Neonchameleon
Legend
One interesting example here is the barbarian class. Introduced in 3.0 I find the 3.0 and 3.5 barbarian entirely surplus to requirements - there is nothing in the conceptual design of the 3.X barbarian that a 3.X fighter shouldn't be able to do. So shoehorning in the barbarian probably stole design space from the fighter and made the fighter worse.Yes, well, I have the converse issue, where I find that a lot of people are incredibly keen on class reductionism über alles, even when that means shoehorning together things that would really, really benefit from being separated. E.g., the people who say we don't NEED Paladins and Rangers, because they can just be divine and nature-y versions of Eldritch Knight; and we don't NEED Barbarians, because they can just be a rage-centric version of Fighter; and we don't NEED Clerics and Druids and Warlocks, because those can be just divine/nature-y/patron-linked Wizards with a stat swap; and we don't NEED Monks because they can just be a pugilist Fighter type; etc., etc., ad nauseam.
Parsimony is not the only virtue in game design. Sometimes, giving a concept or an idea the space to stretch out can make the difference between "functional" and "bad" (or at least dull/ineffectual/etc.; consider the Purple Dragon Knight/Banneret subclass, which has been generally panned as a pretty ineffectual attempt to kludge a "Warlord Fighter" that isn't based around maneuvers. The big problem being that the Fighter is so hard-coded to do a lot of damage by itself, have a lot of personal bonuses, and take a lot of actions personally, that it's really hard to build into it much in the way of ally support.
By contrast 4e said "we've got this thing called the barbarian. What can we do with it?" And then dropped it in the primal power source, effectively incubating it into a real class that both didn't overlap with the now more focused fighter and allowing designers to explore the space. Which is how we get things like Storm Herald and Totem Warrior barbarians in 5e.
So just hacking off a bit of design space from the fighter IMO meant that the 3.X barbarian made the fighter worse. But actively incubating the barbarian in 4e meant that it's now a worthy class on its own in both 4e and 5e.
And sometimes it's not this bit of the aesthetics that matter but the desire to explore them.Now, of course, this gets into thorny, aesthetically-driven arguments. But parsimony is already an aesthetic argument. "Less is more" is only true when, y'know, you actually DO get more out of it. Sometimes, surprisingly enough, less is less.