All Sectors

"Wholly Avoidable": What the Verity / Polesie Collision Teaches Every Watchkeeper About the COLREGS

🕑 5 min read words Safety • Incident

4:55am, German Bight

In the dark, early hours of 24 October 2023, the general cargo ship Verity and the bulk carrier Polesie collided approximately 12 nautical miles southwest of Helgoland, in the German Bight. Verity sank within minutes. Two of her crew survived. Five did not.

MAIB Report 5/2026, published in early 2026, doesn't describe a freak event or an unforeseeable mechanical failure. Chief Inspector of Marine Accidents Andrew Moll's conclusion is blunt: this accident was wholly avoidable. Both watchkeepers had radar. Both had time. Both had the COLREGS — rules that exist precisely to resolve situations like this one. Neither applied them diligently.

What Happened: The Factual Record

Verity, as the give-way vessel in a crossing situation, was required under Rule 16 to take early and substantial action to keep well clear. Instead, her officers made a series of small course adjustments via the autopilot — changes too subtle for Polesie's watchkeeper to recognise as deliberate collision-avoidance action. Verity then attempted to pass ahead of the much larger Polesie, a manoeuvre the investigation found added risk rather than resolving it.

On Polesie, the stand-on vessel under Rule 17, the watchkeeper also made only small autopilot adjustments. At 04:52:55 — just under two and a half minutes before impact — the officer still had the ability to turn hard away from Verity. Instead, Polesie made a late, small alteration of course to port — a move MAIB found compounded the danger and added further ambiguity to an already confused picture.

The two vessels collided at 04:55. Verity began flooding immediately and sank within minutes — too fast for her crew to don lifejackets or immersion suits. Verity did not carry a Voyage Data Recorder, which significantly hampered the investigation's ability to reconstruct exactly what was said and done on her bridge in the final minutes.

Separately, the German VTS covering the area operates on a duplex VHF channel requiring an assistant to relay messages between vessels rather than allowing direct ship-to-ship contact. By the time VTS intervened — using non-standard warning markers — the two ships were already too close for outside help to change the outcome.

Root Causes: Two Watchkeepers, One Avoidable Outcome

1. "Small and Late" Instead of "Early and Substantial"

The COLREGS don't just say what a give-way vessel must do — they say how. Rule 8 requires that any action taken to avoid collision be "large enough to be readily apparent to another vessel observing visually or by radar." A 5° autopilot nudge, repeated every few minutes, fails that test even if the cumulative effect would eventually have been a large alteration. To the other ship, it looks like nothing is happening — because, from their perspective, nothing visible is.

2. The Stand-On Vessel's Moment Came and Went

Rule 17 exists for exactly this scenario: a point where the stand-on vessel realises the give-way vessel isn't acting, and is not just permitted but required to act herself. Polesie's watchkeeper reached that point — the investigation notes the ability to turn hard away still existed at 04:52:55 — and chose a small adjustment instead. Worse, that adjustment was to port, across the bow of the situation, adding ambiguity at the exact moment clarity was needed most.

3. No VDR, and a VHF System That Couldn't Help in Time

Verity's lack of a VDR wasn't a cause of the collision, but it shaped everything about understanding it afterwards — and it's a gap MAIB specifically recommended addressing for vessels of her size. The VTS's duplex VHF arrangement meant that even if either bridge team had wanted external input, the channel itself added delay at the one moment delay couldn't be afforded.

What If? — Scenario-Based Escalation

Verity and Polesie collided in the dark, in open water, with no other complicating factors. Now change a few of those variables.

What if this had happened in a busy shipping lane with a third vessel nearby? Two ships making small, hard-to-read course changes in a crossing situation is dangerous enough. Add a third vessel reacting to either of them, and the small-adjustment problem multiplies — each ship now has to read two sets of ambiguous signals instead of one.

What if Verity's crew had been at their lifejacket muster stations for an unrelated drill 20 minutes earlier — and assumed the gear was already to hand? The rapid sinking gave no time to retrieve survival equipment from stowage. The margin between "survived" and "didn't" in this accident was measured in the seconds it takes to put on a lifejacket that's already within arm's reach versus one that isn't.

What if the OOW on either vessel had called the Master to the bridge as the situation developed, rather than managing it alone? Neither watchkeeper's actions suggest a second, more senior perspective was brought in before the situation became critical. An extra set of eyes — and the authority to order a large, unambiguous alteration without hesitation — changes the calculus.

What if VTS had been on a single, open VHF channel both ships could hear directly? The duplex relay system added time to an already late intervention. A direct "Verity, Polesie, this is VTS, you are on a collision course" broadcast on a common channel, heard by both bridges simultaneously, is a different intervention to one routed through an assistant.

Role-Based Lessons

Officers of the Watch

If you're the give-way vessel, "early and substantial" isn't a vague aspiration — it's the legal standard, and it exists because the other ship needs to see that you're acting, not just know internally that you've adjusted the autopilot. If you're the stand-on vessel and the other ship still isn't acting as the situation closes, Rule 17 doesn't just permit you to act — at some point it requires it. And when you do act, a turn that adds ambiguity (like altering toward the other vessel's projected track) is worse than no action at all.

Know whether your vessel carries a VDR, and if it does, that it's tested and recording. If something does go wrong, that data is often the only objective record of what each bridge team actually did — and the only way lessons like this one get learned at all.

Chief Officers

A watch handover in open water with other traffic isn't complete without a clear picture of what's being monitored and what's already been considered. "There's a ship out there somewhere" is not a handover; "that vessel's been on a steady bearing for the last 20 minutes, CPA looks like half a mile, I haven't taken action yet" is. The relieving officer shouldn't have to rebuild the picture from scratch while the situation is already developing.

Masters

Verity's crew had no time to don lifejackets because the ship sank in minutes — not because lifejackets weren't on board. Whether your crew has "immediate access" to survival equipment during periods of elevated risk (restricted visibility, congested waters, the early hours of a night watch) is a culture and routine question, not a stowage-plan question. And when a watchkeeper is managing a developing close-quarters situation alone in the small hours, your bridge night orders should make it unambiguous when and why to call you — before the situation, not after.

Test Your Knowledge

Think you've got this covered? Put it to the test in Crew Connect's free Knowledge Checker — try a quick COLREGS round and see how you score.