Originally Posted by
whiskeybravo
After some number of weeks the percentiles begin to become gear locked. It doesn't matter how well you play if you can't get an orange until you get weapon, therefore percentiles to do not automatically represent a given level of "skill". After some point in time your item level and gear simply has a much larger influence on breaking through to higher percentiles. That is why gear acquisition must play a role in the spread of the statistics. If the more dominant jobs are gearing up faster, it pushes their higher percentiles to a further degree compared to jobs which aren't being played as much. By the very same token as people start gearing up more jobs and running them through, the higher percentiles get pushed further out of reach for those lacking the same gear. Even if the jobs are played at an identical skill level the higher percentile is out of reach until you get the gear. To further illustrate this point, we can compare the number of e4s logs between DRG (10,312) and NIN (2,399). This would certainly seem to indicate there are more geared DRGs out there than geared NINs, and so the overall average of the 95th percentile DRG would be higher than the overall average of 95th percentile NIN - because the 95th percentile NIN doesn't have as many players pushing it's average up. Another way of putting it would be NIN is further away from realizing it's full potential compared to DRG. It doesn't have to be a large degree for this to make sense, as I stated previously even 1% is relatively meaningful within the current operating margin.