In a badly balanced game, given that one class is significantly weaker than the others, the solution becomes to simply not play that class if you ever want to win. Or if one class is significantly better than the others, the popular option becomes to simply play that class. That sort of thing happens. In fact it already happened in PBEM, with Necromancers dominating hard enough that a community PBEM rebalancing patch was made.
Yes, that sort of thing happens. Among competitive players, who only see a sense in playing if it is for victory.
And unfortunately, competitive multiplayer is the standard way to go when balancing/designing even for titles where it makes little to no sense as matches just take too long for the average player of such games* - like AoW. And that then leads to balancing being done as you do for a competitive scene - until every choice is valid as every choice is more or less the same.
What could instead be done is to focus more on cooperative play, maybe having persistent online campaigns that don't even require you to play at the same time (as long as you do not meet each other in battle), or so many other things... that could then be balanced with a focus on how interesting and varied factions are instead of how you can bring them closer to each other to make sure none are too strong/weak.
Balancing asymmetrical factions for competitive gaming is an illusion. It isn't possible. Never has been, never will. And no, StarCraft or WarCraft does not have asymmetrical factions. They have some minor differences, that is all.
*In these cases, I'm not talking about the average player per se, I really mean the average player of turn based strategy games. That is
not a large group. For most gamers, that just involves too much thinking and/or is too slow.
No, even the average player of TB strategy games doesn't want to invest the time that would be required to play matches regularly. Especially not competitively as that would mean an even larger time investment.
By that logic DotA does not have a competitive scene either. After all the overwhelming majority (90%+) of players are casuals following the same standard build paths / hero guide for their heroes every single game or are just dicking around randomly rather than strategically adjusting their playstyle to fit the situation, lineup, and tempo of the match. Yet there is a decidedly pronounced presence of competitive play within the community which informs both game strategy and game balance on the whole.
You are comparing a game that was made entirely for competitive online gameplay with a game that the majority of people play alone. DotA style games do not even have a world, a setting, a story, a campaign, offline scenarios, etc. Well, some may have an attempt at a world, but... I think that discussion would soon end in "lulz".
You cannot seriously believe that in both cases the game should be balanced the same way, with competitive gaming in mind.
What's your point anyway? Never doubt the power of stupid people in large groups so you'd better bend over and cater to them? Stupid people don't even know what makes a game good anyway. I think what you're really saying is that you just don't like competitive scenes so you'd rather downplay and discount them.
My point is that you live in your little competitive bubble, obviously oblivious to the fact that most people playing your game don't give a shit about competitive multiplayer.
And as such, it makes no sense to balance with that in mind. If you are looking to improve your success when balancing, balance for what most of your players actually want.
Just compare those two (and this time in a way that makes sense):
AoW 3 -
http://steamspy.com/app/226840
Players in the last 2 weeks: 27,858 ± 4,932 (4.32%)
Players total: 580,912 ± 22,510 (90.05%)
Playtime total:
40:52 (average) 09:53 (median)
DotA 2 -
http://steamspy.com/app/570
Players in the last 2 weeks: 9,962,506 ± 92,286 (8.57%)
Players total: 116,182,201 ± 276,460 (100%)
Playtime total:
195:31 (average) 04:22 (median)
Now, ignore the total number of players and focus on the percentages and times:
1) The average total playtime of people who actually played liked the game (didn't stop after a few hours). Do you honestly think those 40 hours on average were spent on competitive multiplayer? For real? Compared to all the alternatives like campaigns, scenarios, co-op, etc. ?
The reason the average playtime is so low (well, still way above average on Steam, but anyway...) is certainly not the lack of perfect competitive balancing. It is more the lack of things to do that interest the players - within or without multiplayer.
In DotA, it is pretty clear they were spent online in competitive gameplay - as it is the only choice available. You just don't play such games "just for fun" as trying out anything non-effective will just get you flamed. These communities are... famous for that
It makes sense here to focus on that kind of balancing.
2) The percentage of players still actually playing is about 9% for the purely competitive online game. It is 4% for the game where multiplayer is just one of many available choices (and within multiplayer, competitive gaming is just one of the choices).
Let's say that part of players is 1% (which would be 25% - most certainly waaaaay too high, but let's roll with it) and keep in mind the average playtime.
You want a game to be balanced focusing on 1% of all players. Does that make sense to you?
Wouldn't it make more sense to bring in changes that would increase the average playtime because they make things more interesting to the rest, not just "more balanced"?
Again, you are part of a (unfortunately vocal) minority. Have a look for example at what Mr. Vincke said about development and balancing for D:OS 2 in interviews - looking at forums, it seemed that many people did not like something. But looking at their actual statistics, most people were perfectly fine with it. It was just a vocal minority in the forums and they decided not to listen to them too much. And it worked out perfectly - just have a look at the average playtimes for the game: Already more than AoW3 in average despite being much younger - and whopping 28h median! Certainly most people buying the game actually play it for more than two dozen hours. Some right decisions were made.
Now, I really tried my best to make it obvious even to you that to focus on a marginally small part of your game and its community for balancing is just.... downright stupid.
The problem here is that "competitive gameplay" so far is the only available, "tested" theory. It is what all developers go for, because all other developers go for it and so far nobody has dared to step away from it**, even if it is obviously wrong to focus so much on it.
Meanwhile, some people realized that shortcoming and are waiting for the day when it finally won't dominate the thinking behind balancing any more - because it almost never made sense to begin with.
Now, you might want to decide to keep living in your bubble, and there is no official statistic proving even to you that most players of the game do not even touch the multiplayer button (because they do not want to play competitively and think it is the only available choice to them in MP, because it almost always has been).
I wish there was, trust me. And I actually think a lot of companies have that statistic. Some of them should just decide to release it so we can finally move on.
**Actually, IIRC Command & Conquer 4 was developed with co-op campaigns as a focus. Unfortunately the game was utter shit, completely changed the series' formula and so that interesting focus went down mostly unnoticed with the shitty rest of the game.