Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
October 22, 2014
arrowPress Releases
October 22, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Debunking Asymmetry
by Keith Burgun on 09/23/13 08:33:00 pm   Expert Blogs   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutra’s community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.


street-fighter-2-7Like anyone else who grew up playing videogames, I have a great romance for “videogame-style asymmetry”.  What I’m referring to with “videogame-style asymmetry” is a quality that games such as StarCraft or Street Fighter have, where you “choose your character” or “choose your race” before the game even begins.

This kind of asymmetry is so beloved for a number of reasons, all of which I understand on a deep level, having experienced and immersed myself in it for my entire life.  However, after years of grappling with it, I’ve disappointingly come to realize that this kind of asymmetry is less than ideal for game design.  I am sad to have come to this conclusion, but I’m comforted by the fact that as always, I’m only talking about guidelines for ideal game design.  So, this doesn’t mean you can never make a game with videogame-style asymmetry, it just means that if you do, you should understand the costs.

I should qualify that this is distinct from “inherent asymmetry”, as is the case in the children’s game of “tag” (where one person is “it” and the other must escape), or even the popular videogame Counter-Strike.  To illustrate, in Counter-Strike, one team is always “the terrorists”, and they have a distinct objective.  The other team is always “the counter-terrorists”, and they always have a different objective than the terrorists.  It’s not as though teams select whether they’d like their team to be Terrorist or Counter-Terrorist and you could end up with a mirror match or something; there are no CS “matchups”.  And CS still has high levels of asymmetry due to the fact that you can choose your guns, which means it might be SMG versus Shotgun (although in practice, most of the time it’s M4 versus AK).

Beyond that, though, There aren’t many popular examples of more “strong” inherent asymmetry.  One that comes to mind is the boardgame 2 de Mayo, a two-player game wherein one player is the French, who have tons of forces pouring in from all directions on the map, and the Spanish, who have limited forces in the middle.  They have asymmetrical amounts of power, but also asymmetrical objectives: the Spanish simply have to survive at all until the last round in order to win.

The difference between “inherent asymmetry” and “videogame asymmetry” (maybe a better, albeit less-catchy term for it might be “componential asymmetry”) is that with inherent asymmetry, you don’t get to “configure the game”.  Sure, you can choose who will be the French/Spanish, or you can choose who will be “It” in tag, but it’s more like choosing who goes first than the process of selecting a character in Street Fighter.

Interestingly, each “matchup” in, say, Street Fighter, should be looked at as a distinct “inherently asymmetrical game”.  So actually, when you build a game with four characters, you’ve actually built ten “inherently asymmetrical” games in one (4 choose 2 = 6, plus 4 for the mirror matches).

With that said, please understand that this article is about videogame-style asymmetry – componential asymmetry – and not at all about inherent asymmetry like in 2 de Mayo.  So, whenever you see me saying “asymmetry” from here on out, assume I mean “videogame-style, pre-game asymmetry”, unless I specify otherwise.

My Problems with Asymmetry

Before I go about with the debunking, I should first list my general problems with asymmetry.

  • It forces the player to “play designer”.  When you have to make a non-strategic choice that has strategic ramifications, it creates tremendous cognitive dissonance.  “Should I pick the thing that I think is the best, or should I pick the thing that seems like it would be the most fair, or should I pick the thing that seems the most exciting, or should I just pick the thing randomly?”  Ultimately, the player is not free to simply play the game – they must first make decisions about what the game will be.  Other configuration options like choosing stages/maps/item settings/timings make the problem even worse.  In these situations, the player is under tremendous pressure to skirt some weird line between “doing what’s best for the game” and “doing what’s best for me as a player who wants to win”.
  • It tends to cause games to be vastly less elegant than they otherwise could be.  If you make a fighting game with just 4 characters, what you’ve actually done is create ten different games.  Each matchup is a distinct game.  For this reason, as well as others, I can only think of a few asymmetrical games that don’t have a ridiculous amount of content.  Most asymmetrical games – fighting games, card games, real-time strategy games, etc – have truly insane amounts of content.  At the time of this writing, League of Legends has a whopping 115 champions, each with four unique spells and a passive ability, not to mention unique stats.  Having to step into a game like that, or even a fighting game with 30 characters, is crazy.  It’s way too much stuff to have to learn, it causes individual elements to lose contrast, and…
  • It generally causes games to be vastly harder to balance than they should be.  In videogames, instead of pushing towards “balance”, we instead push towards “an acceptable tier list“.  This is to say that there shouldn’t be anyone in God tier(characters so powerful that you can only ever justify playing them), and there shouldn’t be anyone in trash tier (you can NEVER justify playing them because they suck).  But we accept everything else.  It’s just peachy that some characters are flatly better than others, and the reason we’ve accepted this is because with as much information as we cram into these systems, we just can’t really do much better than that.  In fact it’s a tremendous feat that we’re able to get a game with 30 characters to not have a trash/god tier.
  • It constrains dynamics.  Great games, as they are being played, emerge into a massive beautiful and mysterious web of dynamics – a resource is expanding over here, which is tied to some other resource over there, which is dependent on player one taking this action right now, which is possible because he took another action six turns ago, which in turn opens the door for a huge resource gain for player two three turns from now.  Because of this, only a few turns/seconds into most good games, you already have naturally emerging asymmetrical forces at work.  You can think of a player’s set of powers and resources halfway through the game as a “character” that grew out of the system.

