Psychochild's Blog

A developer's musings on game development and writing.

11 April, 2009

Balanced but not equal
Filed under: — Psychochild @ 10:08 AM

My thoughts have been going toward balance again. It’s an interesting topic, and one I haven’t visited in while.

So, let’s take a look at what makes different options balanced, but not identical.

People often think that true balance requires a gross simplification of a system to the point where only truly balanced scenario is one where the options are nearly identical. Using classes as an example, classes are only balanced if every class has nearly identical abilities; the logical consequence is that you can’t truly balance a healer against a damage dealer in a meaningful way.

I don’t agree. In my previous posting on balance, I said:

An option is balanced when that choice is just as appealing as other options.

Now, you still have one tremendous caveat here: You have to consider multiple contexts. One option may be much better in a PvP situation, for example, but it may be balanced against another option that is much more suited to some non-PvP situation. Or, one individual may prefer one power due to his or her playstyle. But, as long as both options are equally appealing across your playerbase and considering the situations in your game, then they are balanced overall.

There are a few problems to evaluating and measuring balance. First, it’s not a hard science, but rather part opinion and part groupthink. You can’t look at game data and determine if something is balanced; things may be perfectly balanced by the designer’s model, but the players decided (or determined) that some aspect is tremendously more important than the designer modeled, and you have perceived imbalances. Players also tend to be very selfish when evaluating balance; they often want more relative power for their individual characters despite how powerful they might actually be.

So, how can different options be equally appealing but not necessarily identical? Let’s take a look at the balance between tanks in the last WoW expansion, The Burning Crusade (TBC).

As most people reading this know, tanks are there to take the damage and keep aggro on monsters. This is one of the “holy trinity” of DIKU gameplay, the other two being damage/DPS and healers. In TBC, there were three primary tanks classes: Feral Druids, Protection Paladins, and Protection Warriors. Each had their relative strengths and weaknesses and were fairly well balanced against each other. All could fill in a general tank role in a pinch in most of the normal content. The place where their differences started to be noticed were in raid content.

Feral Druids were able to assume bear form to tank. Bear tanks got a significant increase in armor value from items; a well-equipped Druid in leather armor would often have much higher armor value than a plate-wearing tank. Bear form also gave a bonus to Stamina, which meant more hit points for the Druid. Druids also gained aggro quickly on enemies. On the downside, Druids only had one “avoidance” stat, dodge, and could not parry or block attacks with a shield as other tanks did. They also had the fewest options of all classes, and could not even use health potions in dire situations. Druid tanks were meat shields in the truest sense, with lots of armor and hit points to outlast an enemy, but they had few options to get out of trouble fast. Druids also had a measure of flexibility; a properly specced Druid could go from tank to DPS roles easily by switching equipment, and could sometimes change roles in the middle of a fight if required.

For what it’s worth, I played a Feral Druid in a raiding guild in TBC. I preferred doing DPS, but I often filled in as an off-tank in guild raids.

Protection Paladins were the second major tank class. They could wear heavy armor and had some pretty nice abilities to give them extra protection. As an added bonus, a Paladin could heal him- or herself if needed. Paladins also had a very handy AoE damage ability that allowed them to grab aggro on many different enemies; this was often useful in situations where swarms of enemies had to be dealt with. However, Paladins had to balance more stats, since they needed Int for a larger mana pool, a stat that the other tanks could ignore; a Paladin with no mana could not tank very effectively.

The last of the main tank classes was the classic mainstay: the Warrior. Warriors wore heavy armor and used shields like Paladins. Warriors also used Rage to fuel their abilities, and Rage regenerated by doing and taking damage. A Warrior taking massive damage from a boss monster usually had all the resources they needed to use the abilities they needed. They also had a wide variety of abilities to help in tight spots; these were often referred to as the “Oh shit! buttons”. They didn’t have any particular weaknesses, overall.

