Happy to explain.
>10 years is really open ended and non specific. Think of how ridiculous their graph would have to be if they had a diver at 11 years, 2 at 15, then 1 who had been certified for 30years. What are they going to do? Have 20 "0"s? Silly from a presentation standpoint, so they consolidated.
Lets take a look at your 3rd point for a moment. 35% under two years. 2. Thats it. The next group (under 10, more than 1) is 5 times that. So a bit less than double the % roughly (35% to 60%), but five times the number of years we're looking at. Its not apples to apples.
Back to the 2nd point, if there are more divers certified in the 'mid range' as defined as greater than 2yrs but less than 10yrs, why is that segment so under represented? If all groups were equally likely to experience a fatal accident, we'd see a higher % than 40%. Heck, looking at the divers with <1 and 1 is nearly that value alone, and its a WAY smaller time period.
Its a little late, so I could be off a little here, but I think its fairly clear cut that year for year, the newer a diver is the more likely they are to experience a fatal accident. There's probably some sort of upward trend eventually (complacency factor), but our sample size is just too small to show it. We don't have the data to draw a meaningful conclusion about that.
I hope what I wrote makes sense, and I'll be happy to clarify anything if need be.
A little bit more precise language would help in this discussion. As far as I can understand, the question is how the risk of death changes depending on the level of experience. Suppose we pick a diver completely at random. The probability space is the set of divers. The probabilistic events of interest are as follows.
A = diver had 0-2 years of experience
B = diver had 2-10 years
C = diver had >10 years
D = diver died in a scuba-diving accident
What we know from data that Omission has posted:
P(A) = 0.35
P(B) = 0.25
P(C) = 0.4
What we know from page 64 of the DAN 2009 report:
P(A|D) = 0.3
P(B|D) = 0.3
P(C|D) = 0.4
What we want to know is the ratios between P(D|A), P(D|B), P(D|C):
P(D|A) = P(A|D)*P(D)/P(A) = 0.86 * P(D)
P(D|B) = P(B|D)*P(D)/P(B) = 1.2 * P(D)
P(D|C) = P(C|D)*P(D)/P(C) = 1.0 * P(D)
Conclusions:
Divers with 0-2 years of experience are at a lower risk compared to other groups, divers with >10 are at a slightly higher risk, but divers with 2-10 years are at the highest risk overall. The risk is similar between all groups, to within ~20% of each other.