    Videogame-style asymmetry, however, gives players a “quick start”, starting you with “forced” asymmetry that you chose before the game even began (i.e., it’s not a strategic decision).  The cost is that the game dynamics are constrained the entire game by a decision you made before the game even began, and they’re forced, not emergent.  That emergent character is now constrained by something you chose before the game even began.
  • It’s a smokescreen, making it harder for designers to really judge the quality of their system, which results in worse systems.  Videogame asymmetry makes a somewhat boring system seem more interesting.  If Street Fighter 2 only had one character, Ryu, then I think that the designers would realize that they probably need to make the system itself a bit more elastic and interesting.  But, since there is a forced-dynamic obfuscating the system itself – now it’s Ryu versus Zangief, I wonder how those two things push up against each other! – it’s harder to see that the system itself is kind of flat.

    Further, there’s a psychological trick that asymmetry pulls on you.  While you’re playing as one character, there’s a bit of a grass-is-greener thing, where you imagine other characters to be more interesting.  Not consciously, but in the back of your mind, that “wonderment” at not just seeing other characters in action, but how they will interact with THIS character, is compelling in a somewhat cheap way.  Even if you’re just going to choose Ryu and never play anyone else, you’re still going to play against other characters, so this effect takes place.

    I think that the above psychological effect is highly noticable with the card game Dominion.  While Dominion isn’t asymmetrical in the way I’ve been describing, it does have a “customizable” card market.  You swap cards in and out, and during the game, you get to combine them and see all kinds of effects happen when they’re put together.  Eventually, you reach a point where you’ve kind of combined everything, and then you either need to get an expansion, or quit.  So in a sense, asymmetry (or customizable-ness) ends up really just being a strange kind of “asset tour”.  You want to see all of the things.
  • It’s just not necessary and therefore shouldn’t be used, which I intend to prove throughout this article.  It dramatically increases the cost of production, and needlessly decreases elegance.  One quick and simple way to put it is this: if asymmetrical forces are necessary for your game to work, then you should disable mirror matches.  If asymmetrical forces aren’t necessary, then you shouldn’t include them.

The Status Quo on Asymmetry

Based on my experience and research, the generally agreed upon opinion right now is that Videogame asymmetry is almost always a “plus” in game design.  Indeed, it has many qualities that seem highly desirable, so this point of view isn’t surprising, even if it’s ultimately mostly wrong.

We can begin with one of the world’s most vocal champions of asymmetry, game designer David Sirlin, who has been striving to bring videogame-style asymmetry to boardgames with his games like Puzzle Strike and Yomi, as well as his game Flash Duel, which is essentially Reiner Knizia’s En Garde, but with asymmetrical characters added.  From his Sirlin Games “manifesto” page:

“It’s boring when every player starts with the same set of moves. All my games are asymmetric, meaning you can choose from 20 characters, each with a different set of moves and powers.”

While I have huge respect for David Sirlin, much of his writing, and his ability to design games (go check out Puzzle Strike if you haven’t already), the first line seems to me like an obvious over-statement.  He can’t really be saying that every non-asymmetrical game is boring, can he?  Keep in mind, I’m certainly not saying that “all games with asymmetry are boring”, or anything even close to that.  Only that it’s less than ideal and has problems.

Most designers don’t go quite as far as to say that symmetrical games are boring, however they do generally echo the idea that asymmetrical games are “less boring”.  Jon Shafer wrote an article for Gamasutra about asymmetry, which mostly echoes David Sirlin’s point of view, although makes a few caveats saying “not all games can be asymmetrical”, with the implication being that if they could all be asymmetrical, they would of course be better for it.

Or listen to this Three Moves Ahead podcast episode, wherein Troy Goodfellow, Tom Chick, and Julian Murdock muse about asymmetry in games.  While they don’t get too deeply into theoretical specifics, the mutually agreement that “asymmetry is better” is clear throughout.  Interestingly, the most telling moment is at one point when one of them said,

“Last year we had Halo Wars which was asymmetrical, and End War, which was symmetrical, and End War was… just a better game.  Which surprised me.”

This was followed by about two or three seconds of dead air, before someone chimed in to essentially change the subject.  Basically, it’s obvious to all of them here that even if End War really was better, that’s obviously just some fluke, some bit of meaningless trivia.  It’s surprising that a symmetrical game would be better than an asymmetrical one!

So overall, the status quo point of view is clear:  ASYMMETRY GOOD.  On why asymmetry is good, well… that’s a bit more fuzzy, although there are certainly some given reasons.  I will now go through some of the specific reasons and debunk them.  They are listed (roughly) in order of how common the defenses are, with Defense #1 being the most common defense.

Defense #1: Variety / Replay Value

I’m coupling “replay value” with “variety” in with this section, because I think they’re very closely related.  Essentially the argument here is, if a game doesn’t have asymmetrical forces, then it’s going to get stale after awhile, because you can’t try the game from a new angle with new powers or whatever.  Or put another way, if a game does have asymmetrical forces, it’s much more likely to have increased variety.