As I said before, each class was a reasonable tank in normal small group situations. A healer might have to adjust slightly, but each was capable of holding their own if played by a competent player.

Raids were a bit different. Different raid encounters called for different types of tanks. If a tank needed to keep lots of small enemies occupied, a Paladin was often the best choice, although a talented tank of another class could also work (although they’d have to work harder). A big monster that hit hard and had unpredictably spiky damage would probably be best suited to a Warrior, although a Paladin could fill the shoes, too. Druids were best at picking up sudden spawns since they could gain aggro fast, perfectly filling the role of off-tank. A good Feral Druid who didn’t mind swapping equipment to do DPS was also useful the times when you didn’t need as many tanks.

Overall, these different classes felt balanced between themselves. They handled the basic tasks of tanking rather well, but they had different strengths in a raid situation. I enjoyed my roles in a raid, and didn’t feel that I was at a severe overall disadvantage compared to Warriors and Paladins; therefore, I thought the classes were well balanced.

Of course, not everyone is going to agree that these were balanced; many classes looked at the other class’ abilities and were envious. I know more than once I wished my Druid had a Shield Wall or Holy Shield type ability to get out of tight scrapes, or could use potions in those times when the healers were falling just a little behind. I know some other classes got jealous when comparing armor values or total hit points with my character in bear form.

This discussion also ignores strange cases like Rogue tanks, where a specifically geared Rogue gained enough avoidance to be completely unhittable. That’s not balanced, though. ;)

What is interesting is that with the latest expansion, the WoW dev team have stated that they want to focus more on evening all the roles out between the classes, including tanks. For example, the Swipe ability Feral Druids get in bear form now hit all enemies in the area of effect; previously it only hit a maximum of 3 targets. This means that a Druid can now tank large groups of enemies about as well as a Paladin previously could. Many other changes have evened out the tank classes so that none is better than the other. Druids lost the armor multiplier bonus on some of their equipment, bringing their armor values down. In addition, there has been a push to separate out the Tank and DPS roles of Feral Druids so that they have to specialize in one or the other to gain the most efficiency to balance things out; this has lessened my interest in the Feral spec for my character (and my interest in my character, overall). It’s interesting that the game feels less balanced now that the developers are taking the approach that most people say is the only way to be balanced: making everything similar.

So, what types of balance do you see that are not quite equal in games? How could you incorporate this into other areas of MMO games, even beyond DIKU style gameplay?


  1. The best example I can think of are the 5 colors in Magic the Gathering. Mark Rosewater covers the subject from time to time in his column. The most relevant articles are probably the 5 on the colors themselves, but there’s good stuff all ’round:

    The Great White Way
    True Blue
    In the Black
    Seeing Red
    It’s Not Easy Being Green

    Comment by Vargen — 11 April, 2009 @ 1:08 PM

  2. Just to get back on Magic… I never liked how it was “balanced”. The balance comes from the fact that you “never knew” which deck you’ll be facing but that each colors has it owns “unfair” advantage over the other (it just looks like rock/paper/scissor, everything is “balanced” because you don’t know the deck of your opponent). I had this very effective buried alive deck that was useless if I met someone that could destroy my graveyard.

    Might not be as true today but from the time I played, often it ended up to “did you faced the deck using the color that was meant to beat you”. Of course, in tournament colors matters not, it’s just deck building skills and getting the best cards/combo so balance between colors here matters not.

    I’m not sure if “balance” can really be achieved. Mostly because I never experienced it. Balance meaning I can play differently and still get away with “victory”. If I remember right, in SWG at one point, TKM/pistoleer (and maybe a bit of something else) was hard to beat in 1on1. Me I was a doc/pistoleer (meaning people were having a hard time to beat me and I was unable to beat anyone…).

