The Issue of Balance
We have all heard game designers talking about balancing their games. This is a very important issue and one that is by no means easy or quick. So, what is a balanced game? And also, how do we know if we have balance issues in a game? Or better still, how do we balance a game that has balance issues? But what about, is balance necessary in order to have a fun and successful game? Let’s take it one question at a time:
What is a balanced game?
The one form of a balanced game is one in which each player has an equal chance of winning. Another is a game in which each player is given the same tools to win. There is a small distinction between these two ideas of what makes a balanced game. From a strictly mathematical sense, a game is perfectly balanced if it is a zero-sum game.
- Zero-Sum Game: “A zero-sum game is a mathematical representation of a situation in which each participant's gain or loss of utility is exactly balanced by the losses or gains of the utility of the other participants.”
However, in some sense, all multiplayer games that are played for money are externally zero-sum games, regardless of how “fair” they are to one player or the other. Let’s take a closer look at how people view game balance. We will add the following four subtypes of balance:
- Perfect Balance: The zero-sum equation both external and internal to the game
External Balance: The idea that the outcome of the game is a zero-sum
Internal Balance: The idea that the mechanics of gameplay are a zero-sum
Practical Balance: The middle ground most people would call fair and/or fun to play
However, there is still more to it. There are games that are not perfectly balanced and are still fun and successful as games. There are also, things outside of the game and its rules that can affect a player's chance of winning, like player skill. The game might be “balanced” but the players might not be of equal skill. Some games even account for this and try to build in ways for lower skilled players to have a better chance of winning (adding imbalance or luck/chance based rules). Yet another factor to game balance is turn order. The “first player advantage” is cited among the top balance issues within game design.
Other games are made to be “unfair” to the players in the form of challenging difficulty, but can still be internally balanced in the way the systems of the game run. As you can see there is more to game balance than just giving the players the same tools or making sure the outcome of the game is a “fair” win for one player and a “fair” loss for the other.
For this article, we will look at two examples of truly “balanced” games. They are on opposite sides of the game spectrum.
- Example #1: Rock - Paper - Scissors
This is a prime example of a “zero-sum” game. Both players have the same tools (hands) and the same power to choose one of the three options (tactics). Each game is won or lost based on a single factor that is non-variable. Each of the three options has the same statistical chance of winning if chosen by the player. The game is, therefore “balanced” in the pure mathematical sense. While the game “as it is” is an interesting way to settle disagreements in a “fair” way, many would not call it “fun”.
This simple, clean, and perfectly balanced game is the basis for many sub-mechanics in other games and has become one of the most widely used methods or guides for adding balance to how the internal structure of a more complex game operates.
- Example #2: Yahtzee (competitive version not the solitaire version)
This is a prime example of a luck based game. Both players are given the same tools (dice) and the same power to choose which dice to reroll and what to score each round (tactics). In this sense, the game is balanced in what players can do in the form of actions and choices. The game is a random “press your luck” style game that comes down to who rolls "better" and also who is "better" at managing what is given to them moment by moment by the game rolls. While on some level both players have the same statistical chances of rolling each of the combinations needed, there are very few people who would consider or think of this game as a zero-sum game in the traditional sense.
The game of Yahtzee is from the external sense a zero-sum game as one player wins “nothing” and one player loses “nothing” by playing the game. From an internal sense, the game is zero-sum as no player has any gain or loss that the other player can’t also experience. Both players are subject to the same force of chance in the same way. There is no player interaction and therefore no way for one player to influence the outcome of the other player’s game. I am sure that you would consider the game to be fair and balanced. I am also sure that you don’t think player skill has very much to do with winning or losing the game. As to whether or not it is a zero-sum game I am sure not everyone would agree.
How do we know if we have balance issues in a game?
For simple games, the answer is, your “players” will tell you during playtesting or you will see it the first or second time you test how it works. If you have a complex game you can assume a first draft prototype it is not balanced from the start. In this case, by testing the way the game mechanics interact you will notice one (or more) of the following things:
- You have a mechanic or component that is overused or underused by players
You will see one path to victory as easier than all others
One mechanic in the game completely nullifies another (not by design) making it unusable
The first player always (or almost always) wins if both players are of similar skill levels.
One powerful strategy will emerge and dominate gameplay making all other strategies obsolete.
