Upcoming Events
Unite 2010
11/10 - 11/12 @ Montréal, Canada

GDC China
12/5 - 12/7 @ Shanghai, China

Asia Game Show 2010
12/24 - 12/27  

GDC 2011
2/28 - 3/4 @ San Francisco, CA

More events...
Quick Stats
84 people currently visiting GDNet.
2406 articles in the reference section.

Help us fight cancer!
Join SETI Team GDNet!
Link to us Events 4 Gamers
Intel sponsors gamedev.net search:

 What is Play


 The Basic Play
 Balance Process


 Printable version
 Discuss this article


Once a game is macrocalibrated, game balance must be fine-tuned. If the game is at least somewhat fun, without any huge glaring problems, then it's probably macrocalibrated and ready for the small details. Microcalibration is the process by which the game designer applies small tweaks in order to perfect the state of balance of the game. A tweak can be defined as the set of changes such that the change in value is less than 10% for a "global" value (something that effects many game elements) and less than 30-40% for a "local" value (a particular game element).

The biggest challenge in microcalibrating is simply identifying problems. Once a problem is identified, it is just a question of tweaking values slightly, and in such a way that one is not creating new problems. Good element modularity and pre-planning will really be a life-saver at this stage - without them, it may not be possible to balance the game in a reasonable time frame.

Identifying Minor Imbalances

There are several techniques with which a designer can identify minor imbalances. The most obvious of these is to simply test the game heavily, and watch for approaches that seem consistently favored or dominant, and to watch for approaches that are never used. Another common method is to simply argue hypothetical situations and counters with a tester or another designer, reach an agreement as to what is supposed to happen, and then test it to see if that is how things work out in the game.

If a designer is using the first method, of simply looking for dominant (or never used) strategies, it is important to determine exactly why the strategy is either dominant or never used, and furthermore, whether or not it is supposed to be that way. Additionally, classifying the imbalance as one or several of the typical types of imbalances is helpful as well in understanding the problem. Ultimately, the more certain one is of the imbalance and its characteristics, the more able one is to fix it.

In recent years, it has become increasingly popular to record game outcomes and statistics secretly without player knowledge. Age Of Empires, several games published by Sierra, and Strifeshadow have benefited from this technique. Sometimes such statistics can be enlightening, and sometimes they can be highly misleading.

All data must be taken with a grain of salt. In some cases, an immature testing population can give you very skewed results, simply because they aren't familiar with the game and haven't gotten around to trying things (or just rush to the easiest things to use). Similarly, an overly mature testing population can either be so set in their ways as to ignore the potential of other strategies, or get caught on a very advanced, obscure point that is indeed imbalanced, but perhaps less pressing than other, more obvious imbalances. One highly effective technique that Ethermoon Entertainment used with Strifeshadow was to overstate play balance changes in beta patches, so as to encourage players to actually try new strategies, as opposed to continue to "write off" changes.

The second method of spotting imbalances, which sometimes is referred to as "hunting for imbalances" is when a hypothetical situation is formulated, and the various possible responses and their outcomes are agreed upon. For instance, it might be agreed that a tank rush should beat a light vehicle rush, but with light losses, and at the same time lose badly to the anti-tank infantry counter. If in the actual game, the tank rush completely massacred a light vehicle rush, and broke even against anti-tank infantry, there would be a perceived imbalance of tanks being too strong. Hunting imbalances is very important, and if done rigorously, can easily spot 75%+ of minor balance problems. Just because a designer expects something to work a particular way, doesn't mean it necessarily does in the game - especially when a very small difference in a local balance value separates balance from imbalance in a competitive multiplayer game!

One thing to keep in mind whenever hunting imbalances is that game elements that come into play early in the game is always much more balance sensitive than late game elements, simply because an imbalance early game element will affect everything after it, while a late game element has much less of a timeframe in which to cause trouble. Just as one needs to macrocalibrate a game before one can do useful microcalibration, one needs to balance early game elements before late ones.

Fixing Minor Imbalances

Once an imbalance is identified and demonstrated, fixing it is a pretty simple thing… assuming the game is designed to be easily tweaked! A very balanceable game has the property that the designer can specifically attack an imbalance, without collaterally affecting other game elements.

The most important thing to keep in mind when tweaking is to keep it at a tweak level (think small), especially when an upgrade is involved. A game element that is too strong can easily make many other game elements useless, while a game element that is too weak is just ignored and impacts little.

It is also important to implement tweaks that do not cascade onto other values. For instance, consider a spell in a role-playing game called "fireball", which is a subset of fire spells. If fireball is proving too powerful, the options a designer has range from downgrading the potency of fire magic in general, to downgrading fireball in some manner. Clearly, you want to choose the localized solution, downgrading fireball, before tweaking fire magic in general. This is a highly simplified case, but in many situations there will be a certain level of interdependence between game elements. Carefully consider what impacts a change will have, and try to use the one that specifically attacks the problem without affecting other game elements.

Finally, avoid "Over-solving" imbalances. "Over-solving" an imbalance is when a designer applies multiple different types of tweaks at once to fix one particular problem. Over-solving makes it very difficult to determine the effects of the changes simply because you are using multiple independent variables to affect one dependant variable. Over-solving is also a recipe for trouble in terms of accidentally affecting other game elements.

Wrapping it up

When developing a game, it is easy to occasionally lose sight of the end-objectives when confronted with an enormous onslaught of details. Remaining true to the desired gameplay of the project, and to play balance theory in general at all times is quite trying, but ultimately necessary to ensure both high quality play balance and a beta period of appropriate duration. With multiplayer games becoming more and more popular every passing day, it is becoming increasing necessary for play balance to be as good as it can be. Too many promising multiplayer games have already been compromised by lackluster play balance.

About the Author

Tom Cadwell is the Lead Designer at Ethermoon Entertainment. Titles he has worked on include Strifeshadow, a multiplayer RTS (http://www.ethermoon.com), Dark Reign 2 and several web games and MUDs. Tom welcomes questions and comments regarding this article, and can be contacted at tcadwell@alum.mit.edu

Credits and Thanks

Thanks to Erin Daly and his associates at Relic Entertainment for providing feedback on this article. Thanks also to Dave Custer, my college technical writing professor, and the members of my writing workshop.