Do I Really Need a Life Raft?

[lestersails]

Originally Posted by chrisr

...of course in item 2, all they have to do is refuse a permit to operate but it would be a massive call to fine or arrest a british vessel that being used by a spanish resident...

Or your insurer deciding not to pay a claim because you are not in conformity with applicable safety regulations?

[rslifkin]

Originally Posted by chrisr

of course in item 2, all they have to do is refuse a permit to operate but it would be a massive call to fine or arrest a british vessel that being used by a spanish resident

If it's being operated by a Spanish resident, but owned by someone else, that might be tough to enforce. But if you're a Spanish resident or citizen and also the registered owner of the vessel (even though it's registered elsewhere), I'd expect that would make it a bit more enforceable.

[Greg K]

Originally Posted by chrisr

are you referring to RD 339 / 2021 ?

such states it applies to recreational craft in any of the following scenarios:

1. Registered or pre-registered craft in Spain.
2. Craft carrying out activities with commercial and profit-making purposes in Spanish waters.
3. Craft sailing through Spanish waters whose owners or users reside in / have a registered office in Spain.

i would be really interested to see an actual legal opinion on this as frankly i don't believe any port state has the power to override the flag state and regulate 2 & 3 (as new zealand found out)

of course in item 2, all they have to do is refuse a permit to operate but it would be a massive call to fine or arrest a british vessel that being used by a spanish resident

has it ever been tested in court ?

cheers,

Yes, this is the law I'm referring to. Had a one year grace period, so became effective in May 2022. Originally aimed at Spanish nationals who increasingly have opted for other flags of convenience to circumvent the onerous safety regulations pertaining to Spanish flagged boats, I would speculate. But, unfortunately, foreign residents were also included as is clear in #3 scenario.

No idea if this has been challenged in court or to what degree it is being enforced.

[Greg K]

Originally Posted by Mike OReilly

Having these mandatory life raft countries opens the opportunity to study what impact they are having on safety. Is anyone aware of any studies? It would be obvious to look at outcomes before and after these mandates. Could also compare places with, and without, such laws.

Tricky research as always, but it should be doable.


ADD: I went looking, but can't find any... so far. I'll keep looking.

So you think these mandates are about enforcing public safety?

[Cwens]

Originally Posted by Mike OReilly

Having these mandatory life raft countries opens the opportunity to study what impact they are having on safety. Is anyone aware of any studies? It would be obvious to look at outcomes before and after these mandates. Could also compare places with, and without, such laws.

Tricky research as always, but it should be doable.


ADD: I went looking, but can't find any... so far. I'll keep looking.

The problem with low-probability, high-consequence events (for example, things that would cause you to need a liferaft, or airliner crashes) is that they are so infrequent that tiny random variability (a.k.a. "statistical noise") tends to drown out the signal. Airline A flies 5 million flights and has one crash. Airline B flies 5 million flights and has two crashes. Can you conclude that Airline B is, on the whole, twice as likely to kill you as Airline A? Or did Airline B just have an unlucky day?

Given the (relatively) tiny number of people sailing around in sailboats, and the even tinier occurrence of serious emergencies, I doubt that it's feasible to extract meaningful data by comparing countries that do and do not require life rafts and counting sinking occurrences.

[Cwens]

Originally Posted by Cyrus Safdari

I bet you're a hoot at parties lol this isn't a grad school seminar

It's unreasonable to make deliberately misleading statistical arguments and not expect people to call them out.

[SailingHarmonie]

For all of you who are planning on using a dinghy as your life raft, have you ever practiced getting into it from the water, in choppy sea? If you have, and you can, good for you. If you have not, you’re probably deluding yourself about the possibility.

As anyone who has practiced boarding a raft, I can assure you it’s not easy. There is a reason practicing this is part of many certification programs. But it was designed to make it possible.

A large inflatable tender is virtually impossible to get aboard in rough conditions unless prior arrangements have been made, and actually tested in a sea state like you’d expect offshore.

[Mike OReilly]

Originally Posted by Greg K

So you think these mandates are about enforcing public safety?

That's the claim. I think we can test this claim.

Originally Posted by ChrisOwens

The problem with low-probability, high-consequence events (for example, things that would cause you to need a liferaft, or airliner crashes) is that they are so infrequent that tiny random variability (a.k.a. "statistical noise") tends to drown out the signal. Airline A flies 5 million flights and has one crash. Airline B flies 5 million flights and has two crashes. Can you conclude that Airline B is, on the whole, twice as likely to kill you as Airline A? Or did Airline B just have an unlucky day?

Given the (relatively) tiny number of people sailing around in sailboats, and the even tinier occurrence of serious emergencies, I doubt that it's feasible to extract meaningful data by comparing countries that do and do not require life rafts and counting sinking occurrences.

Yes, good point. That would make it very difficult. I don't know if that means it is impossible, but I agree, it would make meaningful country to country comparison difficult; not just because of your reasoning (which I agree with), but also the multitude of other local confounding factors.

But it should certainly be possible to compare pre and post changes to the law in the same jurisdiction.

I also think your analysis hints at an answer. If the noise exceeds the effect, then statistically speaking, the action is irrelevant. BTW, you probably know all this, but a good read on this subject is a recent book by Daniel Kahneman, Olivier Sibony & Cass Sunstein: Noise: A Flaw in Human Judgment.

Originally Posted by ChrisOwens

It's unreasonable to make deliberately misleading statistical arguments and not expect people to call them out.

[rslifkin]

Originally Posted by SailingHarmonie

For all of you who are planning on using a dinghy as your life raft, have you ever practiced getting into it from the water, in choppy sea? If you have, and you can, good for you. If you have not, you’re probably deluding yourself about the possibility.