In all of this, you are looking for the sources of the power and control players have to change how the game is played and to affect the outcome. The weak spots for balance are “things” that do not cost enough for the benefit the player gets from that “thing”. Quantifying and tracking this cost to benefit ratio for actions and mechanics is part of the game’s power curve.
One misconception about balance comes from there being different levels of power and control within a game. As long as the power levels all fall along the game’s power curve this is not necessary an issue with balance. The fact that there can be units or resources with different values do not make the game unbalanced. Even giving players unique powers might not make the game unbalanced as long as they all fall in the power curve of the game at the same level of power (cost and/or benefit level).
Another red flag might be if the cost of what one player can do costs another player more to do that same level of thing. Keep in mind that, even if this is the case, the game might be designed to shift the cost from one player to the next to compensate for another factor like “first player advantage”.
The main thing with spotting issues with balance is to "stay alert" and watch for anything that is not “acting” or “behaving” in the game's mechanics the way you designed them to act or behave. If you keep the cost to benefit ratio for each player as close to equal as you can, there should be only minor issues to work out.
How do we balance a game that has balance issues?
For this, the best advice anyone can give is what the “little engine that could” said. “If at first, you don’t succeed, try try again.” and I will add to that, “to get something you have never had you must do something you have never done.”
For game designers this means, while testing a mechanic or system, that you have found to be out of balance with the rest of the game, you will need to "change" something. This can mean trial and error without end.
- Helpful Tip: "Only make one change at a time, the test the game again to see if that change made a difference."
If you find an issue and then change three or more things before testing again, you will not know which change made the difference and also might create new issues and not know the cause.
This is the main reason for getting a game to the playtesting stage as fast as you can. It will save you countless hours of “designing” balance in your head to try and find what “might” work. Through playtesting you will “know” or “discover” what works (and what does not work) faster than just thinking or planning or brainstorming.
How do you balance the scales when you have many items of differing weights? You mix and match items on both sides of the scale until you achieve balance. In design, the two sides of the scale are the parts of the game that are not “working” and the parts that are “working”. One way to think outside the box is:
“If it is not broke don’t fix it” might not always be the best advice in game design, because you might need to change part of the game that is working in order to incorporate something that is not working back in things that are (say that five times fast). This might mean the “change” you need to make could be to part of the game that is not working or the “change” might need to be to another part (that is working) so that the two will fit together. The part that is not working might be a great mechanic alone and the other mechanics of the game just need to be adapted to fit it in. Both are options that should be considered.
Is balance necessary in order to have a fun and successful game?
The answer is “no” if we are talking about perfect balance. A game does not need to be perfectly balanced to be fun and there are many games that are not perfectly balanced that have sold quite well and been very popular. Does this mean other forms of balance is not important? No, they are all important. The forms that are found in most if not all fun and successful games are practical balance or at least internal balance. On the other hand, there are a few games out there that are fun and successful games without much practical balance or internal balance. Finding the right balance is important, just not as important as some designers might think.
If your game plays the way you designed it to function, and players have fun playing it, you already have something good. If there are some balance issues, don’t lose sleep over it. Remember, the goal is to make a playable game that is fun for the intended audience.
This is intended only as "Food for Thought". Please let me know what you think, I am by no means the authority on this subject so any input from other designers is greatly appreciated.
"Always remember to think outside the box so your games will fit inside!"
@BHFuturist
Comments
I can't get past...
I can't get past the phrase "However, in some sense, all games that are played for money are almost automatically zero-sum games regardless of how “fair” they are to one player or the other."
This is almost certainly false. Casino games are completely non-zero-sum as the balance favors the house more than anything. Most gambling games ALWAYS favors the house, unless it's "legit", and there's not many of those out there, imo.
The lure of "winning" is what makes the games "fun" in the example above. It destroys "Perfect", "External", "Internal", and "Practical". In this sense it's psychological, but it doesn't declassify it as a "game".
You could say SOME games played for money are zero-sum, but certainly not ALL games played for money.
Regardless of this discrepancy, there's a lot of truth to this post. When I teach "balance" it falls into 2 simple categories: Mathematical and Feel. Mathematical is what you'd refer to as zero-sum. "Feel" breaks mathematical and gives a sense of balance, even when the numbers are not totally balanced.
It's hard to classify "balance" because there's always exceptions. That's the beauty of game design; it continues to push and break conventional thought to simulate or stimulate situations that challenge our minds.
Still, not a bad article.
