Balancing Act

Balance, what is it good for?

This is the second in my apparent series on design bugaboos. Balance is one of those things, like Randomness, that drives a lot of discussion.

Should Games Be Balanced?
This may seem like a strange question. Balance is often held up as an end point of design… a Holy Grail that, even if you can never reach it, should always be sought after.

But before we set our course by that distant star, we have to ask ourselves if it’ll lead us somewhere we want to go or lead us anywhere at all.

Going Nowhere Fast
The problem you’ll soon find is that balance is not a big concept. It’s a very, very small one.

I could say chess is balanced, but I mean the players’ starting positions are balanced. The pieces were never intended to be balanced, the turn structure is unbalanced, and strategies are not balanced against each other.

“Is this game balanced?” It’s what we’re supposed to be asking, but it rapidly becomes “is this subsystem balanced” and “what kind of balance are we actually going for anyway?”

So the star we’re supposed to be letting guide us is just “what does balance mean for this game?”

Set some goals, design towards those goals, and adjust when you drift off course. Balance is just like any other design goal you may have.

Not a Holy Grail, just something you might want to pick up. Not a quest, just a journey.

Hearts and Feathers
Balance can be a pretty good metaphor, though. It’s one way to remove extraneous factors from the equation so you can weigh specific qualities.

By balancing things, you can remove, say, character selection as a variable and get the result based purely on tactical skill and chance. Or, purely on tactical skill if there’s no chance involved.

On the other hand, if you want character selection to be a strategic consideration, that consideration is defined by local imbalance. Without pairings that tilt one way or the other, the character you’ll be facing becomes extraneous information.

So, in some cases, we want to actively avoid balance. To put it another way, if rock has a 50/50 record against scissors, that’s balanced but you need to fix it.

The Optimizer’s Dilemma
Given any game, a lot of players will drift towards the mechanically optimal route. Even people with other priorities that come first will usually factor in mechanics as a tiebreaker.

Depending on what sort of balance you’re designing towards, balance can be an tangential to optimization or directly opposed to it.

The latter is easier to see. If the choice in the moment is balanced, your optimization buys you nothing. All your thought and energy get you is “it doesn’t actually matter.”

This is important, because “it doesn’t matter if you choose A or B” is actually worse than “A is clearly better,” the theoretical bane of balanced games. At some point, you had to figure out that A is better and now that information is serving you well. Having A always be better isn’t itself the best design, but it’s also not the worst.

Better Sometimes
In other cases, balance is achieved by making different options best in different situations. That’s great for deep gameplay and challenging optimization.

But the balance is actually extraneous there. It doesn’t matter if each option gets an equal piece of the pie, just that they all get a piece.

Personally, I think having one option with most of the pie can be helpful in solo or cooperative play. It provides a stepping stone between choosing randomly and having a full understanding of the situation.

But in a competitive game, the top level of choices needs to form a balanced network. It doesn’t much matter what the bad (or even slightly suboptimal) choices are doing, as long as there is a group at the top that forms a healthy metagame.

Which is an interesting topic in its own right: how you can use balance and imbalance to create the sort of competitive landscape you want. And, of course, to what degree it’s the designer’s role to make that call.

On the Balance
So, this comes back to the main point: balance is a test to be applied after deliberation, deciding what you want balanced and how you’re going to test it.


Tags: , ,