    But a source of the problem is there. If you want balance but not equal, one class will do better than the other. But players will request to be able to beat anyone 1on1 whatever class they are. It seems they don’t want to hear that X class is better than anything else 1on1 even if yours is too powerful in a group. That’s when classes look all the same. For example, one has more HP and the other more avoidance skills. In the end, that’s just the same. It just depends on what you like most. That’s balance but boring balance. You could be of the same class that it wouldn’t changed a thing…

    That’s the boring thing about computer RPGs. Players will always find the perfect combination (they are way better at this then devs…). If you don’t want to hear complaints, you make all the calculations end up to the same result. That’s when I miss my days of AD&D. We had GMs that were getting players to fight against each other and I have great memories of beating “boosted” characters (black books anyone?) not because of my class but because I had some clever ways of using it (my favorite character was a magician that invented some stupid but very annoying spells like a sneezing spell that would remove all bonus to initiative… you can attack before I can even think about it? ah! sneezing spell! you’re as slow as everyone…). In a computer RPG, you’re limit is the rules set by the computer. Your math skill matters most than your imagination.

    So my take is that if you want balance as players see it, you’ll keep changing data over and over until all classes are the same in the end. Doesn’t matter if this priest is too powerful in a group, it’s just bad 1on1… So give multiple similar options to these players and alternatives to players that don’t mind to do bad in duels?

    Isn’t the question “who said it’s not balanced and in what circumstances”?

    Comment by Over00 — 11 April, 2009 @ 8:49 PM

  3. As you say though, aren’t there two different views of balance? What designers perceive as necessary, and what players perceive as essential? Though I’m only the latter, in matters of overall game design it irks me when players are listened to too much in these situations, since it’s quite evident that as a whole, players are out to claw as much advantage to themselves as they can (and then call it “balance” when it happens to another class, or “nerf” when it happens to theirs).

    Sure, players can provide very important information about gameplay that the people making the game may not otherwise have discovered… but beyond that, I’d say it’s foolishness to listen too closely to the general demands, which tend only toward the “make my class more powerful at the expense of everyone else” extreme.

    Balance itself is a slippery concept in games. I would *love* it if it weren’t necessary, because it would open up tons of play/playstyle/content creation opportunities; but one of the underlying laws of most games we play is that the rules need to be equitable to everyone playing (aka Balance), so that everyone playing has more or less the same experience. This works well — and is probably essential — in Scrabble and Monopoly, but I’m not so sure it really works for MMOs, in the long run.

    However, if you remove the balance-the-classes concept, what do you replace it with so that you don’t end up with a) one class to rule them all (in which case the design is flawed) and/or b) an incredibly irate player-base (assuming you have a player-base at all by that point)? I’ve been gnawing on this one for a long time, because while I am fanatical about “fairness” in general, I don’t think “balance at all costs” makes a good game — but I’ve no solutions yet.

    Comment by Ysharros — 12 April, 2009 @ 7:46 AM

  4. I’m skeptical about Blizzards push to make the tanking classes more similar to each other. I wouldn’t really like that as a player, I think. Let me explain. Players like having a unique role in their group: “I’m the one that can X” or “I’m the one that can Y better than anyone in my guild”. It’s part and parcel of feeling valuable and needed for most people.

    So the paladins had a thing they could do, tank multiple mobs better than anyone. That was kind of a niche, a point of pride. So why ruin it? What interest is served. Was nobody playing a feral druid? It doesn’t sound like it.

    When a class has abilities that have no credible use, that’s a problem with class design. When a class has good abilities, but no opportunity to use those abilities (Enchanters in EQ2 have had this both with mez and with power drain.), that’s a problem with level design.

    What’s the evidence for poor balance? Certain classes are played not just less, but a LOT less. Some ecological niches are just smaller. But if nobody raids with a Conjuror any more, then maybe something’s up. If nobody so much as plays a conjuror any more, something is definitely up, and maybe it’s balance, or maybe it’s just no fun to play the class. As always, the players could be wrong about something, it could be they just don’t understand how the class is supposed to work.

