RPG Design – The Illusion of Game Balance

   A lot of work goes into the search for “game balance”. Answers to “frequently asked questions”, errata, rules patches, rules revisions, and entire new editions appear in an endless stream in attempts to “balance” printed games while online RPG’s constantly revise, recode, and expand.

   The problem is that “balance” in a complex rules system is as elusive as perpetual motion – and the two searches bear a strong resemblance, right down to the long-discarded ideas which get dragged out again and again by new would-be machine or game designers.

   For an example, lets take a look at the largest, longest-running, most well-supported, and most elaborate and throughly patched rules system in history. It’s been worked on by tens of thousands of well-educated and dedicated designers, testers, and modifiers for many decades. It constrains itself to real-world physics, no special powers, a simplified version of human history, and only covers humans (and thus avoids all the complications of coming up with systems for magic, psionics, and similar fantastic elements). It’s the most elaborate and throughly researched rules system in history.

   It’s called “Law”.

   Now, anyone out there who thinks that the law has a shorteage of loopholes, unbalanced situations, and exploits should go and watch the news for a bit.

   You really can do better than that in a game system. The more you limit the options and player input, the more balanced you can be.

   Here’s a well-balanced game system:

   Roll a six sided die. On a 1-3, I win. On a 4-6 you win. To avoid player manipulation of the roll, we’ll have a random shaking machine do it and test the die to make sure that it’s fair. If something weird happens to the die or machine, get another and reroll until a winner emerges.

   There. If we eliminate any real player input, make the rules absolutely rigid, limit things to an extremely simple situation, and make sure that the physical apparatus is unbiased, we CAN have a balanced game.

   Of course, no one will be interested in playing unless there’s a prize involved. After all, there’s no reason to.

   Lets move up a bit. Take Chess. Chess is pretty balanced isn’t it?

   Well no. Now we’re allowing the player’s skills to have an input. That makes the game interesting – but in actual play suddenly things are no longer balanced.

    Relatively recently, Kiril Georgiev played a total of 360 games simultaneously – winning 284, drawing seventy, and losing six – during a marathon that lasted fourteen hours and eight minutes. In the 1920’s Alekhine played blindfold chess against 28 teams of four players each, scoring 22 wins, three draws and three losses.

   That kind of imbalance really won’t do for a casual recreational game. It’s not much fun.

   We can reduce this problem though. We can make sure that the game involves a good deal of hidden information and has random factors. That will lessen the advantages offered by player skill without eliminating them – thus retaining interest longer and adding the welcome element of uncovering mysteries to the game play. Sheer statistics will still offer a massive advantage to skilled and clever players if the game goes on for long enough though.

   You want the game to go on for a very long time and yet you still want weaker players to stick with it instead of opting out?

   You can stretch things a bit by making it a team event, so that stronger players will tend to carry the weaker ones along – but that’s a stopgap at best. The players who aren’t making as large a contribution, or who find themselves constantly in the background, will tend to lose interest anyway. It’s always the stars of the team that get the most attention.

   A sufficiently simple game may be able to get along with some sort of handicapping system, such as you find in Golf. The trouble with that is that you have to get people to play a good deal, and keep accurate records of it, to establish those handicaps – and even so the upper levels of golf require constant rulings on variations in the courses and equipment. Just as importantly for our purposes, golf games are relatively short – so chance still plays a substantial role.

   To sustain a truly long-term game, what you really need is an active human presence, who can make exceptions or overrule the letter of the rules when he or she needs to correct a blatant imbalance. As a side-benefit, they can regulate that hidden information option.

   In law, that individual is called a “Judge”. You don’t want people opting out of the law and going in for vigilantism because they see the law as “unfair”. Similarly, you don’t want people opting out of your game, whether it’s by single-handedly ignoring rules they don’t like (“Cheating”) or by not playing at all. Whether as a publisher or as a buyer, you’ve invested time and money in the game – and you want people to play.

   That’s also why online MMORPG’s keep getting revised. They still have judges, they’re just a little more remote than they are when everyone is sitting around a table and implement their decisions by code changes instead of by announcing a decision and making a few notes.

   So: unless you have a group of puzzle-obsessed theoriticians, if you want a truly long-term game, you really can’t get along without a Judge. It’s going to be his or her job to manage the hidden information, correct imbalances, plug – or allow if they’re fun – loopholes, and compensate for the differences in skill and luck between the players.

   I’ve had a player or two that I – in second edition AD&D – gave twelfth level characters to when the rest of the players were using characters of under level six. Those players planned so poorly, and fell so in love with their own poor ideas, that giving them far more powerful characters was the only way to keep their actual in-play contributions in line with those of the other players.