It’s easy to understand this claim.  If you are playing Street Fighter, and you’re starting to get bored playing as Ryu, well, you could always switch to Blanka, or E. Honda, or any other character, and get a different experience.

The problem is that this is a short-term solution that comes with all of the negative effects that I listed in the “problems” section of this article.  But the real myth here is that variety / replay value can only come from asymmetry.  You can create – and people have created – symmetrical games that are every bit as replayable, varied and interesting as any StarCraft or Street Fighter, but which is also way more well-balanced.

A quick example would be the great boardgame Puerto Rico, which quickly unfolds into natural, emergent asymmetry.  Puerto Rico uses a deep, interesting system that unfolds into a wide array of interesting outcomes, combined with some very light input randomness, with nothing even close to resembling videogame asymmetry.

Puerto Rico

Defense #2:  “Flavor”

There’s much talk about flavor – about different asymmetrical forces “feeling” different in a significant way, and the assumption is that this can only be achieved through asymmetry.  While I agree that it certainly is achieved through well-built asymmetry, my point would simply be that any great game, symmetrical or asymmetrical, would already provide a wide range of flavors for different strategies. For instance, this bit from Jon Shafer’s Gamasutra article on asymmetry:

The Zerg and Terrans could maintain their excellent art direction, but if they had basically the same units with the same abilities, only with different coats of paint, it would no longer feel like StarCraft. The Zerg are an organic swarm – their units should be more numerous and should have unique abilities like burrowing out of the ground.

So, “feel” also seems to be a major defense, and it’s not just “thematic” feel, but also mechanical.  I agree that the Zerg do mechanically feel different to play than the Protoss, and that’s of value.  The question is, can that kind of “feel” only be captured with asymmetry?


Unless you are speaking purely thematically, I think it’s without doubt that asymmetry is not actually needed to have this kind of “different feel”.  Good symmetrical games have enough strategic space so that players can exhibit vastly different playstyles.  For instance in the game of Go, playing more defensively and shooting for influence (a loose spreading of many weak stones around the board, often closer to the center) feels very different than a player who is more concerned with winning fights now and grabbing territory (and those are just two obvious examples; a more serious Go player can give you better ones, I’m sure).

Even if you look at an asymmetrical game such as StarCraft, you can already have vastly different playstyles even in a mirror match.  Maybe in a Terran vs. Terran match, one player chooses to put a lot of pressure on with harassment, or another playstyle could involve using lots of air units, or rushing.

A great example of how to do non-asymmetrical flavor is something like heroes in Warcraft III.  If the game was only Human vs. Human, there would still be plenty of emergent asymmetry in hero choice.  If you get an Archmage first, and I get a Mountain King, that’s a significant difference that really feels different.  If you get a second hero and I don’t, that makes it even more different.  And this game wasn’t even designed to be single-matchup; imagine a game like that that was.

In just the way that these designers are talking about how different races or characters need to have a strong sense of “flavor”, taking on different strategies also needs to have a strong sense of flavor – needs to feel different.  Actually, I think that the “feeling different” is just a byproduct of them being different, which is obviously important.  So if your game has strategies that really are different, then why do you need asymmetry?

The only remaining answer is “to support the theme/fantasy” – to make it feel like that scene from Starship Troopers.  Obviously, this is not a game design motivation, so it can’t be used to defend asymmetry’s role in ideal game design.

Defense #3: “Personal Expression”

The idea here is that players find a character that matches their “playstyle”, which we are to understand already existed on some level before the player ever played the game, and this game, being asymmetrical, allows them to express themselves more truly because of this.  So for instance, a very aggressive or impatient player might find that a rushdown character allows him to play the way he wants, or a person who is generally pensive might prefer a more defensive character.

My counter to this one is that I simply don’t even accept that this is really a thing.  I do not think that a person necessarily comes to a game with a “playstyle” before they even know how to play.  It does not make sense that a person would have such a thing.  It needs to be demonstrated that this actually exists.  At what point does it develop?  Was it there before they ever played?  I have a hard time believing that I could have a natural playstyle in some strange abstract game that I haven’t played or even heard of yet.  I suppose I could see it happening a bit for a player in two games that are similar, like if they went from Street Fighter to Super Smash Brothers or something, but hopefully going forward, fewer games will be so similar to each other.

But even if you do somehow “inherently” have a “rushdown personality” or something, a great game should have a wide range of creative strategy that allows you to express that.

Finally, though – even if this phenomenon really did exist, it conflicts with getting good at a thing anyway.  If you’re trying to play optimally, you can’t just keep doing what you “felt like doing before you even played the game”.  Playing games is about the creative search for more optimal moves, and good players know to try to ignore their own biases and that often times, getting good involves doing the counter-intuitive.

Defense #4:  “More is Better”

This is different from variety, in that it doesn’t actually care about the game experience at all.  This simply means, on the back of your box you can brag about having “50 characters!!!!“, which can seem like a good thing to many people.

