Why did the ‘Bayesian’ sink?A ZDF documentary featuring simulations and unanswered questions

Lars Bolle

 · 03.08.2026

Simulation of the drag caused by the rig and superstructure in a virtual wind tunnel.
Photo: ZDF/Terra X
In a 45-minute documentary, ZDF reconstructs the sinking of the 56-metre superyacht ‘Bayesian’ off the coast of Sicily. Simulations offer new possible explanations, but key questions remain unanswered due to ongoing legal proceedings.​

Topics in this article

ZDF has, in a 45-minute documentary reconstructed the events that led to the dramatic sinking of the 56-metre superyacht ‘Bayesian’ in 2024 off the Sicilian coast. Seven people lost their lives in the sinking, whilst 15 were rescued.

​ZDF reconstructs the night of terror off the coast of Sicily

The authors frequently use the subjunctive in their reconstruction. This is hardly surprising. The parties involved have filed lawsuits against one another running into millions, and other journalists have also been sued for alleged false claims. The reconstruction of events to date is based primarily on the laws of physics and simulations. There is as yet no official final report, and legal proceedings are ongoing. Furthermore, according to the documentary’s authors, all parties – from the owners to the designer and the three crew members facing charges (of negligent shipwreck and manslaughter) – were given the opportunity to comment on the details of the documentary. However, their lawyers reportedly declined to do so.

​Well worth seeing, but not entirely convincing in every respect

Essentially, the documentary reflects the current state of knowledge, which has also been published by YACHT and summarised further down in this article. Nevertheless, it is well worth watching. It provides a meticulously crafted and dramatically told reconstruction of that night of horror.

The only point of criticism is that the question in the subtitle of the documentary – ‘Why do ships sink?’ – is not answered in a remotely satisfactory manner. The other shipping accidents cited have little or no connection to the events on the ‘Bayesian’. On this point, the filmmakers have set their sights too high.

Most read

1

2

3

4

5

​How the rig became a windbreak despite the sails having been stowed away

What is interesting, however, is a model of the Spanish Association of Marine and Naval Engineers featured in the documentary. Their wind simulation revealed that, when the ship was heeled over, the rigging and the pairs of spreaders in particular generated a wake whose drag acted like a flying kite. Accordingly, although all the sails had been furled, the ‘Bayesian’ was, from a ‘wind perspective’, still sailing solely because of its rig.

​​Ventilation openings as a possible cause of flooding

Also of interest are the Spanish simulations shown in the documentary regarding the yacht taking on water. If, as is suspected, the ventilation openings had not been closed, the yacht would have started taking on water as early as a 35-degree list (galley ventilation). If, as is also suspected, the crew were preparing to drop anchor, the ventilation hatches for the engines would have had to have been open for that purpose.

The engines only run when the ventilation hatches are open. According to the modelling, the engine hatches would have started taking on water at an angle of 40 degrees or more, at a rate of one tonne per second. Eyewitnesses reported that the yacht sank stern-first, which is consistent with the engine rooms being flooded. Whether this was in fact the case and whether the crew can be accused of negligence as a result is the subject of legal proceedings.

Seven people lost their lives in the accident. Their bodies were recovered one by one by divers several days later. The documentary’s makers do not comment on the circumstances in which they were found – which gave rise to much speculation, for example as to which cabins they were in and what condition they were in.

​Timeline: The sinking off Sicily

This was reported by YACHT:

During the night of 18–19 August 2024, the 56-metre sailing yacht ‘Bayesian’ was at anchor about half a nautical mile off Porticello on the north coast of Sicily. There were twelve guests and ten crew members on board. During a violent thunderstorm, the yacht was caught in extreme gusts, capsized and sank to a depth of around 50 metres within a few minutes.

In our first report, dated 20 August, we described, under the heading “56-metre-long ‘Bayesian’ sinks off Palermo, six people missing” based on the information available at the time. Fifteen people had been rescued from a liferaft, one crew member had been found dead, and six people were still missing. The ‘Sir Robert Baden Powell’, which was anchored nearby, played a crucial role in the rescue. Its captain, Karsten Börner, and his crew spotted the liferaft and took the survivors on board. A waterspout was initially suspected as the possible cause.

Two days later, we reported that Owner Mike Lynch had been recovered from the wreckage, dead. By this point, five of the six missing people had been found, whilst the search for Lynch’s 18-year-old daughter, Hannah, continued. At the same time, the first technical considerations began: discussions centred on open fuselage flaps, the wind resistance of the unusually tall aluminium mast, and the possibility that the lift keel had been retracted.

Initial attempts at a technical explanation

On 28 August, we published the Assessment by Stephen Edwards, former captain of the ‘Bayesian’. Edwards had captained the yacht from 2015 to 2020 and highlighted its unique status: it was the only sloop on Perini Navi’s 56-metre platform, which was otherwise designed as a ketch, and featured an exceptionally tall mast.

According to his account, the fixed ballast consisted of around 200 tonnes of lead, whilst the movable keel served primarily as a centreboard to prevent the vessel from drifting off course. The so-called flooding angle was said to have been the decisive factor. At a list of around 40 to 45 degrees, water could have entered via ventilation openings, provided the relevant systems were in operation. Edwards, however, considered it unlikely that a large hull hatch had been left open. His explanations outlined the yacht’s technical limitations, but could not yet substantiate the specific sequence of events leading to the accident.

