Nick,
I must say that my tongue was in cheek when refering to a helicopter that doesn't crash. I am also not so naive to believe that old helicopters weren't without their teething problems. But it seems that lessons are not always learnt.
Would I be right to think that where old helicopters were designed with a slide rule and a bigger safety factor, the new ones are designed by computer and a lower safety factor? If so, why haven't helicopter manufacturers recognised that a small increase in safety factor for components might reduce costly re engineering at a later date? I am thinking of one or two specific cases here relating to a new design and of course the constant battle with airframe cracking.
I don't believe that modern analysis tools can model (with sufficient accuracy) the myriad of load cases that each component on a helicopter will see.
Also I don't believe that the oil companies are a) as bothered about safety as they make out and b) know as much about helicopters as they think. Politics (and personalities) have as much if not more to do with helicopter choice than technology, safety and reliability.