Comprehensive list of all accidents

A few people, me included, have raised this before. I thought that Alex had addressed it (it does change the Inspo stats considerably). Somewhere, there is a post that breaks down the fatalities into a) non-backlit, old battery box Classic, b) upgraded Classic and c) Vision/Evo. The vast majority of deaths, IIRC, was on a).

I can understand that outsiders may view all Inspos as one unit, but I don't think that is true or fair (particularly to APD). APD have addressed some serious flaws in the early Inspirations and the death rate has fallen. For that reason, those units should be separated out and due credit should be given. For one thing, it shows that any manufacturer can turn around issues on a unit.

Alex,

Please split them out. I'm sure you said in the past that you would. Apart from communicating a more meaningful picture to purchasers of a unit (or god forbid, the families of divers who pass on while diving Inspirations), it'll reduce the impression that you are on a crusade against APD and Martin. I am not crusading against anyone, other than for all companies to improve their safety and adopt basic safety practices in their business.

Cheers,

I would love to split them out.

The only reason I have not, is I do not have the data to do so except from a confidential source that I can't use.

In gathering the data the third party sources (i.e. not the confidential source above), has not have the breakdown of Vision or Classic in most cases. If anyone can provide that, I would separate the two sets of figures immediately. My impression is that Vision stats are making Classic stats look better, which has got to be good in that Vision is APD's newer product.

Alex

well anything pre vision can't be vision for a start.
why not mark every other one as classic, vision (where known) and inspiration where not known.
At least that would be a start.

I think Phil's point is totally legitimate... if it's for lack of numbers from manufacturers then lets stop putting up with this BS and start pressuring the manufacturers to put out the numbers of units produced, known fatalities and dates of design change... Give them a date to have the data together, relaease all data from all manufacturers simultaneously to ensure no undue mud slinging and publically call out the manufacturers who refuse... we'll do our part to hammer the ones who refuse with "encouragement" and reward those who do with less speculation.

the fact that we are left with so much to speculate is no fault of our own, times have changed and so should the standards.

IMHO

Do we know when Classics started leaving the factory with backlit handsets and the new battery box? That could be a split point, though some incidents on units after that could have been on unmodified units (and clearly were). <snip>

Thanks for offering to do it. Any chance of going back to your confidential source to ask for permission to do it?

Cheers,

... Thanks to Kevin (UWSojourner) for a discussion on this (still ongoing) in helping ensure figures are presented in a way that can be understood and assumptions are listed clearly.

Some units have had a huge increase in sales in the past year, particularly the Meg ...

To take examples of the effect of these assumptions:
1. Meg, 300 units in use based on 50% sales. Linear growth gives mean population of 150 from start of 2003 to 2008 (5 years). In latest accident list we have 12 fatal accidents listed for the Meg (up one, unfortunately).
This gives 750 dive years of experience, with an average risk of 1.6%.


Alex

My problem with this discussion is that we have no way to tell for what kind of and how many dives the units are normally used. Thus all the -very earnest- approaches at calculating a risk factor per unit seem to me rather futile, even if and when we had all other relevant data.
Personally, I bought the HammerHead to ultimately be able to do dives I can´t do now with my rather elaborate trimix OC equipment. A year of accident free deep/long/cold/... "tec" dives would show up in this statistic with the same weight as my first (this) year on a rebreather with only shallow/no deco/normoxic dives is going to count.
I still believe that collecting and publishing all available data is a good thing and will improve our ability to identify fields of possible improvement and risk factors when diving a rebreather but from a mere statistical point of view I´d rather not try to calculate a number representing risk/unit.