By “many people”, I am mostly referring to young children, people trying to market to young children, and adults with the minds of young children.  I don’t think most decent game designers give this much credence, since it’s pretty obviously stupid.

At least, on paper.  I think that after 30 years of the more-is-better arms race, a lot of totally solid designers – myself included – have had their perspective a bit warped in terms of amount of content.  Videogames are expected to have a certain amount of content to be “a real game”, and asymmetry is often one of the ways that designers fulfill this quota, many times, unwittingly.  I think we all have to take a serious self-assessment of the damage that living through the past 30 years has done to us.  Especially since so many of us played Magic: The Gathering as kids.  That can’t have had good effects.

Mugen is mostly a gag game, thank god.

Mugen is mostly a gag game, thank god.

Defense #5:  “Easier to Learn”

I also listened to a recent Game Design Roundtable podcast episode that featured Mr. Sirlin, where he mentioned that in asymmetrical games, “the player really only has to learn one side”, suggesting that that games are easier to learn, since you technically only have to learn one character to “play”.

I put the words “play” in quotes, because if you don’t know the character you’re playing against as well, then you’re really just inputting moves into a black box and it’s spitting out a win or a loss – you can’t possibly understand what you’re doing or form strategies.

And unless you “restrict the game to 1 or 2 characters”, in which case the game isn’t really asymmetrical, you actually have to learn all of the characters to really play.

For example, I’ve been playing a bit of League of Legends recently, largely for research purposes, but also because I’m pretty desperate for a decently well-made modern online competitive game.  The claim that I only have to learn one character in this game is obviously pure trash – I’m trying to play that way and it isn’t working out so well.  I’m frequently killed by a character ability that I had no idea existed, or I waste a key spell on a character that has a way of getting out of it, etc.

Asymmetry lets you learn how to “control your character” perhaps more quickly, but “controlling your character” and “playing the game” are two different things, the latter involving “forming strategy” which is only possible if you know the other characters’ abilities.

Defense #6:  “Metagaming”

I played a ton of Warcraft III back in the day, and I can tell you: arguing about different races, heroes, and units is way more fun than actually playing the game.  I think that this statement is true or close to true for a lot of popular online multiplayer videogames.  Arguing about “tier lists” and such probably takes up at least as much time as does playing the game for many players.

Then there’s stuff like “popular strategies”, where some build or set of tactics becomes well-known and copied throughout the world, and then people start to develop counter-strategies, which themselves become popular, and so on.  This is a pretty cool thing, for sure, and an important part of a game’s “life”.

However, this isn’t at all a unique thing to asymmetrical games.  Research professional / high-level Go replays and commentaries online and you’ll see that it’s got all the same stuff going on as does League of Legends or StarCraft.

There’s just one thing that comes to mind that asymmetrical games provide for a “metagame” that symmetrical games don’t, and it’s a really ugly, horrible quality: “counter-picking”.  You see, not only are some characters squarely better than others in asymmetrical games, but some have significantly high win rates versus specific characters.  So, you can “counter-pick” your opponent by choosing a character that is better than that character.  This results in all kinds of weird, ugly meta-game rules, starting with a blind first pick, and then stuff like “the loser gets to change character and the winner can’t”.  So, this bouncing back and forth of picking, counter-picking, and counter-counter picking is actually optimal play.  Horrible.


I’m not saying you should never use asymmetry in your games.  There are some real, tangible reasons to use asymmetry, it’s just that almost none of them are good game design reasons.  However, I’m willing to accept that there could be some good game design reasons out there; it’s just that none of them have been used yet.

So here are some of the not game design related, but still good reasons to use asymmetry.

For one thing, asymmetrical forces can be really helpful in building a world/theme/setting, which sometimes can be of great value to a company that’s trying to get people attached to its products.

There may be some very technical thing that’s specific to your game system that no one has really seen before.  Some have talked about using characters as difficulty modes, which is a bit weird, but could work in a party game or something.

Asymmetrical forces could be used to create a smokescreen for a lackluster game.  Let’s face it: most games we make just aren’t all that great.  Sometimes having a “character select” might be just the sauce you need to get people to play your game even for a little while.  They played your game for a couple weeks, you made a sale – everybody’s happy.

Or, you just want to.  You, like me, have a great romance for asymmetrical forces, having been brought up in a world that romanticizes them.  I understand that, and there’s a really good chance that I’ll design an asymmetrical game or two in my lifetime.

Actually, wait – I made 100 Rogues, which has asymmetrical characters, didn’t I?  Huh.  And the funny thing is, I didn’t understand any of this back in 2008 when I started working on that game.  So I’m not saying don’t make asymmetrical games.  I just want people to understand the costs and benefits of doing so.

Oh, and if you have any good counter-points, I would love for you to convert me to a pro-asymmetry position.

This article was originally posted at my blog.

Related Jobs

Petroglyph Games
Petroglyph Games — Las Vegas, Nevada, United States

Illustrator / Concept Artist
DoubleDown Interactive
DoubleDown Interactive — Seattle, Washington, United States

Game Designer
Zindagi Games
Zindagi Games — Camarillo, California, United States

MOBILE Art Director
Treyarch / Activision
Treyarch / Activision — Santa Monica, California, United States

Senior UI Artist (temporary) Treyarch


Sean Hyde-Moyer
profile image
Wow. You lead with Star Craft, Team Fortress and Street Fighter, and your argument is you can't find a compelling reason for this kind of asymmetry?

