# Balance is not about Viability This is a reply to [this post](http://www.sirlin.net/articles/balancing-multiplayer-games-part-1-definitions.html), which is apparently no longer accepting comments. "Viable options" is a poor way to characterize the kind of choices that a balanced game needs to have. All the options in rock paper scissors are viable; there are even variants where you have 5 or 17 viable options rather than just 3. But none of those variants is any more balanced. I think the defining characteristic of a balanced game is *choices which are matters of skill*. That rules out both choices where the "correct" option is obvious, and choices that are made effectively at random. It also goes some way towards explaining the problem with some theoretical "multiplayer solitaire starcraft". Such a game would not really be "unbalanced"[1] as long as there were still interesting choices to be made about your build order. But once the optimal build order was "solved" it would cease to even be balanced, because the game would now just be a matter of rote memorization rather than skill[2]. This also provides a more solid argument for why it's good for games to have some "bad" choices - to make them balanced for beginner play. Imagine a game where you choose to play one of five cards each turn, which all depend on previous cards and your opponent's cards and the cards remaining in the deck in some incredibly complex way which (to a beginner) might as well be random. Such a game might well be balanced in high-level tournament play, but it wouldn't be balanced for beginning play in exactly the same way that rock paper scissors is unbalanced - because there is no skill element to the beginner's choices. Compare that to (Japanese rules) Mahjong, which I've recently been playing quite a lot of. Every turn you have, theoretically, 14 options for which tile to discard. Some of them are obviously terrible choices - discarding a red 5 when you have a plain one would be just crazy[3]. Discarding part of a set when you have unmatched tiles available is *probably* a bad idea - though it might make sense in high-level play if you'd figured out your opponents' hands, or were taking a small chance of getting a much higher value hand. As a beginner you learn rules of thumb like discarding end tiles rather than center tiles; as your skill increases you get better at feeling the shape of your hand and inferring more from your opponents' discards, and start to understand when to break those rules. So there's a whole range of levels on which you can be thinking about the game, but at every level there's skill involved in making decisions, and every increment of skill - whether the difference between a beginner and a novice, or between an expert and a master - gives you a better chance of winning. (Better *chance*, mind you. The flip side of balance is that a game must not simply be won by the most skilled player all the time - if everyone knows who's going to win before the start, there's no point playing the game) Note that the decision here is never about choosing between "viable options" - or at least, it never feels like it. Rather, one seems subjectively to be figuring out the unique best discard - the one where no other options *are* viable[4]. When I really can't tell which tile is better, what I feel is not freedom or balance but frustration - the game has reduced itself to the level of rock-paper-scissors. (Which seems like a perfectly good game under Sirlin's characterization - it features not only perfect balance, but Yomi level 6!) [1] Not that I think it's a good game - a game needs more than just balance [2] Ignoring the manual dexterity that would be quite central to such a game in real life [3] Ok, *technically* if you were in second place and the fourth-place player was close to 0 you might want to lower the value of your hand to avoid wiping them out and ending the game [4] With different levels of ability one might reach different conclusions in the same situation. But the subjective feeling is always the same, and I think it reflects objective reality - in the overwhelming majority of game situations, there really will be a single "best" discard that one should be picking [Home](/) <div id="disqus_thread"></div> comments powered by Disqus