MoonSong
Rules-lawyering drama queen but not a munchkin
Again, take a look at the wizard class. Compare it to the other full casters.
The cleric and the druid have their sprellcasting detailed under a class feature called "spellcasting". But the mage doesn't. It has it under "wizardry" which includes casting using Int and the spellbook.
So it would be super easy to replace "wizardry" with "sorcerery" and keep the same class but cast with Cha and not have a spellbook.
Oh, and look what the subclasses are called. It's not "School of Enchantment", it's "Wizardry: School of Enchantment". Suggesting there could be subclasses called Sorcerery: Wild Magic or Sorcerery: Elementalism.
This seems to be going in circles, my example was meant to show that the d&d wizard has never been a generic wizard, but rather it's particular brand of wizard that doesn't lends itself to be other things. Second, I've already detailed on this same thread that it isn't just the thing placed uinder "Wizardry" that is wrong for the sorcerer and warlock, EVERYTHING (ok probably overstating, a lot of things ) on the class write up is a poor fit, from Esoteric lore (shouldn't be enforced into sorcerers), to scrolls(not very befitting of warlocks nor sorcerers), and potions (maybe could fit if we see warlocks as witches), even the narrative assumes a scholarly caster. In other words the Mage is a D&D wizard and fails to cover sorcerers and warlocks propperly. (Not to mention Light armor and all simple weapons don't quite make up for the loss of ritual caster, and that spell mastery doesn't work outside a slot based preppared caster)
And one more thing, more than the spellbook, Spell Prepparation runs counter to what a sorcerer is, sorcerers are about being reactive and adapt on the run, not about the careful planning that results from prepparing spells.
Still my point remains, the data, even if considered good, doesn't says they can be safely removed, actually the oposite, if someone cared to analyze the numbers we'd find most classes but rogue and wizard are nearly the same within what is statistically significant, a 1-2 percent difference is small taking into account the error margin is very likely 5% or something. Now since this is also skewed towards 4e players it proves nothing of OSR, 2e and 3.x players.It was on the WotC during the run of 4e, so it likely largely favoured 4e players.