First, it's important to remember that there are Many Types of Balance.
Unfortunately, what most people seem to mean when they talk about balance seems to be "combat equivalency". Which, when it comes to a fully-rounded roleplaying game, is a highly suspect measure. OTOH, I would tend to agree that the balance of pre-3E D&D was basically so nonexistent that it's a non-issue. (Case in point: When people talk about how "easy" it was to make house rules for pre-3E versions of the game, what they generally mean -- whether they realize it or not -- is that the balance was so completely screwed up that you could basically change anything you want and it was virtually impossible to end up with anything more busted/unbalanced than the RAW.)
Nonetheless, D&D worked as a game because:
(1) It leveraged spotlight balance exceptionally well.
(2) It was a fully cooperative game.
If your PC fighter and PC wizard get into a deathmatch, the "balance problems" of the game will cause problems. But if the group thinks of itself as acting like a collective unit (able to enjoy and revel in each other's successes), then the strong mechanical spotlight support of the class system pretty much carries the weight.
This is why I generally only run into problems with balance in D&D when I have two PCs who are both in the same niche with wildly disparate builds allowing one PC to be much more effective than the other. (This can generally be solved, particularly in 3E, by tweaking the build.)
(Of course, if you have a player (or are a player) who obsesses over whether or not their DPS is equal to the other PCs, then the balance issues in the rule system are going to cause problems.)
There is another factor at play here, too: 3rd Edition was the first edition to give any sort of truly meaningful guidelines for gauging appropriate encounter strength. This is a useful tool when used correctly.
When people become convinced that the only "right" encounter is a perfectly "balanced" encounter, however, the system is being abused. That's when you get people obsessing over PCs who aren't optimally built; or PCs that are performing "above their level"; or party builds that aren't capable of meeting the challenges they're "supposed" to be facing. The fetishization of balance is poison.
And finally: I think it's also true that a lot more people saw high level play in 3E than they did pre-3E (largely because I think the vast majority of pre-3E groups ignored the "XP for treasure" rules, which meant they leveled much more slowly). Since the big balance issues don't start cropping up until high levels, the problems weren't seen by as many people.
This is an effect of the internet making international discussions on a mass scale possible. If you weren't participating in that kind of community pre-3E, then you didn't see this stuff. If (like me) you were participating in similar communities, then you saw this exact same kind of stuff for 2E. (Except more egregious examples of it because 2E had more balance issues.)
You get a thousand monkeys typing away furiously at their keyboards, and they're pretty quickly going to ferret out all the corner cases in a ruleset that can be horribly abused.
The idea may have eventually been altered in order to become a balancing tool, but one can say with absolute certainty that in a game where all weapons regardless of type do 1d6 points of damage (i.e. OD&D in 1974) that any differences in weapon proficiencies were entirely a matter of flavor text.
Unfortunately, what most people seem to mean when they talk about balance seems to be "combat equivalency". Which, when it comes to a fully-rounded roleplaying game, is a highly suspect measure. OTOH, I would tend to agree that the balance of pre-3E D&D was basically so nonexistent that it's a non-issue. (Case in point: When people talk about how "easy" it was to make house rules for pre-3E versions of the game, what they generally mean -- whether they realize it or not -- is that the balance was so completely screwed up that you could basically change anything you want and it was virtually impossible to end up with anything more busted/unbalanced than the RAW.)
Nonetheless, D&D worked as a game because:
(1) It leveraged spotlight balance exceptionally well.
(2) It was a fully cooperative game.
If your PC fighter and PC wizard get into a deathmatch, the "balance problems" of the game will cause problems. But if the group thinks of itself as acting like a collective unit (able to enjoy and revel in each other's successes), then the strong mechanical spotlight support of the class system pretty much carries the weight.
This is why I generally only run into problems with balance in D&D when I have two PCs who are both in the same niche with wildly disparate builds allowing one PC to be much more effective than the other. (This can generally be solved, particularly in 3E, by tweaking the build.)
(Of course, if you have a player (or are a player) who obsesses over whether or not their DPS is equal to the other PCs, then the balance issues in the rule system are going to cause problems.)
There is another factor at play here, too: 3rd Edition was the first edition to give any sort of truly meaningful guidelines for gauging appropriate encounter strength. This is a useful tool when used correctly.
When people become convinced that the only "right" encounter is a perfectly "balanced" encounter, however, the system is being abused. That's when you get people obsessing over PCs who aren't optimally built; or PCs that are performing "above their level"; or party builds that aren't capable of meeting the challenges they're "supposed" to be facing. The fetishization of balance is poison.
And finally: I think it's also true that a lot more people saw high level play in 3E than they did pre-3E (largely because I think the vast majority of pre-3E groups ignored the "XP for treasure" rules, which meant they leveled much more slowly). Since the big balance issues don't start cropping up until high levels, the problems weren't seen by as many people.
This is an effect of the internet making international discussions on a mass scale possible. If you weren't participating in that kind of community pre-3E, then you didn't see this stuff. If (like me) you were participating in similar communities, then you saw this exact same kind of stuff for 2E. (Except more egregious examples of it because 2E had more balance issues.)
You get a thousand monkeys typing away furiously at their keyboards, and they're pretty quickly going to ferret out all the corner cases in a ruleset that can be horribly abused.
The idea may have eventually been altered in order to become a balancing tool, but one can say with absolute certainty that in a game where all weapons regardless of type do 1d6 points of damage (i.e. OD&D in 1974) that any differences in weapon proficiencies were entirely a matter of flavor text.