   That game continued for several years – good evidence that the solution worked. It was “balanced”, albeit not in any way that most mechanical rules systems would recognize.

   There’s only one final jump to take. A lot of games simulate a world of sorts. Chess can be taken as an (exceedingly) abstract simulation of medevial politics and warfare. Other games, such as Axis and Allies, or Seakrieg, or many other wargames, are far more explicit.

   There are two basic viewpoints on that – the gamist view, which says that “this is a game, with mechanical rules, and that the setting is irrelevant flavor text”, and the simulationist view, which sees the game rules as being there to provide a necessarily-limited simulation of a world.

   The gamist view has a simple answer to players who want to try things that the rules don’t cover, or complain that the rules aren’t accurately simulating a situation. “It’s a game and these are the rules”.

   That’s the classic answer for games like Chess, where questions like “where does the knight go if I move it over the side of the board?” are meaningless – and for a lot of games (in fact, for almost all of them that don’t use a referee) it’s normally (if not quite always) very much the best answer.

   I completely scandalized the chess club back in high school when, during the course of a quick game with a friend of mine who wanted to improve (and was losing as he pretty much always did), he said “Hey, lets change this! Knights are Queens, Rooks are Knights, Bishops are Pawns, Pawns are Bishops, and Queens are Rooks!” and I said “Why not?”. We continued on that basis quite amiably – but there was widespread protest and (literal!) outcries of “You can’t do that!” from the rest of the chess club, who took chess far more seriously than we were.

   Still, some people do ask such questions – and that’s why there’s Fairy Chess.

   The Simulationist view pretty much requires a judge, and – barring unusually cooperative players – won’t work in games like chess which don’t usually have one. You and a friend may be able to agree that “a piece which exits the chessboard leaves the battlefield, and can reappear on any edge square fifteen turns later by traveling around in the wilderness. If the king flees the battlefield, however, the battle is lost” – but it’s hard to get that kind of cooperation even when you’ve only got two people playing. Trying to get it in a group of five or six is close to impossible.

   One of the great benefits of using a Judge in a game system is that a good Judge will allow the players to exercise their creativity. As soon as you start allowing that kind of thing at all though, you’ve made the jump to a role playing game. The players might be taking the roles of gods, or generals, or entire armies – but now they’re playing the role of an entity or entities that exist in an alternative world and are exploring solutions and moves that the rules don’t cover.

   Since you’ve got to have a Judge for a role playing game, you want people to be willing to take the job without being paid for it – which means that you’ll want to make it an enjoyable and relatively easy job.

   That means holding down the number of rules, ensuring that you’re using general systems and principles that are intuitive and easy to apply, and leaving the game master an active role in the game – characters to play and rulings to make. You’ll want to avoid shackling his hands with detailed rules for everything, massive tables of modifiers, or piles of sourcebooks to sort through. That sort of thing isn’t much fun.

   That’s sensible enough anyway. There may be a few gross oversights in any system that need patching – but endlessly tweaking things (that are almost always working just fine for everyone in your audience who really understands the role of a Judge) in search of “balance” only works until people start playing with the rules and find the new exploits. The more special cases, patches, and rules you add, the more exploits you’ll create. You’ll never achieve more than a (very temporary) approximation of “balance” through mechanical means.

   If you could, law libraries would fit into a few feet of shelf space, rather than taking up small buildings.

   Sadly, recognizing that limitation is in direct opposition to the notion of getting more money from your audience by publishing an endless stream of supplements and expansions. True mechanical “balance” may be an impossible fantasy, but it’s often a profitable one. Issues of “balance” – usually quoting issues that only a tiny segment of the participants or Judges involved have any problem with – provide a wonderful justification for an endless string of new rules.

   In law creating new, and ever more specific, rules to deal with rare or irrelevant issues can carry politicians to re-election, and the endless volumes of material that such efforts produce provides employment for swarms of lawyers, clerks, and secretaries. In games this sort of effort usually involves publishing an endless string of new sourcebooks.

   A game system can put up with a few of those – but too many will start to strangle it as they overburden the Judge. A new edition will set the process back for somewhat, but the only way to stop it permanently is to create a basic set of rules that’s as complete as possible in just a few books and publish nothing much more except setting material, how-to-use-it guides, and scenarios after that – all of them entirely optional and setting-specific.

   You doubt that? Look at law again; professionals spend lifetimes studying the law – and you probably can’t pay your game judges to spend a lifetime studying your game system. Eventually all the expansions get to be too much, and they have to start over. In games, that’s called “buying a new game” (or at least starting a basic game without using all that source material you’ve invested your time in – and without buying any more).

   In law that same process is called a “revolution”. It takes longer to build up to that point because people have a lot more invested in their social system than in your game system – but the result is the same in the end.

