off the UK. And Russia won’t move them — By late May 2026, the Admiral Grigorovich had escorted over a dozen shadow fleet tankers through the English Channel, with two Russian vessels (the frigate and a 122-metre supply ship) operating in tandem — the supply ship restocking the frigate at the Galloper wind farm off Suffolk, enabling persistent loitering 30 miles off the UK coast.
15. Russian warship shadowed for whole MONTH prowling off UK shores - The Mirror — The Royal Navy confirmed it tracked the Admiral Grigorovich for the entire month of April 2026 without a single day unmonitored; the frigate escorted approximately six merchant/auxiliary ships and one submarine, moving between the North Sea, Western Approaches, Atlantic, Mediterranean, and Baltic — but no British interception of escorted tankers occurred during this period.
16. Why the West Doesn’t Board Shadow Fleet Tankers — Three structural barriers constrain EU enforcement against shadow fleet tankers: (1) UNCLOS legal architecture channels enforcement toward indirect mechanisms (insurance, port state control, flag state pressure) rather than physical interdiction; (2) the G7/EU price cap sanctions regime was deliberately designed to tolerate evasion to avoid energy supply shocks; and (3) a strategic calculus weighing precedent effects, coalition management, and escalation risk makes routine interdiction unattractive even when technically feasible.
17. EU Commission on US Seized Russian Shadow Fleet and Sanctions — As of early 2026, EU enforcement against the shadow fleet remains primarily a member-state competence under national legal frameworks, not a centralized EU authority. When pressed at a January 8, 2026 press briefing, the European Commission spokesperson confirmed no EU-level legal basis specifically authorizes member states to board or intercept shadow-fleet vessels, with implementation left to national maritime law.
18. Russian frigate spotted in UK waters as Putin brazenly steps up his ‘grey war’ — A NATO official confirmed in May 2026: “We see a trend of Russian Federation Navy ships escorting sanctioned vessels through these waterways,” adding that these escorting activities “take up valuable days at sea for the Russian navy, and it places a strain on their ability to generate naval forces for other missions.”
19. The 205 times Putin has brazenly sent shadow ships into UK waters — As of 29 May 2026, at least 205 sanctioned Russian tankers had sailed through the UK’s 200-mile EEZ since Starmer announced seizure powers (with 84 entering the 12-nautical-mile territorial waters), and shadow fleet tankers returned to English Channel transits after a brief detour around Scotland — often accompanied by Russian naval vessels — as time passed without UK enforcement action.
20. Europe’s coastal states tighten enforcement on Russia’s shadow fleet — By March 2026, European coastal states were shifting from ‘symbolic boardings’ to sustained maritime law enforcement, with notable examples: France intercepted the tanker Grinch in January 2026 (detained with a penalty fine, setting a new precedent); Belgium and France jointly boarded the shadow-fleet tanker Ethera under ‘Operation Blue Intruder’ for false flag and forged documents; and Sweden seized the cargo vessel Caffa on March 6, 2026 in Swedish territorial waters. However, IISS notes the shift is nascent and carries significant NATO escalation implications.
21. Can the West shut down Russia’s shadow fleet? — A key operational barrier to shadow fleet interception is the absence of coordinated targeting criteria across maritime states. Without common inspection standards for flag anomalies, insurance gaps, AIS manipulation and ownership opacity, shadow fleet operators can ‘shop for the easiest port and the quietest regulator.’ Europe’s enforcement is described as ‘episodic and politically fragile’ rather than systematic, as of early 2026.
22. Can the West shut down Russia’s shadow fleet? — The January 2026 U.S. seizure of the shadow fleet tanker Marinera demonstrated that Russia may respond to interdictions by deploying naval assets (including a submarine) to shadow pursued tankers, creating an escalation management challenge. This military signalling means European states must prepare rules of engagement and legal justifications in advance — a structural deterrent to routine enforcement escalation through the summer of 2026.
23. How Germany Can Act Against the “Shadow Fleet” and Shape International Order — UNCLOS Article 110 permits states to approach and inspect vessels suspected of being stateless, but is silent on what states may do subsequently. This legal gap was central to both the U.S. Marinera seizure in January 2026 and debates about EU interception powers; the EU found a limited legal opening by treating tankers as ‘stateless’ after flag renunciation — applicable to 16 specific vessels as of late March 2026.
24. The shadow fleet is undermining the maritime order more brazenly than ever — In 2026, Baltic Sea countries, France, India, and others inspected and detained more shadow-fleet vessels than in prior years, but the Atlantic Council (April 2026) still describes the shadow fleet as ‘undermining the maritime order more brazenly than ever,’ indicating enforcement escalation has not yet structurally suppressed shadow fleet operations heading into summer 2026.
25. Can the West shut down Russia’s shadow fleet? — Shadow fleet operators counter enforcement by rapidly reflagging and rebranding vessels — ships can change name, flag, and paper ownership faster than EU sanctions lists are updated. Policy that focuses primarily on hull listings is therefore described as ‘reactive’ and structurally limited as an enforcement tool through mid-2026.

