Definition: What is Depth? A multiplayer game is deep if it is still strategically interesting to play after
expert players have studied and practiced it for years, decades, or centuries.
We often talk about skillcaps. The more skillcaps a game has the better it is. Ideally, you want a game to have an infinite number of skillcaps. A skillcap is a level of expertise which allows the players in it to dramatically beat the players which belong to the lower level of expertise. The idea behind skillcaps is that, if you, as a player, are in a lower skillcap, it means you're not entirely exploiting all the possibilities of the game. Many skillcaps means that increasing things like reflex is not enough. The player needs to increase its understanding of the game. Incidentally, that game is very deep.
Definition: What Is Balance? A multiplayer game is balanced if a reasonably large number of options
available to the player are viable--especially, but not limited to, during highlevel
play by expert players
Well, not really in my opinion. A game is balanced if you cannot find a single way-of-playing to beat an equally-expert opponent more than 50% of the time when the opponent is not allowed to use that same way-of-playing. (Note we are talking here about robots which automatically play each game with the exact same level of expertise.)
EDIT: oh the author calls that fairness. Fair enough then.
>A skillcap is a level of expertise which allows the players in it to dramatically beat the players which belong to the lower level of expertise.
Where are you getting that definition? I've always seen "skill cap" defined as the highest possible level of skill, beyond which no increased skill can increase performance. "Skill cap" is also not a measure of players, but of game mechanics and games themselves. We might say that checkers has a lower skill cap than chess, or that the bishop has a lower skill cap than the queen, but we wouldn't say that Magnus Carlsen is at a higher skill cap than I am. He's just more skilled.
We want unreachable skill caps, so that there's always room for improvement. I'm not sure any game has an infinitely high skill cap.
Unlimited skill caps become boring. Ideally you want players to use and face a range of strategies over time, but in an unlimited skill cap game with no single dominat strategy or hard counters players get stuck perfecting the same approach over time.
I think it's by analogy to the chess piece version you mentioned. A strategy has a skill cap. Therefore, a player who only knows some subset of strategies, where the max skill cap within that set is X, will always eventually be beaten by some other player who knows a strategy outside that set, with a higher skill cap. That player can be said to "have" the skill cap of the highest-skill-cap strategy they're currently aware of.
To put it another way: a "perfect" player with mastery of every strategy they are aware of, but with imperfect knowledge of the space of possible strategies, will have a skill level at the game equal to the highest-skill-cap strategy they're aware of.
Or, if you only practice the strategies you already have, you'll only ever get as good at the game as your best strategy is powerful. It can be much higher-ROI to explore the game's strategy-space to find higher-skill-cap strategies.
Imagine a game where a player who has 80% mastery of a large number of strategies, beats a player with 90% mastery of fewer strategies. The 80% player probably won because one of the strategies they were competent at had a high-enough skill-cap to dominate the other player who didn't know that strategy.
One method that has proven to work good is to keep statistics and then nerf overused tactics and bump the ones that are rarely used.
It will make some interesting game play!
Also try to have as much diversity as possible. This is one of my favorite quotes: "This game is so imbalanced so that it get balanced".
Completely hijacking here, but if you haven't updated your profile to include an email do so now. I have gotten several interesting mails and no spam that I know of.
Doesn't sound thaaat crazy. Go has the komi system, which gets updated occasionally - that is, white starts with points to compensate for blacks first-move advantage (currently 6.5 points in Japanese Go, but its been different over the years).
Of course that means you need massive testing periods and/or have to patch things all the time (which takes effort, and can annoy players a lot), but for complex games that aren't extremely symmetrical it likely is the only way.
Are you thinking of any particular examples? First to my mind was Starcraft 2, which gets regular balance patches. And community bitching because the patches take to long/don't conform to everybody's pet theories/...
Constant updates and patches are annoying, indeed! You have to be very delicate about what to change and what to add. If you manage to do it right though, it will make the game feel fresh and exciting for many years.
We all hate changes. Look for things that annoy ppl and try to fix those problems. Try to add new content, instead of changing old mechanics.
This should be done slowly though, to give the mediate players a chance to get a hang of it. In a player vs player environment, it can take years for strategies to settle in, and for the majority of players to adapt.
> Also try to have as much diversity as possible. This is one of my favorite quotes: "This game is so imbalanced so that it get balanced".
That is embodied by the board game Cosmic Encounter.
It is a sci-fi negotiation game of space conquest. It has tons of special powers that effectivelly bend or break the rules. Some powers are so over-powered that the other players end up ganging up / never helping the OP player. (The game is best played with 5 players)
Cosmic Encounter is the ur-example of a "balance through player interactions" type of game design (Diplomacy is another example). That is, there is just enough balance that one player cannot force a win by themselves, but there are otherwise gross imbalances in player power, and it is up to the players to maintain balance through temporary alliances and inevitable backstabbing. From my many plays of Cosmic Encounter, it does seem that the player with the strongest position does not win very much more often than others.