Assuming that you're serious, and this isn't some masterful troll (if it is, well done, sir!), simply put: not everyone wants to play chess.

Player's don't give a damn about how elegant you design is. They want fun, warts and all, they want the thrill of discovery.

By all means, make your symmetric games. But don't think for a moment that the product is inherently better because it was easier to balance, or because the poor player won't have to learn to player terran and zerg and protoss.

If your thesis was: asymmetrical games are more expensive to produce, maybe. But your hyperbolic assault on all things asymmetric is hard to take seriously, given the number of hugely popular/lucrative asymmetrical games in the world.

You might also want to check out League of Legends, which I hear is doing ok.

Keith Burgun
profile image
The choices are not "either play an asymmetrical game, or chess." There are already tons of excellent symmetrical, modern games out there that millions of people play.

Also, your last line about LOL demonstrates that you didn't read the article (since I mention that I play League of Legends *in the article*), so I highly recommend you do check it out at some point if you have time.

Mike Jenkins
profile image
Chess isn't symmetrical.

1 player moves first, 1 player has the king on their right, the other on their left, etc. Like you said about learning the role of one side vs learning how to play: learning openings and defenses for black and white.

I also wonder where you draw the line. You talk about WC3 human vs human, how one player could choose one hero vs another. Why is the line drawn there? Why not when you enter the lobby? One player chooses human, the other chooses orc.

Keith Burgun
profile image
Right, well if chess isn't symmetrical, then almost no games are actually symmetrical, so I think the word loses utility. Also, I explain what I mean by symmetry / asymmetry specifically in this article - choosing a faction before the game begins, videogame-asymmetry.

Mike Jenkins
profile image
So regarding your Warcraft 3 example, a minute before the game launches, video game asymmetry. Under a minute into a 45 minute game, videogame symmetry?

Is DOTA symmetrical? You choose your hero after the game loads, like your hypothetical Warcraft 3 humans vs humans.

E McNeill
profile image
I'm sure I need to get more familiar with your concept of "ideal game design", but that definitely seems like a loaded start. Symmetry is surely more elegant and smooth, but you can't translate those directly to "good" or "ideal", especially considering the success (in many dimensions, not just popularity) of highly asymmetric games.

It's a little weird to me that you consider a character selection to be outside of a game system while you consider something like choosing a hero to build in Warcraft III to be inside the system. Can't we accept the initial asymmetry as part of the game? Why isn't that choice "strategic" (especially when there's a draft involved a la Dota 2, or when the different choices are well-balanced)? Sure, maybe there are flaws in the balance of some games, resulting in tier lists and the like. But those tier lists are always shifting subjects of debate. The metagame of Starcraft or Dota 2 shifts drastically all the time (even in between balance patches) as players discover and evolve their strategies over time. The competitive scene is full of strategic innovations and surprises! Sure, that can still happen in a symmetric game, but if it's a desirable thing (which I believe) and it occurs more easily when asymmetry is present, how can we consider that asymmetry to be a negative?

I should stop writing and go read the comments on your own site; I see that some of my points have already been made over there!

Keith Burgun
profile image
You can kind of swap out "ideal" with "good" if it makes it feel less loaded for you - that's all I'm really talking about. I'm concerned with the problem of establishing guidelines for good - or perhaps even "better" - game design.

E McNeill
profile image
Well, that's not my issue; I was trying to use "good" and "ideal" interchangeably in my comment above.

I was just saying that you can't say elegant == good, since "ornate" or "crunchy" or "chaotic" can also be good (and bad, but that's beside the point).

Keith Burgun
profile image
Oh, and I wouldn't normally post a link to my blog in the comments here, but being as the likes of David Sirlin, Raph Koster, Greg Costikyan and Jon Shafer have all posted there, I figure people might like to go check out that comment stream:

Ian Richard
profile image
Having great systems does not mean you made a great game. The player experience is what makes it a good game.

While you might not believe that choosing a character to fit your playstyle exists... many people do. I pick rogues and fast moving characters because I prefer precision games to tanking. This IS my playstyle and I feel happy when I pick a character who moves the way I want him to.
It doesn't matter whether it's real or in my head... my experience was better. It improved the game in the eyes of a player... and that made it a wise design choice.

"Game Design" isn't all crunch. Influencing a player's perception is a major part of game design. You hide the flaws while showing off the successes.

Heck... your own "It can be a smoke-screen to make a bad game look good"... is a clear indication that it can serve a positive purpose for a designer.

Sometimes symmetry is the right choice, sometimes it's not. A wise designer knows that it's about picking the right tool for the job... not one being better than all the others.

