Game Design: Playing on the Seesaw

In my previous post, I talked at some length about balance, particularly in single player environments. This time around, I'm going to dive into multi-player balance. As previously explored, the core of single player balance is making sure that the player continues to be able to make interesting and valid choices. This continues to be true in multi-player games, but requires additional care.

Before diving into multi-player examples, lets take a quick look at how Skyrim would fare if it were a multi-player game. A rogue (specializing in daggers) would be able to instantly kill any other player, as well as defeating NPCs faster, allowing them a greater rate of progress. While in single player, the other styles of combat are effective, they would have little to no defense when competing with a rogue. This shows that while Skyrim is balanced for single player, it is not balanced for multi-player.

Let's start simple. Rock-Paper-Scissors (RPS) is a balanced game. Given equal skill, each player has a fifty percent chance to win. If I throw rock, I win (and you lose) if you throw scissors. I lose (and you win) if you throw paper. Each move I can make has one move that it beats and one move that it loses to. This system of balancing is cyclic and is used in many games.

In classic RPGs, the Rogue beats the Wizard, the Wizard beats the Warrior and the Warrior beats the Rogue. In Age of Empires, Archers beat Cavalry, Cavalry beat Infantry and Infantry beat Archers. These configurations show up all the time, and are often a good design choice because players can recognize the pattern. Most people are familiar with Rock-Paper-Scissors and so can grasp this concept and use it in their planning fairly quickly. These systems can be larger than just three options, for example, the World of Warcraft pet battle system, which has ten different options, each of which is strong to one type and weak to another.

There is also the Pokemon weakness and resistances. Unlike the other listed systems, the Pokemon weakness and resistances are not perfectly balanced. As you can see, if you look at the effectiveness of attacks on Normal type Pokemon (the first column), and average the non-standard values (so one x2 and one x0) you get a result of x1 damage, which is standard. If you look at the effectiveness of attacks on Fire type Pokemon (the second column), and average the non-standard values you get 1.06. This is offset by their strength in attacking, which shows that their attacks on average do 1.25 times base damage. If you delve further into the game, and figure out the rarity of the various types (how often you fight them, how likely you are to have one of that type) the values should be about balanced. So while Pokemon may be much more complex, it can be boiled down to a Rock-Paper-Scissors system.

Other designers also take advantage of the cyclic nature of RPS in a different way to balance their game. In Starcraft 2, their are three races that can be played, Terran, Protoss and Zerg. Each of these races is balanced with each other, but particular units that can be created beat other particular units. In a battle between Terran and Protoss, the Terran player will likely create some Marines, to counter this, the Protoss makes some Colossi, countered by Vikings, which in turn are countered by Stalkers, which get taken down by Marauders, which lose to Void Rays (look up units on the wiki). This evolution of fighting style mid game requires the player to shift their tactics and strategies.

Now, while Starcraft 2 uses the Rock-Paper-Scissors system for smaller scale balancing, on the whole, they use a system that is equal-but-different. Each of the three races has a base military unit. For 100 Minerals (one of the resources) and 2 Supply (which counts towards a unit-cap), the Protoss can make a Zealot, the Terran can train two Marines and the Zerg can hatch four Zerglings. Any one of these groups has approximately the same total health and damage capacity. The differences in their stats are offset by other abilities. The Zerglings numbers allow them to surround an enemy unit to prevent it from moving, the Marines have a ranged attack which lets attack air units and the Zealots are simply statistically superior. This for of equal-but-different reaches through the ranks, but comparisons aren't often as clear cut as it is for the basic unit.

A third commonly used system is just giving players the same resources from the starting gate. In Age of Empires, each of the many civilizations has access to the same units. They all get the militia, the longswordsman, the archer, the crossbowman, the knight etc. There are minor differences, and each gets a unique unit, but a lot of the balancing was achieved by giving each civilization the same opportunities. Similarly, in Risk, a unit is a unit, it doesn't matter what color it is, or where it came from, it has the same value as any other unit.

So, as it stands I have covered what I feel are the three biggest types of balance seen in multi-player games.

1. Rock-Paper-Scissors or A > B > C > A
2. Equal-but-Different or A = B = C
3. Cake-for-Everyone or A = A = A

Each of these systems has its value. Cake-for-Everyone is the easiest to design and it leads to simpler gameplay. If you want the game to be easier to follow, and easier to master, giving cake to everyone is a good idea. The downside is that because of the relative simplicity, the game can get stale as players figure out proper strategies. Rock-Paper-Scissors is the next easiest to design, as you can create obvious strengths and weaknesses. Once the cyclical nature is explained RPS isn't too much harder to follow than Cake, but allows for a game that evolves as particular strategies becoming popular. Equal-but-Different is the hardest to design, as each individual piece has to have several values that you can tweak up and down to bring to the intended level of power. However, once achieved, it is the most rewarding system and allows for the greatest depth of play.