There are downsides to the design: one, it's hostile to new players who don't have enough knowledge of the game (this is true of any skill based multiplayer game, but in games like CE the effect of an unskilled player on the game can be profoundly wonky). The other issue I have is that players with a weak initial position have to win through negotiations, and so have fewer paths to victory than the players in powerful initial positions. I rarely ever see the "weak" player win in Cosmic Encounter (whereas the moderately-powerful and very-powerful players seem to have about equal chance at winning if they're playing competently).
I've casually worked on MUDs for a while and whenever I help a friend, the first thing I do is put their game in a spreadsheet. It usually reveals several things they didn't realize about their game.
They tend to tweak their games over and over based upon feedback until they eventually don't know what their own game is doing.
I've toyed with MUDs a bit as well. I'm actually thinking about doing it again because I want to play around with a component-entity-system idea I had. Anyway, I've always figured that keeping everything as a zero-sum-rule is the only way to really balance things. A spreadsheet makes this much more visible (because you can list all your assumptions (weights etc) in a sheet and calculate if everything balances or not. Then if you learn the assumptions are wrong, modify them and rebalance). So yeah, definitely good advice!
Definition: What Is Balance?
A multiplayer game is balanced if a reasonably large number of options available to the player are viable--especially, but not limited to, during high-level play by expert players.
Interesting. I think of 'balance' in multiplayer games as a different thing altogether. I am usually thinking of the match itself, rather than the design of the game. In that sense, a game is "balanced" if the overall skill level on each side is reasonably close to the overall skill level on the opposing side. Games where one team is just too much better than the other are boring. "Balanced" games -- where you really don't know who will win until the game is over -- are usually the most fun.
I have a beef with the article's definition of "balance", but you are describing balanced matchmaking, rather than game design. Important, and a hugely difficult problem, but a different thing.
I thought it was obvious from my response that I did know that I was talking about a different thing. In fact I was intentionally pointing out that I was thinking of a different thing.
My point was that when I hear the term "balance" in regard to gaming, I'm thinking about the relative equality of skill of the teams, not things that relate to the game design itself.
In that case we need some new term to differentiate the games of "Starcraft II" and "Starcraft II, but without Terran and Zerg as viable options for competitive play."
I think you're confusing "game" and "match", where "game" means the rules defining how players interact, and "match" means one instance of a game being played by a specific group of players.
"game" and "match" can sometimes be used interchangeably (e.g. "that was a good game" can mean a single match), but a discussion on "game balance" typically refers to the rules themselves being fair.
What you described, grouping players into opposing teams of roughly equal skill, is called matchmaking, and it's a different problem from game balance. In fact, in a competitive match where game balance is arguably the most important, matchmaking is not a big problem since it's solved by a bracket of some sort.
"I am usually thinking of the match itself, rather than the design of the game".
By which I mean pretty much exactly what you said. "Game" can mean either a single match, or the actual thing that you are playing.
My point was that the term "balance" isn't limited to the dynamics of the game (those that exist regardless of the players skills in using them), but that it is often used to refer to the equality of the teams -- and that when I hear the term "balance" in regard to gaming I am more likely to think of the team equality than the other possibility.
Where do you go, that you hear so many complaints about unbalanced matchmaking?
Tournaments and ladder systems almost automatically cause match "balance" in the way you're speaking of. Some games go further and introduce handicaps, so weaker players can play stronger players with the result not being predictable. In either case, nobody is really thinking about balance; it's just built into the social systems around the game.
The only time one might hear complaints about match "balance" is in non-competitive "casual" play. (This is actually a large part of what Nintendo's game design is about. The Smash Bros, Mario Kart, Mario Party, etc. series are all about letting people with unbalanced skill—family members, say—play casually against one-another while still having fun.)
I think attempting to balance games where the number of character/choices are large enough is a futile process (especially if the design process aims at the "cream of the rop" players). The truth is that at the end of the day, competitions are still going to be populated with players would use the top ~20% something (depending on the game) of the character cast, with a very small minority actually bothering with the other characters due to the playing-to-win mentality. I see this pattern being very common with a lot of games, and attempts at balancing them only shuffle the casts around i.e. a new set of characters now dominate, replacing the old ones. Part of the reason is because the characters/game mechanics are so intricately tied to one another, a small change could flip the entire "metagame" around.
This symptom is more apparent with MOBA games such as Dota 2. After each balance patch, the game becomes chaotic enough to raise interests among players, although the metagame will eventually converge to a singular point. Players will then complain and the process continues. Each patch hardly makes the game more balanced, they just make the game different. I think at stages like this, you simply have to adopt a new definition of "balance".