    Comment by Toldain — 12 April, 2009 @ 7:58 AM

  5. I feel like the issue of balance in an MMO with a strong tileset consists essentially of two major sources of player joy coming to loggerheads. As Toldain points out, players want to feel special, but players also don’t want to lose. The extent to which a Player legitimately perceives their abilities as special is the extent to which another Player is likely to ‘lose’, that is, to be bested in a hostile encounter with the first player, or to feel like he’d be better off playing the class/build that the first player is playing.

    One possible solution is to create ways of feeling special that don’t rely on zero-sum or high-score games, but which have social context. Requiring the recitation of a chant that only a Paladin knows, for example. A dark room where only Rogues can see. A cone of silence where only classes with no magic can speak.

    Comment by Bret — 12 April, 2009 @ 8:48 AM

  6. Balancing PvE and PvP have very different rules. In PvP situations player skill should play a large role in determining the outcome. If this is not the case (assuming relatively equal gear and level) then the players could perceive an imbalance.

    PvE is a completely different beast however. For example healers in MMOs can rarely solo at the levels that tanks can and yet most people consider healers balanced with tanks. In this case it seems that the players who are healers have a perceived benefit that offsets the ability to solo; for example they have a very easy time finding groups because they are a very critical team member whereas the role of a tank can often be filled by a number of characters though perhaps less optimally.

    In both cases I believe that the balanced but not equal is always the best choice. It may be more difficult to get right and it may be a moving target as the players’ skill levels increase, but I can’t help but feel that making everything equal is just design laziness.

    Comment by Matt Shanker — 12 April, 2009 @ 3:26 PM

  7. “An option is balanced when that choice is just as appealing as other options.”

    I agree… but “appealing” to whom?

    When this old question of balance comes up, there are always two points I want to make:

    1. “Balance” by itself isn’t a goal. It doesn’t tell you what needs to be balanced, nor does it hint at what needs to be measured to determine whether you’ve achieved such balance. The goal for making commercial games is. I think, more specific: a “balance of fun.” Emphasizing that distinction helps to keep the discussion focused on the real goal.

    2. In a MMOG. a balance of fun doesn’t imply that one player must find every single bit of content to be equally enjoyable — it means that every player is likely to have just as much overall fun playing a game as every other player.

    That second point is crucial. It speaks to the misapprehension that leads to dysfunctional designs in which everything is so similar that it has the consistency of tepid mush.

    I interpret “balance of fun” to mean that pretty much any gamer has an equal shot of finding gameplay that they can enjoy. What I think that means is that it’s OK if I run into some content that someone else likes but I don’t, as long I can easily find gameplay that I do enjoy (even if it doesn’t excite other players).

    That’s the big-picture view. The narrower picture of balancing combat mechanics that people have been using as a basis for discussion would work the same way. If I’m playing some role in combat with mechanics that are fun for me, and you’re got a set of role-based actions your character can take that you find enjoyable, then from a customer satisfaction standpoint it doesn’t matter if those mechanics are completely different — we’ve established a balance of fun.

    From a code standpoint, it could be important if gameplay mechanics within one core system very wildly. That can increase implementation, maintenance, and enhancement costs, which has indirect customer satisfaction effects. But in terms of direct perception by players of gameplay mechanics, “I’m having fun” is all that matters. Whether the kind of fun that someone else is having differs from mine is nearly irrelevant as long as we’re both having about the same amount of fun.

    In which case, I conclude that Blizzard’s changes are completely bogus and their game can’t possibly ever make any money now.

    Or something like that. :)

    So what does Blizzard understand about “balance” that I’m missing?

    Comment by Bart Stewart — 13 April, 2009 @ 12:05 PM

  8. Bart Stewart wrote:
    I agree… but “appealing” to whom?

    Appealing to the playerbase for the game, whomever that is. Before launch, that is your target audience; after launch it’s your actual playerbase.

    You can’t build an MMO for a single person. Well, you could, but they’d better be willing to pay a lot to play it. Even then it wouldn’t really be considered “multiplayer” if it only appealed to only one person, correct? Perhaps I should have been more explicit, but you cannot balance a game to only one person’s tastes.

