Of course it was careless of me not to qualify that my statement was made on the basis of a moving average - not a single year. As has been said statistics can be manipulated; for example the width of the averaging window could be used to elimate (or include) significant data. The Chinook accident has been used to illustrate a point in a previous post but the UK CAA uses a data window of 10 years - that accident has therefore not featured for a significant number of reports. Arrival of buses is also not an analogy that I would use - the whole reason for moving averages is to smooth the data. Could it be because of the rising
trend of accidents that attention is being focused on the GOM. Fatal accident rates are not reported on the basis of numbers of passengers carried - although that methodology has been used recently in an attempt to legimise the use of singles. Fact is that with the small number of hours flown by helicopters, skewing is always present. (The reason that the fatal accident rate has not significantly improved in the UK North Sea over the last couple of years is that as the Puma moved out of the (10 year) window the S76 moved in (but there were still only two fatals in that 10 years)).
My curiosity is aroused by your questioning of the OGP in their data collection and reporting. Whilst they do use the ICAO/FAA/CAA definition of serious incident and accident, the quality of their data and analysis for the GOM is better than that of the FAA/NTSB - whilst Mr Tucker can tell you what the 5 year moving average accident rate per 100,000 hours is, the FAA/NTSB definitely could not - unlike some other countries the US has no requirement to report usage. Where there might be a discussion of the representation of incidents/accidents is the removal from the analysis of engine failures that result in a ditching - not reported in the accident figures in accordance with the ICAO/FAA definition (which does need to be addressed to include engine failures in singles).
As we have heard from Nick Lappos, the safety of the more modern FAR 29 aircraft has been enhance by the requirement for a design assessment of the dynamic parts, and the introduction of fault tolerance (neither required on FAR 27 singles). That, and the gradual introduction of Vibration Health Monitoring, should ensure that the single point of failure problem is reduced on these sophisticated helicopters.
I would also disagree that the use of more sophisticated aircraft and the required levels of skill and training could lead to an increasing number of accidents (unless your comment is specifically addressing maintenance). One of the points made by Nick and others is that the increase in complexity should not lead to an increase in workload; this and the extensive use of CRM in multi-crewed helicopters should make it easier to fly and not the reverse (like some others on this thread, my skin has also been saved by the two crew concept). I would qualify your statement about singles being safer than twins and might say that their use could be more appropriate for some activities.
However for me one of the most important statements made on this thread was by Hippolite
Strict enforement of weather minima through operational control, adherence to standard procedures and SOP, helideck regulation, improved weather reporting and communications will go a long way to improving Single Engine accident rates in the GOM. Singles should be able to safely operate in a benign environment within a certain distance from the beach under the majority of weather conditions. For those areas and conditions which fall outside certain parameters, twins should be used with 2 pilots.
to which I heartily subscribe.