wooley_rhino • run 9065d25a • $3.03 • 8m3s

Extracted Percentile Values: {0.1: 1.01, 1.0: 1.01, 5.0: 1.01, 10.0: 1.01, 15.0: 1.01, 20.0: 1.01, 25.0: 1.2, 30.0: 1.6, 35.0: 1.9, 40.0: 1.99, 45.0: 1.99, 46.0: 1.99, 47.0: 2.0, 48.0: 2.0, 49.0: 2.0, 50.0: 2.0, 51.0: 2.0, 52.0: 2.0, 53.0: 2.1, 54.0: 2.2, 55.0: 2.3, 60.0: 2.7, 65.0: 2.95, 70.0: 3.0, 75.0: 3.4, 80.0: 4.0, 85.0: 4.0, 90.0: 5.0, 95.0: 6.0, 99.0: 9.0, 99.9: 26.0}

Model’s Answer: Step one is to anchor on the observed 2026 pace rather than the fleet’s huge size. The most decision-relevant evidence is the recent Europe-wide enforcement tempo: roughly five actions in January through April per ACLED, plus the May Sweden case and the June 1 France-UK Tagor case just before the window. That implies Europe has moved from rare symbolic action to a real but still sparse cadence of boardings and detentions.

Step two is to discount that raw pace because this question has stricter resolution criteria than “newsworthy action.” A qualifying case needs a first physical stop attempt inside the window, EU participation, and later public verification that the vessel was part of the Russian shadow fleet. Some reported inspections will fail one of those tests. That pushes me down from a naive straight-line extrapolation.

Step three is to account for the short horizon. There are only about ten weeks left. Short windows favor the base case and punish tail-heavy stories. Even if Europe is becoming more aggressive, that does not leave much time for a dramatic regime shift to routine interdiction. The most likely story is still a small handful of case-by-case actions, led mainly by France and possibly one or two Nordic/Baltic or Benelux incidents.

My center is around 2 vessels, with substantial mass on effectively zero or one qualifying event and a thinner but meaningful tail to 4 to 6. I think the chance of a genuinely high count needs a trigger, such as a spill scare, a near-miss, or a political decision after another provocatory incident.

A low outcome would result if legal caution dominates, states avoid escalation with Russia, and many suspicious tankers are monitored rather than physically stopped. That low-outcome bucket, substantively zero or one qualifying interception, looks about 45 percent likely.

A high outcome would result if France sustains its pace, another proactive state joins in, and one environmental or security incident creates political cover for clustered interdictions. I put 5 or more at about 12 percent, and truly extreme outcomes above 10 as very unlikely.

The assumptions most likely to matter are fairly clear. The first is that the current case-by-case enforcement posture persists rather than either collapsing back into caution or jumping to a much more aggressive campaign; I put about a 25 percent chance that this assumption is materially wrong. The second is that reported boardings and seizures during the window will map reasonably well onto the resolution criteria; I put about a 20 percent chance that qualification/reporting mismatch is larger than I expect. The third is that France remains the main driver and does not pause after the Tagor case because of legal or diplomatic blowback; I put about a 30 percent chance that this is materially wrong. Those failure modes mostly widen the distribution rather than shifting the median dramatically.

There is an explicit formatting-bound inconsistency in the prompt, since the real-world count could be zero but percentile outputs are required to be above 1. I therefore express the lower tail just above 1 while substantively meaning “very low, including some chance of zero.”

Percentile .1: 1.01 Percentile 1: 1.01 Percentile 5: 1.01 Percentile 10: 1.01 Percentile 15: 1.01 Percentile 20: 1.01 Percentile 25: 1.20 Percentile 30: 1.60 Percentile 35: 1.90 Percentile 40: 1.99 Percentile 45: 1.99 Percentile 46: 1.99 Percentile 47: 2.00 Percentile 48: 2.00 Percentile 49: 2.00 Percentile 50: 2.00 Percentile 51: 2.00 Percentile 52: 2.00 Percentile 53: 2.10 Percentile 54: 2.20 Percentile 55: 2.30 Percentile 60: 2.70 Percentile 65: 2.95 Percentile 70: 3.00 Percentile 75: 3.40 Percentile 80: 4.00 Percentile 85: 4.00 Percentile 90: 5.00 Percentile 95: 6.00 Percentile 99: 9.00 Percentile 99.9: 26.00

Forecast (median): 2.7831 intercepted shadow-fleet vessels (EU countries) between June 4 and Aug 12, 2026.

• Recent enforcement suggests a steady trickle, not a flood. Through spring 2026, actions have been happening at roughly a low “monthly” pace, which over ~2.2 months points to around the low single digits.
• France is the most consistent driver, so outcomes track France’s continued push. The overall total is treated as highly sensitive to whether France keeps up its tempo and whether other states add support.
• Russian escorts and deterrence likely suppress some attempts. There are clear signs of Russia actively managing risk in nearby waters, and one nearby state (Estonia) has stepped back, both pulling counts downward.
• A key upside unknown is whether a mentioned boarding (“Oneiroi”) happens in the window and qualifies. If it does, that would nudge the total upward by about one.
• There’s meaningful uncertainty and a right tail. Most scenarios cluster around 0–4, but a coordinated summer escalation could produce a noticeably higher result (several more than the median).