Good point
I guess that part was not as clear as I would have liked...
I meant: "Multiplayer games that are played for money are externally zero-sum games regardless of how “fair” they are to one player or the other."
Multiplayer gambling is an example of zero-sum because the sum of the amounts won by some players equals the combined losses of the others. This is because the "external result" of all multiplayer gambling is zero-sum. This would be anything from "Matching Pennies" to "Poker".
I was trying to exclude solo games for the purpose of this article but did not make that clear. Playing against a machine designed to always keep more money than it gives out is not really a "board game" that needs to be balanced.
Thanks for pointing out this flaw! Also, thanks for taking the time to read and comment. Your way of explaining it as two type is really good and I will try to incorporate that into the way I think about balance.
-Eamon
Practical Balance (war games as example's)
You didn't explain this, how it exactly works. But here is my version.
Practical Balance is imbalanced, but practises making sure that the outcomes remain valid. Victories can be 100%, and go down to almost 0%. But winning with 0,1% is still winning.
When adding new things to the game. Every new choice has to be observed. And the designer has to make sure that another choice doesn't become obsolete.
Practical Balance can be observed at 2 scale's. 1-Grand scale, 2-Unit scale.
1. The strategy that you use. Strategy A, beats B, beats C and beats A again. Players can choose any strategy. What matters is that you achieve a victory in the game. If one victory leads into having only 1 soldier standing, this looks imbalanced. But it is still a valid strategy if it leads to a victory. Which might not be achieved if another strategy was chosen.
2a. The tools that you use. Unit A, beats B, beats C and beats A again. But this becomes a bit harder, since you can achieve overkill when using enough B to counter A.
2b. Having 2 choices of different strenghts of one particular unit. Always leads to players picking the best unit. This can only be countered with achieving skill differences between the units. Thus a skilled player will pick the best unit if it is only the best unit by using skill (micro managment).
2c. When using 3 tools, you have a circular RPS. But once you have 4 or more. You might risk using mirror units. A familiar example might be having infantry and tanks. There are also anti infantry and anti tank weapons. So we can design 4 types of units. A:i-ai, B:i-at, C:t-ai and D:t-at. A and D are mirror units and, B and C are mirror units.
With most games, infantry are cheaper and weaker. This is ok, but you get this durability effect on the tanks, which imbalances the game. B and C might still remain balanced if designers don't change the statistics for more flavour. But D certainly beats A on equal grounds. This results in A not being used any more, not even when overkill could be used. And you get yourself a 3 way RPS, with A being obsolete.
Practical balance, makes sure that A gets an unique role in another area. And will remain usefull. Since it is used again.
First, let me say
First, let me say that I am not trying to be argumentative, my only goal here is to build a clear understanding of these concepts for myself and anyone else who might read this.
Practical Balance is imbalanced, but practices making sure that the outcomes remain valid.
While it is true I did not dig into how Practical Balance works, for the time bing I will stick with my definition. This is why.
Your definition that Practical Balance "is" or equals a state of "imbalance", is not at all how I see it, and whether the "outcome" of player's actions are "valid" is not what I am concerned with as a designer. let me explain...
leaving invalid outcomes to player action in the game makes failure possible. This is not to be confused with having the ability to lose or be beaten by your opponent. If two players in chess use their own strategies and in using them they make zero mistakes each... one player can still win and one can still loose. Only when the strategies themselves are balanced should the game a draw. By the same token, two balanced strategies can be played and one player can make mistakes. Neither of these scenarios invalidates the outcome of the game (win/lose/draw).
From a game designer's viewpoint, what needs to be balanced are the tactical actions players can make and the tools the players have to take actions with. The outcome is up to the players based on the plans they make using those tools. In other words, no one power of force in the game can outmatch another power of force that has the same level of power in the game's power curve.
Black pawns and white pawns in chess are a zero-sum, but a black queen and a white pawn are not!
We as designers can't (or should not try to) balance strategies only the tactical elements of the game. If we can maintain a practical level of balance within the power curve of the game's tactics and build many meaningful options (tools and choices) for the players to make. This will provide the framework for building strategies both valid and invalid for the players to test.
There are also many reasons for trying to design a state of imbalance in a game. I just don't see that as the goal of practical balance. Instead, I see practical balance as "accepting" some "margin for error" to the mathematical zero-sum that the game's tactics have as long as they are playable and fun.
