The Flow Report

Live Intelligence

Commodity intelligence · Vessel tracking · Physics-informed analytics · Spatial intelligence

0 positions0 vessels·76 engines

FilterAllTechnicalWeeklyAnalysis

🔍AnalysisJun 12, 2026·3 min read

63% of Nighttime Vessel-Hours Concentrate in 8% of H3 Cells. They Are Offshore Anchorage Rings, Not Port Approaches.

63% of nighttime vessel-hours concentrate in 8% of H3 cells — offshore anchorage rings, not port approaches. The 340 cells with the highest nighttime density and elevated AIS gap frequency are where the pattern becomes actionable....

8resolution8cells top pct340overlap cells count45analysis period days

VESSEL-DENSITYH3-HEX-CELLSOVERNIGHT-GRADIENTAIS-OPACITYOFFSHORE-ANCHORAGEDARK-EVENTS

Read article

🔍Analysis

Anchorage Encounters Produce Rule 17 Deviations at a Higher Rate Than Approach Channels. 1.9 Million Encounters Across Three Contexts.

In 45 days and 1.9M encounters: anchorage shows a 53.5% Rule 17 deviation rate — higher than approach channel (47.2%) — despite vessels being farther apart at CPA.

Jun 9, 2026·5 min

PAIRWISE-ENCOUNTERRULE17

🔍Analysis

83.3% Coverage Is Not 83.3% Accuracy: Only 1.24% of Vessel Visits Use a Draft-Survey-Derived Cargo Estimate

The hydrostatic coverage headline counts zero-cargo placeholders and population fallbacks as estimates. In 60 days and 41,377 vessel visits, only 514 used the draft-survey-derived method.

63% of Nighttime Vessel-Hours Concentrate in 8% of H3 Cells. They Are Offshore Anchorage Rings, Not Port Approaches.

Anchorage Encounters Produce Rule 17 Deviations at a Higher Rate Than Approach Channels. 1.9 Million Encounters Across Three Contexts.

83.3% Coverage Is Not 83.3% Accuracy: Only 1.24% of Vessel Visits Use a Draft-Survey-Derived Cargo Estimate

Rule 17 Stand-On Vessel Class: Tankers Are Named in 31% of Extremis Events and Represent 22% of the Tracked Fleet

71,663 Approach Channel Encounters Ended With Give-Way Failure. Average CPA: 0.247nm — the Closest of Any Context in the Dataset.

4,361 STS Transfer Alerts in 45 Days. Zero Acknowledged. Ship-to-Ship Transfer Is Now the Dominant Behavioral Motif.

8,400 Dark Events in 45 Days. The Nav Status at Gap-Start Cuts the Actionable Queue by 14%.

Norfolk Bulk Carriers Wait 127 Hours. Houston Median Is 9. One Statistic Contains Both Facts.

2,894 Vessel Loitering Events in 45 Days. Two Reached High Risk. Context Determines the Classification.

97% of Dark Events Get No Subtype. The 3% That Do Run 8x Longer.

Every vessel page now carries an ownership dossier

2,607 Vessels Reappeared at Physically Impossible Locations in 30 Days. The Subtype Is Called impossible_kinematics.

Mexico Leads US Grain Export Events at 129. China Receives 43. China's Per-Event Volume Is 56% Higher. The Difference Is Vessel Class.

5.3% of the Vessel Graph Is Mutation Edges. Shell Hops Average 13-Day Tenure. That Asymmetry Is the Screening Problem.

4,433 COLREGS Encounters, Zero Analyst Reviews. Head-On Events Need 23.8 Degrees and Carry the Least Classifier Certainty.

Open Water Is Where Vessels Come Closest. 16.7% of Open-Sea Encounters Pass Within 0.1nm — 49% More Than Anchorage.

6% of Active Hex Cells Account for 71% of Encounter Events. The Resolution-8 Density Distribution Breaks Most Anomaly Models.

2,241 Vessels Made an Unexplained 94-Degree Course Change at 7 Knots Last Month. They Were Avoiding Something That Wasn't Broadcasting AIS.

57% of Predicted Hormuz Transits Disappear Before They Reach the Exit Zone. Bab-el-Mandeb Loses 9%.

4,578 High-Severity Behavioral Motifs, Zero Acknowledged: AIS Manipulation Makes Up 42% of the Fleet Alert Stack

197 Shell Hops, 425 MMSI Changes, 383 Renames: The 1,005 Identity Mutation Events in 33 Days

Rule 17 in the Wild: 72,113 Handoffs, 75% Extremis, and What That Says About Who Is Actually Avoiding Collisions

47% of Vessel Visits Do Not Have Draft Data. The Cargo Estimate You Are Reading Is a Fallback.

Port Dwell Distributions: Newcastle 2.8×, Houston 13.6×. The Tail Ratio Predicts Demurrage Risk.

6.4% of AIS Pings Don't Say Where They're Going. The Rest Spell Houston Seven Different Ways.

Recalibrating Risk Tiers Surfaces 1,702 Dark Events; 66% Cluster in the Singapore Strait

Singapore Strait Holds 65% Ship-to-Ship Share While Rotterdam Climbs to 59%: The STS Subtype Reads Differently From the Atlas

Dark Events Across Eight Chokepoints: Every Port Has Its Own Gap-Duration Risk Curve

Draft-Survey Commodity Flow Estimates: 13.2M Tons at US Gulf, 2.4M Tons at Newcastle, and the Mississippi Spring Constraint

Vessel Identity Risk: How Flag State, Class Rarity, Phantom Fleets, and Name Changes Each Contribute to the Same Sanctions Exposure

Vessel-Class Kinematic Signatures: Tanker Noise, Bulk Precision, and the 6× Gap Where Spoofing Hides

From Field to Port: How We Built Commodity Intelligence From Five Government APIs

What Axiom Overwatch Is — and Who It's For

How We Turn AIS Signals into Maritime Intelligence