One could just as validly argue that the very act of optimizing is itself metagaming; that doing anything to give an advantage outside what the character itself would do within its own frame of existence is simply gaming the system.
There are two kinds of build options:
Inherent, which a character, in-game, could not choose.
Learned, which a character, in-game, can decide to learn.
A Race is Inherent. Starting Ability skills are Inherent.
A Class level is Learned, as are most feats.
For Learned options, clearly this is not metagaming. Why shouldn't my character choose the best options? He's in a life-or-death occupation. He'd be a fool not to.
For Inherent options, why is choosing a set of optimal Inherent options any more metagaming than choosing a set of non-optimal Inherent options? Why would any one set of Inherent options be any more metagaming than any other? To use 4E, is playing an Eladrin Wizard somehow more metagamey than playing a Half-orc Wizard? The former is pretty clearly more optimized than the latter.
In 3e, an optimizer's dream if ever there was one,
Not mine. Optimizing 3e is trivial, and does not provide useful results, because optimization just plain breaks the game.
4E is the best game for optimization I've played yet.
I came up with (and played) some very sub-optimal character concepts - not because I was trying to be useless, but because that was the character I wanted to play and the rules almost got in the way. In one instance, a full-on optimizer got hold of one of my characters* and guest-piloted it; he later squawked at me about how terribly built it was. However, once I explained that I was trying to achieve X, Y and Z with this character he had to grudgingly admit that I'd actually done the best I could have with it.
In a more well-balanced game, your character concept would not have lagged behind optimized alternatives nearly as much.
In an ideal game, every possible character concept would have optimized expressions of them. Not attainable, of course, but that's something RPGs should strive for.
Your story expresses one of many reasons why I want game balance.
* - I was trying to build a "heavy Ranger", a woodsman in plate, but as a straight Ranger without going the PrC route.
Now, see,
this sounds like metagaming to me. Is a character really supposed to be aware, or care, that an advancement route is the "Ranger Class" or a "PrC"? Isn't it just a bunch of different capabilities to him? I've always viewed those labels as jargon for the sake of the players, not in-world distinctions.
Can you explain how this is possible? I mean, I presume you want your optimization effort to mean something mechanically....so doesn't that imply that there must be "trap" options that are non-optimized, and that your optimized character will outshine your friends' non-optimized ones? How do you envision this being possible?
If someone doesn't optimize their character, that implies that they do not care about their effectiveness as much as I do. So a small difference in effectiveness should not ruin their fun. In a game without any optimizing players, there's bound to be significant differences in character effectiveness, by pure chance, anyway.
But even if optimization only produced different, novel character capabilities, that weren't any more effective than the obvious builds (but still
at least as effective), I'd still enjoy it. That's highly unlikely to occur, of course.
The only games I have seen with absolutely balanced rules are all narrative-intense games (Capes, frex.) and are far afield from anything recognizable as D&D (although you can do fantasy with them.) Most importantly, optimization is absolutely impossible from a purely mechanical standpoint in these systems.
Does not the existence of "optimized" and "sub-optimized" characters imply imbalance between them?
I don't expect "absolutely balanced rules". I expect rules that are balanced enough that I can optimize my characters, and not ruin the fun of my friends who just grab whatever looks cool at the time.
For the most part, 4E fit this bill (at least, the closest of any edition of DnD). As long as I avoided the few obviously broken combinations (that generally got errata'd anyway), there was lots of room to play. Ideally, 5E would improve on this balance.
Also, there's a difference between "imbalance" that arises because some players care more about, and are more skilled at, a certain aspect of the game than others, and imbalance in the options available, where certain build options are presented as equal to others, but are clearly not. The former is inevitable. That's how games work. The latter should be avoided. Trap options are not good design.
Chess (provided first player is chosen randomly) is an absolutely perfectly balanced game. That does not mean that more skilled chess players won't be "imbalanced" in comparison to others.
