Hmmm.... Seems to me that you're kind of disproving your own point. The averaging idea is implicit in your own statement, i.e. you are effectively saying that, on the average, "being weak at lower levels balances out with the great power they get at upper levels" (your words, not mine). Read without that clause or in any other context, you are effectively saying that the class is balanced because it is unbalanced at 1st level (too weak), unbalanced at 2nd level (too weak) ... unbalanced but in a different way at 14th level (too strong), unbalanced but in a different way at 15th level (too strong), etc. That makes no sense at all. In fact, MUs are really only balanced - kind of - for a couple of levels somewhere in the middle, unless you do look at it as some kind of average across all levels. Sorry, but you can't have your cake and eat it, too.
I'm not disproving my point at all. Again, the flaw in your argument is that you're looking at the MU levels as a pool of stats thrown together and
then averaged out. My argument is that since the game is not played with an average (you spend time playing at every level individually), averaging is the opposite of what you should be doing. I.e., if a class is weaker 50% of the time, and stronger 50% of the time, it's flawed to say the average is balanced (which seems to be the argument you're ascribing to me) because at no point are you really playing that theoretical average character. You're either one or the other. And in the
end result, after an entire campaign, it all works out in the end, then it would be what I consider a balanced class. And I think they were in AD&D, otherwise you wouldn't have each class played often.
Think of it like this. If you have a car that is slow off the start but has a high top end speed, and a car that is quick off the line but a lower top end, those are not the same as a car that is in fact average at both, even if all three reach the finish line at the same time. All three are in fact balanced in that end game context, regardless if one car was much better at first and not as good at the end, and it's very likely the
average speed was never reached. An average, by it's very definition, is analysis only and not reflective of what's actually occurring.
And none of this addresses the fact that the MU may never even make it to a higher level, because of death, retirement or whatever. What is balancing out your low level wimpiness, then? Unless you subscribe to the notion that it is "balanced" on a sort of macro level by all the other NPC MUs who will get past that point. If so, then Arilyn nailed it with her demographics comment. Or do you assume that it's "balanced" because the player's next character might do better? (Assuming she plays again. And she plays an MU. And gets further next time.) 'Cause, if so, well.... ugh. Just... ugh. That is truly absurd game design.
It was addressing that. EVERY character may never make it to higher level in AD&D, and most did not. And even if MUs died at lower levels more often (not my experience in 35 years of playing AD&D, but whatever), putting in that work to get a more powerful class at the end game is in fact the same sort of balancing I'm talking about. Also, tens (if not hundreds) of thousands of gamers disagree with you and think it's just fine game design. I don't like every rule in AD&D, but the class design is just fine for what the game is supposed to be about. And I'm confident I'm not nearly the only one who thinks so.
And, as Arilyn pointed out, what about the poor halfling fighter? How can being stuck at 4th level possibly be considered balanced, when the rest of the group is, say, 12th level? There will never be any point in the halfling's career that he will be terrifyingly powerful enough to balance that out. And, even if there was, it would only be through the very averaging process you were denying. And it would still be terrible game design.
Well, halflings could get to 6th level, not 4th, but again the whole point of those was to emphasize a human centric world. I.e., most of the PCs were intended to be humans. That was a tool to motivate and incentive players to choose humans over other races. And in AD&D, 6th level was pretty high. 9th level was name level, and the end game for many players.
At the end of the day, I just don't get your argument. How can having classes/levels/whatever that are unbalanced for some portion of the game and then unbalanced in the opposite direction for another portion ever equal balance on the grand scale and good game design? And if you are looking at it on some sort of macro scale, involving thousands of other similar non-player characters... Well... why? The players are using their own characters, not those faceless hordes, so that's where any balance mechanics should focus. That's just common sense game design. Otherwise, as Arilyn said, you're talking demographics, not balance.
As for the "if you don't like it, ignore it" comment... Well, obviously. But I got the impression that this conversation is about the RAW, not somebody's house rules. That, too, seemed to be implicit.
I don't know anything about your gaming history, but I get the impression that you never played AD&D back then very much, if at all. Otherwise many of these things would seem pretty obvious. What I hear you saying, is the argument that is the same as the argument "unless every PC can be just as good as every other PC at every pillar and every task, then it's not balanced and horrible game design." That's the argument you're making when you're saying that unless a class is as effective as every other class at every level, then it's not balanced. And if you don't think every PC needs to be as effective as every other PC in every pillar and in every scenario, then it seems you're either holding a contradictory position, or you yourself are also drawing a line, just in a different place than me. In either case, it doesn't make the game horrible design to not fit exactly where you are drawing your own personal line.
And my position is that if you have a PC that is better at exploration, and one that is better at combat, and one that is better at interaction, and you play all equally, then the game is in fact balanced. You need to look at the entire game, in all pillars, for the typical duration of an entire campaign. If a player excels at certain points of the game but doesn't in others, that doesn't mean the game is not balanced as a whole.