Having a zero-sum to the game's tactics is very easy... making them "internally balanced", playable, and fun is not.
This is on a design concept level and not to focused on a single type or example.
My comments, as always, are only ever intended as "food for thought" or to present my non-authoritative viewpoints. I really am a calm, rational and reasonable guy, once you get to know me. I do not intend offense of any kind should I seem to disagree with you or anyone else on any subject.
-Eamon
eamon wrote:First, let me say
If two players in chess use their own strategies and in using them they make zero mistakes each... one player can still win and one can still loose.
I disagree with this statement for chess (chess is a puzzle to me). I do get your point. Perhaps Risk is a better example in this to make your point completely valid.
We as designers can't (or should not try to) balance strategies only the tactical elements of the game. If we can maintain a practical level of balance within the power curve of the game's tactics and build many meaningful options (tools and choices) for the players to make. This will provide the framework for building strategies both valid and invalid for the players to test.
Instead, I see practical balance as "accepting" some "margin for error" to the mathematical zero-sum that the game's tactics have as long as they are playable and fun.
Question?
Did you mean disagree? Because the rest of what you said sounds like that is what you meant. If so I don't understand why you disagree but am interested to know more. I sure don't think I am right about everything... so understanding the way you see this will help me to frame it better in my mind.
About the Chess vs. Risk comment, if my point is partly valid for one... it is partly valid for the other. I am quite sure I am not 100% right about everything. These are very deep concepts and my crude explanations are far from the mark of the best explanation for them.
I enjoy hearing other views on things! Otherwise, I never learn.
-Eamon
eamon wrote:Did you mean
Could you please quote the part where it looks like that I am contradicting myself?
Sorry, I guess I was not clear (too tired)
To me, it sounds like you are saying that we as designers should be focused on the strategies then make changes to adjust things in the design depending on the outcome of the strategies.
You started out with "I totally agree" then you say something is the "other way" and go on to make an opposite point, to what I was saying.
This makes it sound like you meant to say "I totally disagree with this".
I could be wrong, I was quite tired the first time I read that last night.
It might be better to focus on the overall strategies and try and design them into the game... I just think that doing so might lead to some less than ideal results in the overall design of a game. This does not mean as designers we can't (or should not) think about what strategies are developing from the game we are designing, just that I don't feel, that should be the focus of the design process. I feel the focus should be on keeping the game balanced at the tactical level based on the power curves we want in the game. This comes from the way I use and think about tactical & strategic concepts.
Also, check out this video I just watch:
https://www.youtube.com/watch?v=e31OSVZF77w
-Eamon
I see. I try to explain,
I see. I try to explain, since it is a hard one to understand what I mean.
One cannot change strategies, as you said. They are build by the players. And as 1 designer, you don't know what a community can come up with.
So it has no use focusing on them.
However, if a strategy is to strong or to weak, it is most likely caused by 1 or more tactics.
I say, we pick one of the tactics in that strategy. And focus on making that tactic better, regarding the possible desired outcomes.
An well known example (that is annoying me to no end as of today, since no one actually puts effort in this):
A well known strategy in Red Alert is building mass tanks. Why? Here are the tactics:
1-Resource management is maxed out in an easy way. This opens up to use the "imbalance" of durability. Tanks health > anything else in health.
2-They have the best weapons against armor. It just so happens, most tanks and structures are!
3-They can easily crush infantry.
4-They can fire and move at the same time, if you are a skilled player and know to find the Q key on your keyboard.
5-Clicking the same stupid icon for building tanks over and over, is the simplest thing to do. This is the opposite to having to try to click multiple icon's for a nicely mixed army.
6-Speed! Obviously of better use on big maps.
The second, third, fourth and sixth tactic can be used at the same time!!!
There is plenty of money, so it all depends on the fastest production. Thus 1 adds up to 5!!!
Now, you want this strategy to be removed? Let's focus on 1 tactic at a time. Because changing all tactics will immediately revert the strategy in "all the time", into "never ever again".
Personally, I suggest removing tactic number 3. Since it is very OP. And works in tandem with 2, 4 and 5. Once it is removed, the tanks need to shoot at infantry. Tactic 2 becomes less used this way. Tactic 4 can even be stopped now: Since your tanks might get stuck behind enemy infantry, thus easy pickings. And tactic 5 will not supply benefits to 3, since 3 doesn't exist any more.
