Archive for October, 2010

The Balance Myth

Today my thoughts turn to the concept of balance in an MMO. This topic invariably comes up (usually within the first week after a game launches) in almost every game I’ve ever played. That includes MMORPG, MMORTS, and even single player games!

Defining Balance

This may not be as easy as you might think it is. The core concept is that each player wants to have an equal chance at success with their chosen character. However, as they say the devil is in the details. A PvP player doesn’t want the outcome of a 1:1 battle to be dependent on which class or set of powers a character has. Another PvP player may extend that to the equipment that the characters are using shouldn’t matter.

PvE players are not immune to calling for balance either. In this case, they may complain that one class has an easier time leveling than another. Or that when there are multiple classes which can fill one of the traditional roles on a team (healer/tank/dps) that specific classes are more highly desired than others in filling raid team slots. This is often identified as being “unfair” to players who choose characters of the “non-preferred” classes.

Is It Possible?

Can balance be achieved? Certainly in theory it can. Games like tic-tac-toe, checkers, and chess show that it is possible to create a game where the outcome of the game does not rely on the abilities of the pieces (each side is completely equal) but on the skill of the player.

In an MMORPG, the balance is much more difficult. To truly create a balanced system, you would need to have equal powers for all characters. That could mean a classless system where everyone picks from the same pool of powers. All this would do, however, is lead to the “Build of the Week” where you’re replacing a class selection at character creation time with power selections. So you could set it up so every character has the same static set of powers. Or even if the power names, graphics, and animations are different the underlying data (the numbers that really matter) could stay the same.

MMORPGs have also always suffered in the balance area from the leveling mechanism. A level 2 character doesn’t stand a chance against a level 62 character. Several fixes have been attempted to eliminate this including buffing or debuffing one of the two characters, or making it so attributes and powers don’t scale according to your levels. Level locking combats to those within a certain range of levels also mitigates this to a degree, but unless the level ranges are very narrow indeed there will always be an advantage to the higher level character.

Finally the issues of equipment (armor, weapons, etc.) has raised balance issues. The person who spent the last 8 weekends raiding the same dungeon for 16 hours a day has a logical expectation that the armor set that they gathered is better than something a more casual player picked up from an NPC vendor for a few coin. It’s a reasonable expectation, but it makes balancing much more difficult.

Game design aside I’m not convinced it is possible for any online game given technical issues involved. Even if every player has exactly the same abilities in their character, there will still be advantages that favor one over another. Perhaps a player has a better Internet connection and lower latency to the server. The data sent between their client and the server travels faster and therefore they have an advantage of being able to react sooner to what their opponent is doing. This could be true even if the network connections are equal, but one person’s computer is two years old and the other one just bought a new Alienware machine this week.

Then, of course, there is player imbalance. If my reflexes are faster than yours, is it balanced that you lose because I can move my thumb 5.2ms faster than you can? Or maybe it isn’t twitch based, but because you can analyze the battlefield and identify terrain that makes winning easier (i.e. you’re smarter than me), you have the advantage. Most would agree that if we’re getting to this level, then the game balance has been achieved, but that won’t stop people from complaining that they shouldn’t lose just because their slower or dumber than their opponents.

Do We Really Want It?

Could we overcome all of these issues? Certainly. Here’s what I think it would take:

  1. Use a turn-based system administered by the server. Eliminating any network or PC benefits.
  2. Provide all characters with the same static power sets. No classes or power choices; so all characters are equal.
  3. Eliminate loot or equipment stats. If there are equipment differences, only in aesthetics.
  4. Eliminate leveling / progression. A character that has been in play for 3 years has the same power strength as one playing for 3 minutes.

My question is, do players really want to play that game? Yes, I suppose some do, but they are playing first person shooter games (and even then those players would have issues with point #1). Someone choosing to play an MMORPG or MMORTS usually wants to have choices. They typically rant and rail against feeling like the game designers are forcing them into a specific character choice or combat tactic.

Rock – Paper – Scissors anyone?

The typical solution to the balance issue is the Rock Paper Scissors game design. The concept is that a particular character class has strengths and weaknesses. One class’s strengths are specifically designed to play against another class’s weaknesses. Class designs have to be very carefully considered, mathematical simulations need to be designed, etc. And even then the balance isn’t as easily reached as one may think.

Let’s take the common example of a warrior archetype versus a ranged dps archetype. The warrior has great melee attack damage, but their range is limited. The ranger has good range to hit targets, but they have weak armor against melee attacks. On paper this looks like a good balance between classes. However in game play, the complaint usually is that the ranger can “kite” their opponents (meaning strike from range, and keep moving so the warrior never gets close enough to hit them) resulting in the ranger usually winning. So the designers counter that by giving the warrior a rooting or snaring ability which keeps the ranger from running away. But now the warrior always wins because the ranger never gets to use their strength (ranged attacks). So they give the ranger a way to break free of a root / snare, but on a cooldown which means they can’t use it all the time.  Let’s say that the design is so well done now that each character can inflict the same damage per minute against their opponent over the course of the battle. If both players play perfectly, it should end up more tic-tac-toe than rock-paper-scissors.

But now that those two are balanced, how did adding the rooting ability affect the warrior versus mage combat? Or what happens when it isn’t 1:1 but 2:2 or 4:4? Now the balancing has been complicated exponentially with all those potential combinations.

How Do We Solve This?

I believe that any design which achieves perfect balance will have to do so by eliminating player choices and uniqueness. While such a design may be balanced, it will not make the game more fun to play.

Having the game log data about combat and then mining that data may help somewhat. However say that we find that rangers are killing warriors 4 times as often as the other way. Does this mean that rangers are overpowered? Does it mean that warriors are underpowered? Does it mean that rangers are standing off while their warrior buddies lock up on the other warriors and just happen to be good at getting the final hit in for the killing blow? Does it mean that warriors are targeting healers and ignoring the rangers?

Player feedback (via forums, surveys, etc) can also help. However my time in the MMO community tells me that the number of players who can objectively analyze their own characters and determine that they have an unfair advantage over their opponents is vanishingly small. Much more common are those who complain loudly because they lose a lot. And at that point you don’t really know if they are complaining because of a real issue or just because they play badly.

I think the only way to really know is for the game designers and developers to play the game. These are the people who are most able to determine if the gameplay is falling according to their design or if some unforeseen complication is making things unbalanced despite their best efforts to the contrary. But they have to play on the public servers against other opponents. I say this for the same reason that I say that programmers are terrible at testing their own code.

If a game designer is playing a game against other people who worked on the project, they know how things are supposed to work. They will naturally tend to play their characters in the way they were intended. Only by playing against other players who are not privy to the design decisions and development implementations will they get a chance to see how the “real” players are using their creations.

And even then, they will need to consider all the unintended consequences to changing something in the game. It will never be as simple as “The Circle of Death is too strong, it wipes out cloth wearing characters immediately, we need to nerf that damage way back”. Because doing that may make that ability totally ineffective against heavy armored characters. The effects of even simple changes are much more complex than it first appears.

Any Hope?

If you are an MMO player, I’m afraid there will never be a time when you see a diverse complicated gaming system that is perfectly balanced on launch day. We should resign ourselves to the idea that adjustments will need to be made continually. After all, we are in effect changing the game ourselves as we discover new tactics and combinations of things that the designers hadn’t considered when they were building the game.

Furthermore, given that each new content expansion and balancing patch itself introduces new factors into an already complicated system, it seems that the holy grail of balance will continue to be something we pursue, but may never attain.