   Personally, I’d recommend against trying to take that route – but then I hope that people will still be using and playing my games many years from now and I never expected publishing games to make me rich anyway.

   “Balance” is a chimaera and an impossible dream, but it’s one that an awful lot of players and Judges believe in. Unfortunately, you can never find “Balance” in any simulationist game system that’s complex enough to be really interesting to play – and there’s no point in wasting your time looking. You’ve got a Judge. If you’re in need of “balance”, go ahead and let him or her rule on the issue and forget about consulting another book of rules.


5 Responses

  1. Hi, thanks for a thoughtful post. I don’t entirely agree with your conclusion, though I do agree with you that in complex systems “Game Balance” is almost impossible to achieve. Mostly because the requirements for balance get expodentially more complicated with every rule that gets added to a system. However, that said, I think it is possible to create a relatively simple system that still covers most traditional RPG tropes, and make that simplified system reasonably well balanced. I say “reasonably well balanced” because balance is actually for the most part subjective. Even in a reasonably simple system, people will quibble about what factors should be balanced over others. Such as, are even hit points for all classes more balancing than weapons damage being even across all weapons? Bad example, I suppose, but you get the idea. Anyway, that said, a fine and thoughtful post. Thanks.

    • Oh you’re welcome. I’m glad you found todays random thoughts interesting – although I suppose that there actually is a way to keep any game system, no matter how complex, balanced.

      All you have to do is leave it in the box. The trouble comes in as soon as you introduce players, since they too are a part of any game system that calls for them – and there’s no way to balance them.

  2. Totally agree with game balance as an illusory idea that can’t be achieved. The comparison of RPG rules with Societal Laws was genius. I wish I had made the connection.

    The most profound of your observations, for me, is that of Player skill.

    One additional consideration is player-conduct. It’s easy for the social-leader of a gaming group to hijack a game they’re not enjoying by digression, derision or denigration.

    So, as you rightly say, it is about the co-operation of the game group to achieve their gaming preferences. That’s the way real game balance occurs, IMO.

  3. Rules exist because people are frequently not competent to make compromises or decisions without solid guidelines. Otherwise we could handle everything with gentlemen’s agreements instead of Laws.

    Problem being, then, that there’s no guarantee the Judge (Gamemaster) is even minimally competent with the basic law he’s supposed to be using as a guideline.

    See as evidence all the people who castigate Psionics and ToB as broken and call for “core only” games of D&D 3.5. When classes from neither splatbook can even hope to match a “core single classed Druid with natural spell” in power or versatility. And meanwhile the Druid’s animal companion (a class feature that isn’t even her main feature), is better at fighting than the core Fighter class.

    • Sorry this took so long to get back to; I’ve been unwell for a bit.

      I can’t entirely agree there; in role-playing games rules exist to establish a setting and the expectations of the players; people who aren’t competent to make decisions generally have too much trouble with their own lives – unless they’re young or handicapped enough to be totally under the care of others – to worry all that much about role-playing games.

      Of course, getting along without formal rules and by “gentleman’s agreements” is actually fairly easy for human beings; it’s their natural behavior pattern. Admittedly, like many other natural behavior patterns it stops working after the small-tribe level – but I have yet to see a playing group that didn’t fall within that size limit.

      Now, the notion that the judge needs to be particularly competent with the rules he or she is supposed to be using as a guideline is a secondary delusion based on the original delusion that mechanical game balance is possible; a good game master can run a good game by making up a rules system on the spot. I’ve done that plenty of times – and gotten requests to revisit those worlds and “systems”. Most judges will need more to work with than that – but the principle is the same; a good game master can produce a good game from bad rules; a bad game master will produce a bad game, no matter how good the rules may be.

      As far as “see as evidence” goes… that proves the point perfectly; if “game balance” wasn’t a subjective illusion, it could be measured, evaluated, and one side or the other of such debates proved wrong. Instead, all we see are endless anecdotes and personal opinions – demonstrating subjective “evidence” of balance or imbalance with respect to particular groups of players. Yours own statement is a case in point: I’ve seen optimized fighter designs who could easily dish out five or six hundred points of damage a round; I have yet to see an animal companion build do that.

      Does that make such builds unbalanced? In an MMORPG run by a computer it would – but in an actual game with a live game master, no it doesn’t; it simply means that the judge needs to make sure that the players continue to experience the illusion of balance – just as he’d have to do if one character was built poorly.

      Some approximation of mechanical balance may make maintaining that illusion easier, but it’s not required for it. Rules serve the same purpose as a stage magician’s gear; they help with the most impressive illusions, but a good stage magician can do tricks with a few ordinary coins, cards, and pieces of cloth that will amaze an audience withouth any elaborate mechanical tricks.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: