The basics of game balance

Moderator: JaggedAppliance

The basics of game balance

Postby ZenTractor » 12 Jul 2015, 05:37

This is going to be the first in a series of posts that I am going to make about game balance. This post in particular will work as a framework for future discussions and particulars.


I see a lot of discussion on these forums about balancing various things, but a lot of the time it seems marred by the fact that there is an apparent lack of understanding of game balance on a fundamental level. I hope, in this post, to give the reader an understanding of what balance means, and what appropriate balance suggestions look like.

During this post I am arguing from the perspective of a game designer, not from the perspective of a player of the game.

Types of Balance

There are two types of overarching balance in games that I will talk about: Theoretical Balance, and Practical Balance. These things probably have other names, but they're what I'm going to use.

A Theoretically balanced game is a game where every player has the same chance of winning. Examples include: Tennis, A game of paintball in an empty warehouse, Golf, or a game of chess where both people go first. The idea is that no choices about the game before the game has begun have any effect on its outcome. A theoretically balanced game will have a lot of choices that the players can make, with all possible choices being viable if the player wants to win.

A Practically balanced game is a game in which the balance is conditional on things within the game. I think this is easiest to illustrate with an example: Imagine a building that one particular faction could build in SupCom that, when completed, instantly detonates a nuke above all enemy commanders. However, this building costs 10,000,000,000 mass to complete and a comparable amount of build time. Such a building is 'balanced', because it will never actually get built in any competitive game; you would be better off spending your mass and time on other units. If you have ever heard the argument "well you let him build <X>, so he deserved to win", then that was an argument from practical balance.

I argue that Theoretical balance is ideal and Practical balance is generally unwanted. One is never really going to be able to obtain a perfect Theoretical balance in a game with as many moving parts as an RTS, but I still hold that the direction which balance reworks should take is towards Theoretical balance: One wants their players to have as many viable options at as many points in the game as possible.

  • Percivals are an example of Practical balance. Percivals are kinda OP. Mass for mass they beat pretty much anything else on the field near their tech level. However, they are 'balanced' around the fact that they are in the UEF faction, and the UEF don't have a direct fire experimental. The percival makes up for this deficit in another area in some way.
  • T2 fighterbombers. Fighter-bombers are currently under a fair amount of discussion in another thread. However they are proving to be a problem because of a Theoretical balance issue: If a fighterbomber is a valid unit to build as a fighter, then it immediately becomes amazing because the moment you win air control you can leverage that control to start destroying important things. ie: bombing power/t2mex/commanders.
  • T1 tanks are (mostly) balanced Theoretically. UEF get extra hitpoints, cybran get faster units with more damage, Aeon get good damage and range but have massively less hitpoints and movespeed. All the factions have their own strengths and weaknesses, and all are viable choices. (disregarding concerns about Aurora power. We shall get to that later.)
  • The Cybran ACU Vs. the Seraphim ACU as they spawn is an example of Theoretical balance. The Seraphim ACU has more HP, but the Cybran ACU has more regen. The Seraphim has an advantage in direct confrontations while the Cybran can win over an extended engagement.

Balance Points, or, Why DPS Doesn't Matter (Until it Does)

Sometimes DPS matters and sometimes it doesn't. When a few tanks are shooting at a building, a small change in DPS changes the time it takes for them to take the building offline. However, when an ACU is shooting at Thaams it doesn't actually matter if it does 100 damage per shot or 105: It will still die in three hits. What is important is the Time to Kill, and the closely related Shots to Kill. If you have played any FPS game you will know that the value of a sniper rifle is not the amount of damage per second it can deal, but rather the fact that it can instantly remove someone from the fight.

The way these different emergent statistics interact can lead to some very strange behaviour. For example: did you know that, mass for mass, Strikers beat Percivals? Put simply, while the Percival deals an abominable amount of damage with each shot it can only kill one Striker every 4 seconds, while the Strikers will happily take the rather paltry hit to their total damage and the rest of them keep shooting.

In SupCom the T1 stage is very interesting; the units hit points are balanced around the time it takes for a commander to kill them. More precisely: they are balanced around the number of shots it takes a commander to kill them. A T1 tank dies in three shots, an artillery dies in two, and scouts and light assault bots die in one. If we were to increase the hit points of a Thaam or a Mantis by 15 or so it won't change how quickly they die to commanders. However increasing a Striker's hitpoints by a mere 1 will drastically alter their balance, as they will suddenly take four shots from a commander before they cease functioning. (Again, we will get to the dreaded Aurora later)

  • T1 Land is the epitome of this.
  • Again T2 Fighter-bombers rear their head. If they want to have enough damage to destroy land units in a single pass they start to dps that can put away commanders disturbingly quickly, and their balance must take that into account.
  • Mercies are very 'have or have not'. They can be destroyed by a single hit from more or less any anti-air gun in the game, and that fact is what allows them to have such a great amount of power in their suicide weapon.

Types of Power

Power comes in many different flavours in an RTS. A unit might be strong because it has a very large amount of damage output. Or a unit might be strong because it can hit things without being hit back. Or a unit might offer some other ephemeral power not directly, but simply by being there, which we will call utility.

One of the biggest sources of power in video games is speed. Suppose you were in a first person shooter, and had a button which halved your hit points and doubled your movespeed. Any experienced player knows that this is insanely powerful. If you asked such a player how many times you should press the button, they will reply that you press it as often as you can, until the next press renders you unable to control your character any more. This is because if you don't ever get hit, your hitpoints don't matter.

Supreme commander has a number of different sources of power. A few important ones are: Dps, Range, Speed and Density.
DPS is kind of obvious. If a unit does more damge it is more valuable, and hence costs more.
Range is valuable because if you can deal your damage without taking return fire you have gained an advantage over your opponent 'for free'.
Speed is universally translatable. If you can get your units in behind your opponents then you can start shooting things that you want to shoot (economy), instead of the units your opponent wants you to shoot (tanks).
Density is something that is relatively unique to SupCom. Because supreme commander has large maps and (relatively) slow movement, being able to distribute your power into smaller, tighter clumps is valuable in itself. A bunch of units in a ball will beat a few units trickling in.

Different types of power are not necessarily interchangeable. You can't just double something's range and halve its movespeed and expect it to function as a balanced unit. To illustrate this consider the following: You have the ability to pay the cost (mass, power and buildtime) of a Mantis to add the Mantis' turret to any other land unit, along with its hitpoints and repair ability and whathaveyou.
  • If you add the Mantis turret to a Rhino, you gain very little. The change in damage is rather meagre, and it is at a much shorter range than the Rhino's main gun, so it will get used less. It's still a good thing to have and probably worth the cost, but it isn't amazing.
  • However, if you add the Mantis turret to a Mantis, then you suddenly have an extremely powerful unit. You've doubled the damage this new mantis deals, and doubled its health. Sure, it costs twice as much, but it is strictly better than having two separate mantii.

The point of this section is to say that when rebalancing units you have to be very very careful when suggesting trading one sort of power for another.

Soft and Hard Counters

This is a pair of terms taken mostly from other strategy games. A Hard Counter is something which invalidates or destroys another avenue of play. A Soft Counter is something which renders an avenue of play less valuable or powerful, without destroying its utility entirely, or while being useful for other reasons.

For example, an anti-nuke is a Hard Counter to a nuke launcher. A Soft Counter to a nuke launcher is to spread your base out so that no one location is worth nuking. Sure, the player with the launcher could take out something with his nuke, but it's probably an inefficient use of resources to do so. But the nuke still has value because it can be used to destroy an army, or snipe an ACU or whatever. It still lands, but it's not always 100% a win.

I think that Hard Counters are generally something to be avoided. Nobody likes having their cunning plan foiled by a single unit. It might be a valid form of balance, but I think it's an un-fun one.

Supreme commander is remarkable in that it has very few hard counters when compared to other games. All land units can deal their full damage to any other unit, without having to worry about armor class, or infantry cover bonuses or anything silly like that. However, there are a few standout exceptions:
  • Air units can only be hit by anti-air units. This extends to interceptors and ASF: planes whose sole purpose is to kill other planes. They cannot win you the game directly (as the ACU is a land unit).
  • Nukes and anti-nukes, and to a lesser extent TML and TMD
  • Navy and especially Torpedoes.
Yet supreme commander has a wonderful way of bridging these hard counters with soft counters: TML do not home into their targets, and hence can be (soft) countered by any unit with a move speed. Hover units are land units that can fight against navy.

I believe that SupCom's emphasis on soft counters is one of it's core values. There are very few problems which cannot be solved by simply building more tanks.


TL;DR: Let's talk about balance.

I hope that this rather long talk will provide a framework for balance discussions in general, and will give a good foundation for what I hope will be a fun series of balance discussions to write in the coming weeks. I have a few ideas of what to tackle next: Either 1) Factional Diversity, 2) Air balance and proposed rework, or 3) fundamental unit value rebalance (or: why are you guys hating on hover so much?).

I will say now that I expect this series of discussions will result in something that is more of a total conversion rebalance than a simple balancing of what we have. In essence the end result will be a game that 'feels' like SupCom, but is not the same as SupCom.

However many individual things will be useful for balancing FAF, and this thread in particular should give a 'language' of sorts which any and all future balance discussion can utilise.

Tell me what you think! I am posting this because I like to think about and discuss balance. If you think I am wrong on any level then please tell me. I'd love to hear other points of view, and getting stuck in one's own 'echo chamber' is something I would like to avoid.
Posts: 67
Joined: 10 Apr 2014, 04:09
Has liked: 10 times
Been liked: 2 times
FAF User Name: ZenTractor

Re: The basics of game balance

Postby KD7BCH » 14 Jul 2015, 22:59

Our FB component is about making FBs effective against T3 strats, not against Interceptors, or ASFs or even Swiftwinds
The Gun Down
Posts: 424
Joined: 25 Feb 2015, 18:06
Has liked: 29 times
Been liked: 12 times
FAF User Name: KD7BCH

Return to Balance Discussion

Who is online

Users browsing this forum: No registered users and 1 guest