Note: When I refer to asymmetry below, I’m only referring to old school pick-your-character-and-keep-it type asymmetry as found in fighting games and co., none of this new-fangled character-switching-in-match, random select, or draft picks. I think those things are flawed as well, but that’s a discussion for another article.
Much of the difficulty and entertainment of decisions comes from the in-game context that frames them: in chess, for example, there’s a move called castling. It happens to be a very strong move in many situations, and in games between the best chess players, castling often occurs. Because it protects the king and develops a rook in just one move, there is no dispute that castling is a top tier move.
However, it’s not always correct to castle. In a given state in chess where you could castle, you are often better served by making an aggressive move on the center of the board or developing a different piece. Sometimes you want to avoid giving initiative to your opponent, so you’d rather wait to castle on a later move. There are a lot of different game states where you can castle, and less (though still a lot) where you should castle in order to win. It’s determining whether a given state falls into the second category that makes the decision about whether or not to castle an interesting one.
If for some reason there were only ONE gamestate, or only a few gamestates where you could castle, having that tool in your toolkit would be less interesting, because as soon as you’ve been to all the positions where you could castle, you will come to a conclusion as to whether or not castling is a good move in each of those positions. Once that’s figured out, you’re no longer making any new decisions when you come to a gamestate where you could castle; you just go to your five or six item list of gamestates where you can castle and look up the solution you’ve already figured out.
In many “asymmetrical” games, though, you’re asked to do exactly that: the beginning of a game is the only time you’re allowed to select a character, and every gamestate at the beginning of a game is exactly the same. Even if the game has fifty characters, the decision of which one to use only shows up in exactly one gamestate, so it’s extremely context free, and the correct character choice will be the same every time.
Sometimes it’s interesting to look at each of the characters and test each one out in order to determine which one is the best. That’s a lot of fun, and it definitely requires a good understanding of the system to make that determination. Unfortunately, as soon as the “best” character or characters are determined to a reasonable level of confidence, there’s not much to be gained by thinking about it any more. As players become more and more certain what the correct answer to the “which character should I pick” question is, the question becomes boring and solved. A lot of questions can show up in games which will eventually be answered to a high degree of certainty, and that’s fine, but hopefully those questions don’t show up predictably at the beginning of every game, which is what this question does.
If the game turns out have some characters better than others, then if players are playing to win, a tremendous amount of the content of the game, in the form of characters, will never see the light of day in a serious match ever. That’s a massive failure, and you might as well not have included those characters in the game in the first place.
Some games are very well balanced, and someone might raise the objection that in those games, asymmetry is ok, because the characters are so close in power level that the question as to which character to pick is much harder, and because more characters see play in those games. To those people I ask—really? As far as I know the competitive game that I’ve played with the best balance is Yomi, the fighting card game from Sirlin games. Towards the end of that game’s life, the best players were only picking a few characters (more than one because the best characters had different matchup spreads) in serious, high level tournament matches, even though the game was very well balanced. So I challenge anyone who seriously raises this objection to bring me a better balanced game where their objection holds true.
Even if they did, though, they then run into a second problem. Great, your game has 5 characters that are so close in power level that when deciding which one will give them the best chance of winning, a player is indifferent. You’ve just created a decision which, by definition, has NO EFFECT on a player’s chance of winning. From the player’s perspective, this is a meaningless decision. If the decision is so meaningless, there’s no reason to include it in the game at all. (Side note: at this point I think you should make character select a forced random, so that your absurdly well balanced content doesn’t go to waste.)
Another objection that some people might raise is that different players have different “pet characters”, or might be more competent at one character’s playstyle than another’s. If you’re better at doing thing A than doing thing B, that’s fine. My game can be about JUST thing B, and have nothing to do with thing A. It’s not my responsibility as a game designer to put activities that you’re already good at into my games. If you prefer doing thing A, though, that’s a different ball game, because it is my responsibility as a game designer to put enjoyable activities into my games. In this case, though, correct response is likely to make separate games about doing thing A and thing B. There is no reason that I should feel pressure as a designer to force these things into a single game.
That’s the long and short of my objections to asymmetry. In spite of them I’ve enjoyed many an asymmetrical game, and I don’t think that asymmetry takes anything away from a game except elegance, though I don’t think it adds anything in the long run either. This is a contentious topic, so if you have your own opinions about the merits of asymmetry please enrich the discussion and make them known.