Matthew Sydlosky
profile image
I very much resonated with your assertion that asymmetry can cause a great deal of cognitive dissonance, as any person who has played DnD can tell you - it will take you longer to make your character than it will to kill it, most often. At the same time, that ~is~ a big part of the game, and it really has been used as a difficulty control for decades. The Virtua Fighter series even started marking different characters as being better for beginners or experts about eight or ten years ago, and tabletop games have had mechanical aspects of the same for even longer.

A drow wizard is totally a harder experience than a half-orc barbarian in a surface-world campaign where there's the sunlight and things that like to kill you running around and you've got a whopping three level one spells/abilities prepared with your 1d4 hp while the half-orc is sitting pretty in hide armor and has thrice your hp, a higher ac, and a greataxe.

You seem to either be alien to or have totally avoided engaging in the concept of role-playing. You argue that it seems preposterous that people develop playstyles, yet there isn't a genre of gaming in which people don't develop a playstyle - be it sports, racing, fps, tps, rts, rpg, or mario. Heck, even symmetrical games like tetris can show different playstyles and strategies. Some people ~do~ play fast and hard, others like to be subtle, others like to emphasize utility or survivability. This can be totally integral to making a gameworld seem realistic and alive, while also providing the ability for ANY player to relate to their side or protagonist.

As an example of a symmetrical game that is excellent, but not without flaws, we have the recent title "The Last of Us." A beautiful game, world, story, and experience for the player. Truth. Yet, at the same time it is problematic in a number of ways. Foremost among them in that it alienates women and permanently prevents female players from being able to relate to Joel (your protagonist) in the way a male player can. Throughout the game you eventually play as a female for a brief period of time, but it isn't a flattering experience per se. Ultimately, this inherently hurts the game in some ways. Now, as you've noted, the production costs to skin, model, voice, and write another protagonist may have been prohibitive, however it would have allowed greater market attachment, and made the game more emotionally engaging for half the population. Still a great game, but definitely lost some points with various demographics because of its linear symmetry. Yet the game has almost no replay value, you can collect a few things you may have missed, but if you play it on Hard the first time through, losing your hearing ability in Survival is practically moot, because you already know the game... every area of the whole game. It took me less than twenty hours, and I poked around considerably.

By comparison, I spent hundreds of hours in the original Neverwinter Nights, played through the campaigns several times with different characters, and did indeed have to develop new strategies to do so optimally as a result.

Other games, like Dota 2, LoL, any-fighting-game-ever, most shooters, racing games, and real-time-strategy games, and I suppose most rpg's as well (that intend to have multiplayer functionality, or replay value) are going to benefit from asymmetry. I want to be able to play the game as this character, and then I want to play as that character. I want to have different cultures that build different styles of fortifications, or use different troops to my advantage. Otherwise it really is as your previous detractor had said: playing chess.

You yourself built a hellacious strawman in your article by drawing on a gang of board games to champion symmetry in video games. While there are relations, there is also something about this that just doesn't sit well with me. Go has been around for a very very long time, and has a great deal of strategy involved, but that is all it has. Is it a great game? In some senses, sure, but would it be a great video game? No. It would just be digitized Go.

I think there's an app for that, it may even be free, but when we are talking about video games, we're typically talking about a customer investment of around 60 bucks, plus their utilities and console platform or pc. This demands a certain value/content ratio, and has a set of standards that players demand. You can say that most of them are stupid, and you may even be right, but when you say they are wrong, you are the one who is wrong - and that's stupid.

Your job as a designer isn't always to create the most fluid game, or the most psychologically accommodating game (though that argument is pretty bunk in games with characters, because characterizing and developing empathy between game characters and players will always suffer without asymmetrical pregame options) your job is to build the right game for the job, and the job is to produce a game that sells and makes everyone happy with the sale.

While games like Limbo, and The Last of Us, do a fine job making asymmetry their mode, there is nothing any less awesome about games like The Elder Scrolls, StarCraft, D&D, or even Splinter Cell, as a result of them having asymmetrical components, and to argue that they are 'less good' on those grounds is simply preposterous hipstering.

A brush for every stroke.

Rowan Carmichael
profile image
Wonderful comment! I love it a lot.

I pretty much agree with you on most points, however just a few things. Firstly, Keith is talking primarily about competitive games as such RPG and single player experiences are not really the area he is talking about. So, while I super appreciate your comment, I'm not sure how much Keith actually even desires to touch on those issues.

I don't really want to divert this too much... However the gender comment really gets me. I agree, player character male only games are off putting to women somewhat, and as such the power of choice is great. However, gender is not irrelevant either. Sometimes a story requires a male and sometimes requires a female. Many people have different relationships with their fathers compared to mothers, and they often have different relationships with their children for a variety of reasons.

Of course, not all stories lean on those aspects of gender, however, I feel confident in saying that The Last of Us covers perhaps more specifically a male story, as it relates to a father/daughter kind of relationship. Making that a mother/daughter relationship is very different.

I should say I'm not advocating not having female options, but some games do require a certain gendered protagonist for the purposes of their story. If you read about Mass Effect, you'll most certainly have run into player write ups who feel the game story, while technically the same, actually has different impacts when you play as a woman, but even just certain lines can feel like they have different impact.

Anyway, that's enough of gendered stuff.