As anyone who has practiced boarding a raft, I can assure you it’s not easy. There is a reason practicing this is part of many certification programs. But it was designed to make it possible.

A large inflatable tender is virtually impossible to get aboard in rough conditions unless prior arrangements have been made, and actually tested in a sea state like you’d expect offshore.

I think the hope is that you'd be hopping from the boat into the dinghy rather than the water. But yes, that's a big concern. And it's why in my opinion, if you have a life raft and a dinghy and choose to go for the dinghy, take the raft with you if at all possible. If conditions deteriorate and the dinghy becomes a no longer viable option, you can then swap to the raft.

[boatman61]

Originally Posted by SailingHarmonie

For all of you who are planning on using a dinghy as your life raft, have you ever practiced getting into it from the water, in choppy sea? If you have, and you can, good for you. If you have not, you’re probably deluding yourself about the possibility.

As anyone who has practiced boarding a raft, I can assure you it’s not easy. There is a reason practicing this is part of many certification programs. But it was designed to make it possible.

A large inflatable tender is virtually impossible to get aboard in rough conditions unless prior arrangements have been made, and actually tested in a sea state like you’d expect offshore.

Part of one's preparation.. attached and stowed at the stern., elsewhere if one plans on keeping an engine fitted, I always remove mine.

[Liza Lee]

Further to my last reply: When life rafts were first carried on small sailboats, we didn't have such wonderful equipment as AIS transponders and EPERBs. Nearly all sinking's are caused by hitting something. IE rocks and/or ships. Most yachts these days are constructed using GRP. They are strong enough to manage almost any weather. Gone are the days of springing a plank!

[Cwens]

Originally Posted by Liza Lee

Further to my last reply: When life rafts were first carried on small sailboats, we didn't have such wonderful equipment as AIS transponders and EPERBs. Nearly all sinking's are caused by hitting something. IE rocks and/or ships. Most yachts these days are constructed using GRP. They are strong enough to manage almost any weather. Gone are the days of springing a plank!

Fire.

[Cwens]

Originally Posted by Mike OReilly

That's the claim. I think we can test this claim.



Yes, good point. That would make it very difficult. I don't know if that means it is impossible, but I agree, it would make meaningful country to country comparison difficult; not just because of your reasoning (which I agree with), but also the multitude of other local confounding factors.

But it should certainly be possible to compare pre and post changes to the law in the same jurisdiction.

I also think your analysis hints at an answer. If the noise exceeds the effect, then statistically speaking, the action is irrelevant. BTW, you probably know all this, but a good read on this subject is a recent book by Daniel Kahneman, Olivier Sibony & Cass Sunstein: Noise: A Flaw in Human Judgment.

My point about low frequency events is that you can't really use them directly to assess, for example, the effects of a safety-related change you need to find other metrics that are inherently higher frequency.


Let's say your airline decides to change the rules and let pilots fly drunk. Chances are, the the number of fatal accidents in the year before the change will be zero, and the number of fatal accidents in the year after the change will also be zero. Or, possibly, given the randomness in such things, there might be one fatal accident in the year before and none in the year after, or none in the year before and one in the year after. So if you define "effect" to be "the rate of fatal accidents," then you could argue that the noise dwarfs the effect, and the action is irrelevant and hasn't made the airline less safe.



But, on the other hand, if you measure a bunch of far higher frequency events, say for example hard landings, cockpit mistakes requiring correction by the other members, radio calls requiring repetition, incorrect settings of instruments later corrected, pilot deviations from clearances, etc. (I'm just making this list up), you'd surely see a significant difference in the "before" and "after" condition, and you'd get a clear picture that the airline is now less safe.


Admittedly, the analogy breaks down in some regard: that carrying a liferaft is unlike insisting on pilot sobriety, in that the former provides additional safety in pretty much one circumstance (the boat becomes untenable) while the latter is more pervasive in its impact. But the measurement point still holds: (and with serious respect for Kahneman et al) Just because you don't see an effect over a one year measurement doesn't mean it's not there.

[Mike OReilly]

Originally Posted by ChrisOwens

My point about low frequency events is that you can't really use them directly to assess, for example, the effects of a safety-related change you need to find other metrics that are inherently higher frequency.

Yes, I take your point (did I not say that?). Good study design always considers proxy metrics if primary measurements are difficult, or imprecise.

Originally Posted by ChrisOwens

Admittedly, the analogy breaks down in some regard: that carrying a liferaft is unlike insisting on pilot sobriety, in that the former provides additional safety in pretty much one circumstance (the boat becomes untenable) while the latter is more pervasive in its impact. But the measurement point still holds: (and with serious respect for Kahneman et al) Just because you don't see an effect over a one year measurement doesn't mean it's not there.

The drunky pilot analogy is amusing, but just speculation. What I am saying is, there is an opportunity to study the effects of these legislative changes. Since the desired outcome is improved accident outcomes, this should be open to measurement and comparison pre and post change. Sticking to the same jurisdiction greatly reduces the complexity of comparison, but I don't think it is necessarily impossible to do this kind of analysis cross-jurisdiction.

An effect not measured does not mean it's not there. But neither does it mean it is there. And I'm not sure why you're referencing just one year. I'd use all available data if I was designing such a study. The more the better.

Once again, if the noise overwhelms the effect, then the correct response is to accept that one cannot make a claim, either way.

[BBill]

Chances are you won't need it so what is your comfort level for drifting around in a life jacket compared to a life raft with ditch bag holding survival gear? If the weather is bad it may be impossible to stay attached to a floating cat or even a piece of it.

The saying we've all heard applies here...better to have it and not need it than need it and not have it.