Looks pretty reasonable, but it might make more sense to remove necromancers, cabal leaders, and early death outliers from the sample since they have one of the following: artificially long lives (cabal leaders), have a player caused age death so the number of hours is necessarily inaccurate (necromancer) or likely fell victim to +age gear, or other such manipulations that resulted in age deaths of under 200 hours (early death outliers). My assumption is that you are trying to get a good idea of the normal age limit of characters so it makes sense to remove those characters that can be determined with a degree of certainty to have had either a premature age death or had an extra long life due to being a cabal leader as those characteristic are not determined at character generation.
I think the over all results will then be a more narrowly defined range for the normal expected lifespan.