Also I disagree with "Design Self-balancing Forces". While the mechanics proposed are good at creating depth for the game, I think it's misguided to add them with balance in mind. An example is the game Street Fighter 3 (specifically the third edition, Third Strike): the designer added a "parry" mechanic in which allows a player (with a good guess, timing and some memorization) to counter any attack. This could be perceived as an attempt to give any character a chance, but in practice players argue that this makes the strongest characters even stronger (of course it's hard to verify this because an identically copied game with no "parry" doesn't exist) because they can counter attack for much higher damage.
Also, Sirlin (the author) spends a lot of his time with versus fighting games, and the article reflects that. I think if you're looking to balance a game of another genre, it's unlikely that his philosophies will be very useful. Just for reference, versus fighting games usually have these metrics: character pool is between 10-30 (usually don't go up much), there's a heavy emphasis on specialising on one character (could go up to 3), heavy emphasis on reaction and guessing, long intervals between balance patches (if there are going to be patches at all), at competitions there rarely are rules (i.e. usually no banning, save for a few games where balance is pretty bad). If you compare this to other genres (FPS, RTS, MOBA, ...) these metrics will be very, very different.
31 comments
[ 2.9 ms ] story [ 46.7 ms ] threadDefinition: What is Depth? A multiplayer game is deep if it is still strategically interesting to play after expert players have studied and practiced it for years, decades, or centuries.
We often talk about skillcaps. The more skillcaps a game has the better it is. Ideally, you want a game to have an infinite number of skillcaps. A skillcap is a level of expertise which allows the players in it to dramatically beat the players which belong to the lower level of expertise. The idea behind skillcaps is that, if you, as a player, are in a lower skillcap, it means you're not entirely exploiting all the possibilities of the game. Many skillcaps means that increasing things like reflex is not enough. The player needs to increase its understanding of the game. Incidentally, that game is very deep.
Also, there is a measure for player expertise called the Elo rating system: http://en.wikipedia.org/wiki/Elo_rating_system
Definition: What Is Balance? A multiplayer game is balanced if a reasonably large number of options available to the player are viable--especially, but not limited to, during highlevel play by expert players
Well, not really in my opinion. A game is balanced if you cannot find a single way-of-playing to beat an equally-expert opponent more than 50% of the time when the opponent is not allowed to use that same way-of-playing. (Note we are talking here about robots which automatically play each game with the exact same level of expertise.)
EDIT: oh the author calls that fairness. Fair enough then.
Where are you getting that definition? I've always seen "skill cap" defined as the highest possible level of skill, beyond which no increased skill can increase performance. "Skill cap" is also not a measure of players, but of game mechanics and games themselves. We might say that checkers has a lower skill cap than chess, or that the bishop has a lower skill cap than the queen, but we wouldn't say that Magnus Carlsen is at a higher skill cap than I am. He's just more skilled.
We want unreachable skill caps, so that there's always room for improvement. I'm not sure any game has an infinitely high skill cap.
To put it another way: a "perfect" player with mastery of every strategy they are aware of, but with imperfect knowledge of the space of possible strategies, will have a skill level at the game equal to the highest-skill-cap strategy they're aware of.
Or, if you only practice the strategies you already have, you'll only ever get as good at the game as your best strategy is powerful. It can be much higher-ROI to explore the game's strategy-space to find higher-skill-cap strategies.
Imagine a game where a player who has 80% mastery of a large number of strategies, beats a player with 90% mastery of fewer strategies. The 80% player probably won because one of the strategies they were competent at had a high-enough skill-cap to dominate the other player who didn't know that strategy.
For instance, winning 1 in 10 may allow more options to be considered viable than if one insist on a 1 in 5 ratio.
I need help balancing my RTS for this weeks indiecade submission if anyone is interested in helping ?
Also try to have as much diversity as possible. This is one of my favorite quotes: "This game is so imbalanced so that it get balanced".
Are you thinking of any particular examples? First to my mind was Starcraft 2, which gets regular balance patches. And community bitching because the patches take to long/don't conform to everybody's pet theories/...
We all hate changes. Look for things that annoy ppl and try to fix those problems. Try to add new content, instead of changing old mechanics. This should be done slowly though, to give the mediate players a chance to get a hang of it. In a player vs player environment, it can take years for strategies to settle in, and for the majority of players to adapt.
That is embodied by the board game Cosmic Encounter.
It is a sci-fi negotiation game of space conquest. It has tons of special powers that effectivelly bend or break the rules. Some powers are so over-powered that the other players end up ganging up / never helping the OP player. (The game is best played with 5 players)
There are downsides to the design: one, it's hostile to new players who don't have enough knowledge of the game (this is true of any skill based multiplayer game, but in games like CE the effect of an unskilled player on the game can be profoundly wonky). The other issue I have is that players with a weak initial position have to win through negotiations, and so have fewer paths to victory than the players in powerful initial positions. I rarely ever see the "weak" player win in Cosmic Encounter (whereas the moderately-powerful and very-powerful players seem to have about equal chance at winning if they're playing competently).