Some might suggest that we calculate some new prices to the tanks instead. But tactic 1 is in the way of this. And unfortunately, tactic 1 is a choice by the players. We can't remove it!
Changing 6 instead of 3. Might work too. But I bet that the game evolves in an entirely different way. Both have never been tried out in Red Alert.
***
And here is where I agree with you on not focussing on the strategy itself. But only individual tactics that make a strategy. After changing, only the players will show the designer if the change worked. New strategies will emerge. You can't counter this by focussing on strategies only. It is actually relatively unpredictable what strategies might emerge.
One final note:
Once a tactic has changed. Not only the targeted strategy will change. Other strategies might change as well. Since strategies have to be build up by players. We must wait and see.
Yes... but
I think you explained what you mean quite well, and I don't disagree in any way, "they" are both things that need to be considered (tactics and potential strategies). I guess the way I should "frame" what I was going for this way:
As a designer you should focus on the balance of the tactical mechanics and systems in the game, because in the end, that is the "part" you the designer have control over. You can never totally control what a player might try to do with the mechanics and systems you have made so in the end, spending your time focusing on that "what if" might distract you from the real issue of balancing the tactical mechanics. I am saying it is our job as designers to make sure there is no one tactic (other than intended end game units) out of balanced cost to power level or no equal cost counter.
Your example is what I am talking about. You are trying as a designer to fix the problem of "A player created strategy". Any fix action when looking at the issue from this angle is "player action" focused. (saying that what the player is doing needs to be stopped from being a valid strategy)
What I am asking everyone to consider is; looking at the issue from the angle of the "the player strategy" has pointed out a hole or imbalance in the games overall power curve. The tank is there for the one part.
Is it in the right place on the power to cost curve? If so, can something else be moved up or down the power curve to compensate for where the tank sits?
The issue (I think) is, that there is no equal cost-power-benefit tactic to "counter" the mass tank force.
The solution you give is to "Nerf" tank power against infantry. (if I understand you right)
While this "might" bring some practical balance to the game, it "overlooks" the fact that "tanks do roll over infantry" with very little effort in real life.
Consider that in World War I/II this was a "real world problem" that actually happened! (this balancing act has seen action in the real world)
One solution (in real life) was to arm some infantry units with anti-tank rifle grenades. In a real-time strategy game this could be a unit that is attached to a normal infantry unit (or a Medium/High cost upgrade that converts all infantry into an anti-tank force). Then we add a new "armor" upgrade to tanks that brings them into balance again. Further up the power curve, we add air units like the A-10 Warthog (an infantry support/Anti-tank plane with no chance when fighting against other planes).
The issue here is with the cost-power-benefit vs. cost-power-benefit of each unit and the level each one falls along the power curve. These are all "levels" along a power cure that need to be designed and balanced independent of "one strategy thinking".
When you have only one strategy in mind you might over focus on the symptoms of the problem and miss the problem itself (sometimes).
This balance issue was that "one tactic" did not have a "counter force" or "proportional cost" along the game's power curve. The one tactic might just be at the wrong cost scale for the power it gives. Making tanks cost more, or take longer to build might also be the "part" that needed work.
Anything we might come up with, is going to be based on, how we are looking at the problem. I am suggesting that there are other ways to look at the problem.
Just more "food for thought"
-Eamon
Well, this is a 50% off topic fact. But one that I like to share
True that they roll over infantry in real life. And there is the mistake.
The game is considered by me as a slowed down simulation of real life. The tank fire on infantry is weak. This because one infantry unit in the game actually resembles, I dunno, a squad of 10 at least. Normally, it is one shot, one kill. This often counts for tanks vs tanks as well. My infantry units in my board game actually represent 3, since they have 3 health.
Most agree that the balance is better when only taking into account the firepower. Not the squishing. In the game, a tank can squish 5 to 10 infantry within a second. That would mean in real life at least 50 to 100?? The tank is not that big. And the people aren't lined up in the field when they see a tank.
I once estimated that the costs of these tanks in the game. Also including power density, since players go for amass. Would be many times higher. With the light tank going from 700 to 2000 (2 factors involved, not just 1).
So, mathematically, the squishing is wrong. The tank is at the wrong place in the power curve. Anything moved towards where the tank is on the curve. Will also be to strong for the rest of the game. Instead of changing everything. Just that one thing that is incorrect. The tank's squishing has to be nerfed.