What Is Game Balance?

Many games are released each year that commit fundamental game design errors. These games are fatally flawed from the outset, and short of a monumental marketing campaign and a small spate of miracles , they are doomed to failure. There are many obvious reasons for this kind of spectacular failure, such as bad coding, buggy software, poor quality control, and substandard graphics.

However, on many occasions, the cause of failure is not so immediately obvious. The game may look okay, sound okay, and even to some extent play okay, but it still fails commercially. One of the reasons for failure (and the one we are going to concern ourselves with here) is poor game design. In Chapter 7, "Gameplay," we introduced the elements of gameplay that we expect to find in games. We also touched a little on the subject of game balance. Including all the expected elements in a game does no good if they are not in balance with one another and with the player.

But what exactly do we mean by a balanced game? A balanced game is one where the main determining factor for the success of the player is the skill level of that player. That does not mean that random events cannot occur, but a better player should ordinarily be more successful than a poor one unless he has an unusually long run of bad luck.

Traditionally, game balance has been very much a trial-and-error process. The game is played, the game is tweaked, the game is played , the game is tweaked, and then finally, when time runs out, the game is released (and usually tweaked further in the form of patches).

So why are there no formalized , scientifically rigorous methods of game balancing? Well, for a start, it's an extremely difficult and complex process. Essentially, game balancing is a problem involving a fantastically large number of independent variables . It's an optimization problem in n -dimensional space where n is a very large number. No formal rules govern game balancing, except in a very small number of abstract mathematical scenarios.

Classical game theory is simply not suited to this kind of problem. Most areas of research concern themselves with games in which there are discrete player turns and a limited number of variables. This type of theory is ideal for analyzing games of chance, such as poker and coin-toss games, but it would be nearly impossible to use for the analysis of more complex games, such as computer games, which are more often continuous and have hundreds, if not thousands, of independent variables. Also, the majority of game theory is concerned with finding the optimum way to play a game. Using game theory to balance a game would be like playing a twisted version of Jeopardy ”starting with the answer and working back to the best possible question. (It's possible that one day there might be some sort of "game calculus" invented to handle these problems, but we're not going to hold our collective breath . Besides, that still doesn't solve the problem of how to break down the game into a list of strategies and variables to fit into the equations.)

This sort of n- dimensional optimization problem occurs in many areas of science, and it is from these areas that we can borrow techniques to help us solve the problem. This is not to say that all game-balancing problems can be solved by the blanket application of a sterile algorithm. A healthy measure of human finesse is still required in order to make a game feel "just right." Just because a result is mathematically correct does not automatically make it aesthetically pleasing.

In fact, the tweak-play-tweak method of game balancing is a valid approach (and is pretty much the only approach so far). The only problem is that this method is time- and resource- intensive and is extremely prone to error. Worse still, balancing a game is a very difficult concept to grasp. After all, what are you balancing it against? Are you balancing it against itself? The player? And how exactly are you balancing it? Are you balancing it so that it is a fair game? Are you balancing it so that it provides a consistent experience to the player no matter what her ability? The answers to these questions are ”at least to some degree ” subjective and depend upon the nature of the game. For example, a historically accurate simulation of the Anglo-Zulu wars would not be a fair game. The Zulus would have to lose, which would be a bit depressing for the Zulu player.

Even a nonrealistic game has to take some liberties in order to obtain balance. For example, a hypothetical game that simulates the invasion of modern-day Earth by a race of aliens has to give the humans a chance to win. And in spite of what you've seen in films like Independence Day , any race that is advanced enough to move huge ships across hundreds or thousands of light- years probably wouldn't have much difficulty mopping up a small planet like ours ”and they certainly wouldn't have computers that interfaced with an Apple Macintosh, conveniently allowing us to destroy their whole operation. In order to base a game around such a scenario, we'd have to stretch credibility to the breaking point, in that with today's technology we could actually defeat a race of advanced aliens whose sole specialty is enslaving entire worlds . Human " pluck " will only take us so far.

Before we get into the gritty detail, we should briefly describe what it is that we are actually balancing. There are several ways of implementing balance in a game, centered around how the equilibrium is maintained . In particular, there are two broad classes of balance that we will be discussing: static balance and dynamic balance. Traditionally, balance has not been differentiated in this fashion; when game balance is referred to, it is usually referred to as a whole, comprising both the static and the dynamic balance. Often, you will hear discussion in terms of the opening, the midgame, and the endgame, much as you would in chess. This is a perfectly valid, if a little rudimentary, approach; it's fine for the discussion of a game that has already been written, but when we are designing a new game, we would like to be able to go to a finer grain of detail. It has often been said, however, that the degree of understanding of a subject is directly proportional to the sophistication of the available language used to describe it. For our discussion, referring to balance as a whole rather than distinguishing between the two areas of balance is an unnecessary handicap that we would rather avoid.