They tend to tweak their games over and over based upon feedback until they eventually don't know what their own game is doing.
Design in Detail: Changing the Time Between Shots for the Sniper Rifle from 0.5 to 0.7 Seconds for Halo 3
PDF: http://twvideo01.ubm-us.net/o1/vault/gdc10/slides/Griesemer_...
Talk & SlideDeck: http://www.gdcvault.com/play/1012211/Design-in-Detail-Changi...
Interesting. I think of 'balance' in multiplayer games as a different thing altogether. I am usually thinking of the match itself, rather than the design of the game. In that sense, a game is "balanced" if the overall skill level on each side is reasonably close to the overall skill level on the opposing side. Games where one team is just too much better than the other are boring. "Balanced" games -- where you really don't know who will win until the game is over -- are usually the most fun.
I thought it was obvious from my response that I did know that I was talking about a different thing. In fact I was intentionally pointing out that I was thinking of a different thing.
My point was that when I hear the term "balance" in regard to gaming, I'm thinking about the relative equality of skill of the teams, not things that relate to the game design itself.
"game" and "match" can sometimes be used interchangeably (e.g. "that was a good game" can mean a single match), but a discussion on "game balance" typically refers to the rules themselves being fair.
What you described, grouping players into opposing teams of roughly equal skill, is called matchmaking, and it's a different problem from game balance. In fact, in a competitive match where game balance is arguably the most important, matchmaking is not a big problem since it's solved by a bracket of some sort.
I'm not confusing those terms at all. I said:
"I am usually thinking of the match itself, rather than the design of the game".
By which I mean pretty much exactly what you said. "Game" can mean either a single match, or the actual thing that you are playing.
My point was that the term "balance" isn't limited to the dynamics of the game (those that exist regardless of the players skills in using them), but that it is often used to refer to the equality of the teams -- and that when I hear the term "balance" in regard to gaming I am more likely to think of the team equality than the other possibility.
Tournaments and ladder systems almost automatically cause match "balance" in the way you're speaking of. Some games go further and introduce handicaps, so weaker players can play stronger players with the result not being predictable. In either case, nobody is really thinking about balance; it's just built into the social systems around the game.
The only time one might hear complaints about match "balance" is in non-competitive "casual" play. (This is actually a large part of what Nintendo's game design is about. The Smash Bros, Mario Kart, Mario Party, etc. series are all about letting people with unbalanced skill—family members, say—play casually against one-another while still having fun.)
I think attempting to balance games where the number of character/choices are large enough is a futile process (especially if the design process aims at the "cream of the rop" players). The truth is that at the end of the day, competitions are still going to be populated with players would use the top ~20% something (depending on the game) of the character cast, with a very small minority actually bothering with the other characters due to the playing-to-win mentality. I see this pattern being very common with a lot of games, and attempts at balancing them only shuffle the casts around i.e. a new set of characters now dominate, replacing the old ones. Part of the reason is because the characters/game mechanics are so intricately tied to one another, a small change could flip the entire "metagame" around.
This symptom is more apparent with MOBA games such as Dota 2. After each balance patch, the game becomes chaotic enough to raise interests among players, although the metagame will eventually converge to a singular point. Players will then complain and the process continues. Each patch hardly makes the game more balanced, they just make the game different. I think at stages like this, you simply have to adopt a new definition of "balance".
Also I disagree with "Design Self-balancing Forces". While the mechanics proposed are good at creating depth for the game, I think it's misguided to add them with balance in mind. An example is the game Street Fighter 3 (specifically the third edition, Third Strike): the designer added a "parry" mechanic in which allows a player (with a good guess, timing and some memorization) to counter any attack. This could be perceived as an attempt to give any character a chance, but in practice players argue that this makes the strongest characters even stronger (of course it's hard to verify this because an identically copied game with no "parry" doesn't exist) because they can counter attack for much higher damage.
Also, Sirlin (the author) spends a lot of his time with versus fighting games, and the article reflects that. I think if you're looking to balance a game of another genre, it's unlikely that his philosophies will be very useful. Just for reference, versus fighting games usually have these metrics: character pool is between 10-30 (usually don't go up much), there's a heavy emphasis on specialising on one character (could go up to 3), heavy emphasis on reaction and guessing, long intervals between balance patches (if there are going to be patches at all), at competitions there rarely are rules (i.e. usually no banning, save for a few games where balance is pretty bad). If you compare this to other genres (FPS, RTS, MOBA, ...) these metrics will be very, very different.