Keith Burgun
profile image
>>You yourself built a hellacious strawman in your article by drawing on a gang of board games to champion symmetry in video games. While there are relations, there is also something about this that just doesn't sit well with me.

Your first sentence made me excited to hear about the strawman. Sadly, you never explained what it was. The fact that my talking about boardgames "doesn't sit well with you" isn't very convincing.

Stephen Gorinski
profile image
I think you did a great job demonstrating how asymmetry is so often a symptom of story or atmosphere fetishism in games rather than something driven by respect for the underlying mechanics. Too often do game designers make design decisions based on trying to support the "theme" or somesuch, which is missing the point of game design!

There is a clear failure in modern game development to respect games for the systems they really are, and I think you've made a convincing argument that asymmetrical design is a symptom thereof. Bravo!

Rob Solomon
profile image
This was a good, thought-provoking read, even if I don't entirely agree with it. Your counterexamples would have been stronger if they weren't all drawn entirely from turn-based board games. Games played in real time (e.g. basketball) frequently have additional design considerations, and with the physical capabilities and knowledge base of players changing, the rules of these games frequently have to evolve.

Mark McGee
profile image
I like this article, and I think it touches on things in game design that many designers don't think about much.

Relatedly, when I first heard about Divekick (seems like a long while ago), I thought it was a great looking game. It took the fighting genre down to its basics, which is a really good idea. Then they started putting in tons of characters and different matchups and stuff, and I lost interest. Instead of having to learn one game, I'd have to learn dozens of games in order to play it well. The benefits they gained from simplicity were then lost by adding unnecessary complexity.

RJ McManus
profile image
There are definitely some good points here, but as I was reading it gradually occurred to me that a lot of this doesn't apply to the kinds of "games" that I like playing most. This suspicion was confirmed by the seemingly oversimplified assertion that "playing games is about the creative search for more optimal moves", and only then did I remember the "forms" from your "Functional Theory for Game Design". In fact, most of my favorite "games" do not qualify as games by your definition, and I have a much easier time swallowing this argument in the context of goal-driven, competitive decision-making games.

Thus my question is then whether you think asymmetry is equally undesirable for other "forms". My own guess would be that the more control that the player has over their "goals", the more asymmetry is excusable. Asymmetry is simply a large part of the core premise of "games" ranging from Oregon Trail to Crusader Kings/Europa Universalis, ARMA, and RPG's, and to remove asymmetry would destroy the game's very essence. Particularly in role-playing games, starting off with an undifferentiated character is almost unthinkable (though there does seem to be a trend toward this recently). However, these games are less about competition and achieving pre-defined goals than most of the ones you mention in the article. I'm not exactly clear on the distinction that you try to draw between "inherent asymmetry" and "componential asymmetry", so perhaps you would call this "inherent asymmetry", but I do believe that there are some games that would suffer without asymmetry.

Keith Burgun
profile image
I don't even know how asymmetry would really manifest in a puzzle, or a contest for that matter. Like the idea of it seems rather preposterous to me, and I don't think anyone has even really tried it before. Puzzles (portal, layton) don't even have two "opposing sides", so what would "asymmetry" even mean really? I mean if you had a different character in Portal that had some "other" way of solving puzzles, it would just mean that that character essentially had *different* puzzles to solve.

For contests, I guess you could have asymmetry but I have a hard time imagining how that would manifest without adding decisions, and therefore, turning it into a game.

Inherent asymmetry is like in Tag or Counter Strike. You don't get to choose a character, there is only ONE matchup.

RJ McManus
profile image
Well, it seems that your notion of "games" seems to focus heavily on competitive multiplayer games, but for me "competition" as a criterion for "contests" and competitive multiplayer are two different things. Thus a singleplayer game can be a "contest" or a "game", with the player competing directly with the game's programming, and perhaps indirectly with other players in their own respective plays-through. Here I may or may not be re-stating arguments that your theory makes, but I do so in interest of clarification.

Perhaps you would disagree with that (I'm not sure whether your theory might argue that you cannot have competition without a clear goal), but if this is the case, then the player's starting situation (whether that entails character statistics, strategic positioning, or availability of resources) can introduce asymmetry.

Consider a historical grand strategy game such as Europa Universalis; the multiplayer modes are competitive by any definition and there is decision-making (even if the lack of any single goal may disqualify it as a "game"). The many countries one may choose to play as are certainly asymmetrical in many senses, and yet I would have a hard time seeing someone make a case that this is poor design. The same holds true for an ARMA mission that simulates a realistic battle (few of which are truly symmetric in real life), or an RPG in which you modify your character before beginning play.

Perhaps I am misapplying your Theory of Forms, but at any rate the thing that all these games have in common is the lack of any singular objective (whether that disqualifies them from being games or not), and therefore I would assume that this informs their natural asymmetry. It appears that much of your article assumes a strictly "gamist" viewpoint, but gamism is just one of many design concerns for many true games. Or would you not consider these examples of asymmetry in the first place?

Etelberto GomesMontarroyos
profile image
Problems I have with what you seem to be saying with this article:

A) Simple systems become solved puzzles faster. A symmetrical system is "simpler" to understand in terms of a player entering it. They can learn their aspect and by doing so, immediately understand their opponent's aspect.