    This is the main reason why “making a game just for yourself” is a cardinal sin; you’re making the game for an audience of one. Bad move unless you want to be the only person playing it.

    In the tank example above, a person’s individual styles of play and interests are going to probably lead them to favor one class over another (if they even want to play a tank). I preferred the Feral Druid because it allowed me to be flexible; I could fill in different roles as needed. This is something I couldn’t have done quite so easily with a Paladin or a Warrior. Yet, I think when measured against the playerbase as a whole, the tanks were fairly well balanced when I look at things through the lens of a game designer.

    [...]it means that every player is likely to have just as much overall fun playing a game as every other player.

    I think this is a less specific way of saying what I said. I think seeing what choices players make is much more concrete than trying to measure “overall fun”. I’m always wary of definitions that try to measure “fun” as if it were an element we could easily measure. Even though we have some excellent research about why people play games, I don’t think we’ll ever be able to define what “fun” is to an exact science and have it still be meaningful and useful for developing games.

    So what does Blizzard understand about “balance” that I’m missing?

    Honestly, I don’t think Blizzard fundamentally understands more about balance; they do get it right fairly often, though. But, I’m just pointing out a situation where, as a developer looking at the situation, I felt things were “balanced but not equal” in regards to the tanks. I could pull other examples from the same era of the game to show things that weren’t very well balanced (Feral Druid DPS in cat form, as an example), so I don’t think the dev team necessarily had any magical insight. I also think its interesting that they seem to be going toward the “make everything equal (and therefore boring)” false definition of balance in the current era of the game.

    More thoughts,

    Comment by Psychochild — 13 April, 2009 @ 2:53 PM

  9. Broken or Brilliant?

    [...] Balanced but not Equal [...]

    Pingback by Tish Tosh Tesh — 16 April, 2009 @ 8:06 AM

  10. @Over00: I agree, rock/paper/scissors never feels satisfying as a balance mechanism, unless the player is able to switch between each role at will or close to it:

    @Brian: I think Blizzard’s recent push for standardization of tanks and healers is driven much more by a desire to see more players included than having the game balanced per se.

    I think they’re much less worried that individual powers and specs are overpowered/underpowered than they are in having every tree of every class be someone who is allowed to join groups.

    This makes sense to me for casualish audience that WoW caters to. People get sad and quit if they feel left out. A game like EVE, filled with min-maxers makes more sense to me as a game focused on a more classic notion of balance, although I can’t say whether that actually is the case.


    Comment by Mike Darga — 19 April, 2009 @ 2:56 AM

  11. Interesting Mechanics: Multi-role classes

    [...] the Burning Crusade era was one of the better ones for multi-role characters. The system offered balanced but not equal options for characters. Warrior, Paladin, and Druid tanks had different strengths and weaknesses [...]

    Pingback by Psychochild’s Blog — 24 November, 2009 @ 3:56 AM

Leave a comment

I value your comment and think the discussions are the best part of this blog. However, there's this scourge called comment spam, so I choose to moderate comments rather than giving filthy spammers any advantage.

If this is your first comment, it will be held for moderation and therefore will not show up immediately. I will approve your comment when I can, usually within a day. Comments should eventually be approved if not spam. If your comment doesn't show up and it wasn't spam, send me an email as the spam catchers might have caught it by accident.

Line and paragraph breaks automatic, HTML allowed: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Email Subscription

Get posts by email:

Recent Comments


Search the Blog


June 2020
« Aug    



Standard Disclaimer

I speak only for myself, not for any company.

My Book


Around the Internet

Game and Online Developers

Game News Sites

Game Ranters and Discussion

Help for Businesses

Other Fun Stuff

Quiet (aka Dead) Sites

Posts Copyright Brian Green, aka Psychochild. Comments belong to their authors.

Support me and my work on