Tiers?

[Gnorman]

I think people too frequently overlook the amount of optimization that is inherent in even knowing what spells to use. I know when I first picked up 3.5, it was all shocking grasp (5d6!!!) and omg I want to use meteor storm! Knowing what spells are force multipliers or battlefield controllers requires optimization. Do you remember when TLN came out with his treatise on going after saves, divination use, and some generic party buffs? That stuff was revolutionary, and really no less different than the kit optimization that Truenamers require.

I think you're overstating the difficulty here. It doesn't take much system mastery to realize that Sleep is better than Magic Missile. A chance to take multiple foes out of the fight entirely vs. 1d6+1 damage to one of them? That should be obvious to even new players. It does require a basic framework of mathematics and probability, but then so does any other class.

It's really not hard to optimize a wizard. Many people do it accidentally.

A wizard has a demonstrably higher number of choices for dealing with problems. Even a straight-up blaster is liable to take spells that attack Reflex saves and spells that attack armor class (without even perhaps considering the rational behind them), whereas the fighter is restricted to the one.

I think it's also common to forget how hard it can be to pick the number of appropriate spells on a prepared caster. In my experience, I find T2 and sometimes T3 more practically powerful than T1 as the global optima is almost impossible to reach without resorting to foul cheese. A wizard may have a tool to use on every situation, but having a larger array of hammers tends to work better in D&D, especially if there are constraints on taking the 9 hours to reshuffle spells. I am not entirely sure where I am going with this line of reasoning, other than saying in my experience, prepared casters often end the day with spells they don't cast because they couldn't find a use for them. Spontaneously turning them into nature's allies is totally OP, of course, but then I've played for years with a druid who cast a couple goodberries and maybe cure light wounds. Maybe. Never once did anything memorable with spells. I'm not sure if that falls into some special terrible player category, but I think I'd definitely put it on the optimization spectrum.

Just because you didn't play the class according to its intended function (spontaneous nature's ally, in this case, being that intended function, which is supported by the ability existing in the first place) doesn't mean that doing so is overpowered. It may be for different reasons.

Seriously, we need to stop putting a value judgment on "optimization", just like we should not put a value judgment on "roleplaying." Everybody optimizes. Picking Power Attack instead of Spell Focus on a fighter is "optimizing."