The challenging preparations for the rescue

After several months on the seabed, the focus shifted to the recovery operation. In early February 2025, we initially reported that the The salvage of the ‘Bayesian’ was delayed due to the complexity of the operation. The plan was to right the yacht and then lift it using cranes or additional buoyancy systems. Italian and British investigators hoped that an examination of the wreck would provide information on the position of the keel, possible routes of flooding and the behaviour of the crew and the vessel during the capsizing.

Then, in March, it was announced that the Recovery work is due to begin on 20 April 2025. Once the wreck had been raised, it was to be taken to Termini Imerese for forensic examination. At the same time, criminal investigations were underway in Italy, whilst investigations into the deaths of the British victims were being carried out in the UK.

The work eventually began later than planned and was overshadowed by another fatal accident. On 12 May, we reported that a Dutch diver had died during the salvage operation. Around 80 specialists were involved in the operation. The mast, which was more than 70 metres long, was to be cut away before the hull was lifted. Media reports put the cost of the entire salvage operation at around 30 million euros.

The British interim report reconstructs the night of the accident

On 16 May 2025, we published a detailed analysis of the the initial investigation report by the British Marine Accident Investigation Branch. The report provided a much more accurate account of the night’s events than the initial reports.

According to the report, the ‘Bayesian’ had sought shelter from the forecast weather off Porticello. At around 3 o’clock, the thunderstorm approached; shortly afterwards, both the ‘Bayesian’ and the neighbouring ‘Sir Robert Baden Powell’ began to drift with their anchors dragging. At 4.06 am, the ‘Bayesian’ listed approximately 90 degrees to starboard within just 15 seconds. The generators failed, and water poured over the starboard rail and down the stairwells into the interior. At 4.22 am, the crew managed to launch a liferaft. At around 4.45 am, the yacht sank.

The stability calculations were particularly alarming. According to the modelling used in the report, even a gusty wind speed of more than 63.4 knots from a direct crosswind could have caused the yacht to capsize. The actual gusts that occurred were apparently higher than this. The report also noted that, following the extreme list, the yacht was no longer able to right itself. At that point, the investigation was expressly still ongoing, as the wreck had not yet been fully examined.

We updated the report on 23 May with a Podcast episode in which Editor-in-Chief Martin Hager put the results into context. The topics discussed included the low capsize angle, the design of the keel and centreboard, the role of the ‘Sir Robert Baden Powell’, and sudden thunderstorm cells forming in the Mediterranean.

A day later, a comment was posted which was explicitly labelled as an opinion “Bayesian Doomsday: In the wrong place at the wrong time”. In it, Martin Hager argued, on the basis of the British interim report, that blanket accusations against the crew were unjustified. At the same time, he questioned why a superyacht licensed for charter operations had a significantly lower capsize angle than many modern production or racing yachts.

The ‘Bayesian’ is being raised

On 21 June 2025, the Breakthrough in the salvage of the ‘Bayesian’. The mast had previously been cut off using a remote-controlled cutting tool and left on the seabed. Eight lifting slings were threaded under the hull, after which the crane vessel “Hebo Lift 10” lifted the yacht from a depth of around 50 metres. The wreck was stabilised and transported to Termini Imerese, where British and Italian experts were able to continue their investigations.

Four days later, we asked, under the heading “‘Bayesian’ salvaged: a superyacht bargain on the horizon?”, whether the yacht could still be salvaged from a technical point of view. An expert explained that corrosion underwater can proceed relatively slowly due to the reduced exposure to oxygen. However, once a wreck has been brought to the surface, it must be desalinated and preserved without delay. As the yacht was initially secured as evidence, rapid technical treatment was only possible to a limited extent.

On 30 June, we published First pictures from the wreck. They revealed an interior that was largely destroyed and covered in mud and sediment. The saloon, companionways and engine room were badly damaged. The severed aluminium mast was to be salvaged separately. Although the images gave an impression of the consequences of the ten months spent underwater, they did not yet answer the crucial question of the original route of flooding.

A multi-million lawsuit and two conflicting lines of inquiry

In January 2026, the case took on a new dimension, both economically and legally. We reported on a Italian Sea Group’s claim for damages amounting to 456 million euros. The shipbuilding group to which Perini Navi belongs brought the claim against the yacht’s owner, Lynch’s widow Angela Bacares, as well as the captain and crew members. According to the company’s account, it was not design flaws but rather a failure to implement safety measures, unsealed openings, the keel’s position and the handling of weather warnings that led to the sinking. These allegations form part of the legal proceedings and are not facts that have been conclusively established.

In May 2026, we then reported on new findings by the Italian public prosecutor’s office. A meteorological report commissioned by the public prosecutor’s office therefore did not classify the incident as an uncontrollable natural disaster, but rather as a dangerous thunderstorm that, in principle, could have been managed by an experienced crew. As a result, the captain, the chief engineer and a deckhand came under greater scrutiny in the Italian investigation.

This assessment stood in stark contrast to the British investigation. The MAIB had highlighted the yacht’s poor stability and inadequate information in the stability manual as key factors. The Italian side, by contrast, focused more on the anchor position, the keel’s orientation, the night watch and the crew’s response. All those accused are presumed innocent.


What did you think of the ZDF documentary? Have your say in the comments.

Share article:
Lars Bolle

Lars Bolle

Chief Editor Digital

Lars Bolle is Editor-in-Chief Digital and one of the co-founders of YACHT's online presence. He worked for many years as an editor in the Sports and Seamanship section and has covered many sailing events. His personal sailing vita ranges from competitive dinghy sailing (German champion 1992 in the Finn Dinghy) to historic and modern dinghy cruisers and charter trips.

Most read in category Special