Hi Ed:
I'll comment. First, I find .pdf format hard to read on my browser so I didn't even look at your link until you asked for comments. (Wait, that wasn't a helpful comment, was it? Sorry.) Second, the document was about cave diving and not diving in general. Useful, if you're just interested in cave diving, but not at all useful if you're trying to answer a different question. Third, it doesn't clearly state to my read exactly what they're measuring (except when they discuss fatalities). What exactly qualifies as an "incident" to be counted?
Finally, and most importantly, it gives a lot of interesting numerators, but it doesn't give any denominators (except when they discuss fatalities per incident) so the information isn't really helpful in making informed decisions. Take for example their "Cave Certification" chart that shows a total of 190 incidents. The way they label the graph is confusing to me, but it would appear they're saying of the 190 total incidents reported 47 occured in divers with some type of cave/cavern training and 143 occurred in divers with no cave/cavern training at all. So it's safer to have training, right? Well if there were 1,273,856 cave dives by non-trained individuals and 47 dives by trained individuals giving an incident rate of 0.0000007% for the untrained and 100% for the trained, maybe you'd decide to forego training.
Of course my example is ludicrous, but it shows how important it is to know all of the relevant numbers before you make a decision. If no one knows the numbers and there's no way of finding out the numbers, then "statistics" don't really help and can be (sometimes intentionally?) misleading.
HTH,
Bill