.
I was merely explaining that it is not sufficient to divide (catastrophic events) by (fleet flight hours) in order to verify the probability of an event.
"The probability of an event" is not a well defined concept. You need to ask a specific question, then use historical data appropriately to get a quantitative answer.
Dividing the number of catastrophic events by the number of fleet flight hours gives you the answer to the question "What is the probability that an A330 will experience a catastrophic event in any given hour chosen at random from a spectrum of flights similar to those the fleet has undertaken so far?"
This number wouldn't be appropriate, for instance, to figure out the probability of a catastrophic event in a flight that routinely experienced severe turbulence (by which I mean turbulence like the AF flight encountered - I may have the terminology wrong), because the underlying sample of flights is not representative of that situation. For that, you should ask how many A330s encountered severe turbulence to date and use that as the base population.
However, if the question is "should we ground all A330s today?" dividing the catastrophic incidents to date by the total flight hours to date is not an unreasonable way to gain a quantitative piece of information.
Insofar as long haul aircraft probably all fly roughly the same spectrum of routes, there doesn't seem much evidence that the A330 is more dangerous than the others. One question would be "why ground just that one?"