B) Solved puzzles become boring. (I'm referencing a Theory of Fun for Game Design with this point.)

C) Players by definition are asymmetrical. This is why even chess is asymmetrical. Since chess requires two players, the design itself is asymmetrical. (Rob Solomon's point I believe)

I agree with your points for the following reasons:

1) Team based games can have asymmetry in the roles played by the gamers (Tanking, Support, Damage Dealing etc), but could have instead symmetrical choices for those roles (for example in LoL if you played a top lane, you would have the same top lane on both teams, same for jungler, support etc).

2) Players at a certain skill level would negate the asymmetry of players mentioned above.

3) Chess has not stopped being fun despite its nearly solved nature, due to the size of the complexity on the board. Same thing with Go.

Chris Clogg
profile image
I think the bottom line for me is that there's something inherently fun about picking your 'team' or 'hero' and feeling connected to that choice (and different from others).

We tried to do that when creating our 3 races in Stratosphere, but as you pointed out in the article, the issue of balance blows up very fast. Sadly, over time we had to remove some of the differentiation... and the 3 races, while still different, ended up being quite a bit more similar than we wanted. An example is one race had splash-dmg and the other 2 didn't, which ended up leaving those 2 completely screwed from mass waves. So we had to chunk some splash into a tower for each of those teams.

John Trauger
profile image
D&D through version 3.5: high degree of asymmetry.

D&D 4E went for low asymmetry and that is exactly what players hated about it.

I've played both. Both are good games, actually. But the low asymmetry of 4E created an unacceptably different experience for some players.

In both pen and paper RPGs and video games, we have the min/maxers, the power gamers, who love identifying and exploiting an asymmetrical advantage. even in $e the most effective characters were the ones that built up an asymmetrical advantage rather than a jack-of-all-trades generalist.

John Trauger
profile image

Duplicate post, sorry.

Peter Christiansen
profile image
I would agree that there are certain things that work best as symmetrical games. The symmetrical nature of Chess and Go, for example, is why they are considered to be such definitive measures of intelligence and skill. They both have a very simple rule set, which means that the winner is often determined by one player having just the slightest advantage in skill, adaptability or mental stamina. If your ideal of the perfect competitive game is this kind of fine-tuned scale of skill (and for many people it is), then symmetrical gameplay is generally superior.

For many people, however, even those who enjoy competition and rankings, this is not the only (or even the primary) motivation behind playing games. Limiting yourself to symmetrical designs also limits the kinds of experiences you can create. For example, take the differences between Risk and Axis & Allies. Risk is a great game with a lot of replayability and does so while being (mostly) symmetrical. Axis & Allies is a very asymmetrical game, with a team of three taking on a team of two with all five varying greatly in starting strength. Of the two games, I find Axis & Allies far more interesting and fun than Risk. The asymmetry between the players allows for surprising strategies and turns of events that would never be able to materialize in a more symmetrical game.

In addition to the enjoyment of the game, asymmetry is important for the context and framing of the game. A World War II game that started Germany, the US, Russia and their allies off in a symmetrical "Risk" style would both feel less authentic to the theme and be a less useful tool for understanding the strategic aspects of the war.

While symmetry certainly has its place, suggesting that symmetry is generally superior to asymmetry is essentially saying that "People should play games for this reason, not for that reason."

Dusty Hunsaker
profile image
I love how controversial this article is, thanks for writing it. However I'd like to offer a another perspective on some things.

You mentioned that learning the thousands of matchups with different characters is tiring. I'll agree with that. But that could also be part of the design. There is a cascade of information that happens when you first start playing a game like that. At first you get destroyed by everything. Then you come to understand that there are different classes of characters. When a new character is released, they generally excel in one of a few basic archetypes (Tank, Ranged Damage, Melee Damage, etc.). You learn these Archetypes before you learn the thousands of matchups. You know that if you are a Melee character, you have to get close to a Ranged character to do any damage. Then you might learn the nuances of specific characters.

The same applies to a game like Halo. You learn that there are close range weapons and long range weapons, weapons better for shield damage vs health damage. Does it matter if your opponent is shooting at you from across the map with a DMR vs a Covenant Carbine rifle if all you have is an SMG? No, it doesn't, all you need to know is that you are going to lose the fight if you challenge them. When you eventually learn the nuances of the DMR vs the Carbine, you can leverage those advantages. You know that because the carbine shoots faster, but does less damage, you can have a more consistent damage output at closer range, because if you miss once, you only missed 12% of the damage you were doing instead of 25%. You also might learn to strafe or jump right as the DMR person is about to kill you, because they can't follow up as quickly, whereas the Carbine wielder can.

The point, again, being that there is cascading information: General game knowledge -> Archetype knowledge -> Individual nuances. Much of the archetype stuff carries over from game to game as well. I can probably assume that shotguns have a weakness at long range, because that's consistent. What you might not be able to carry over is that a shotgun in Counter-Strike has a much tighter cone of pellets than one from Call of Duty. That's a nuance you have to learn from specific game experience.

I'd also argue that it's just not possible to learn every nuance. Even professional level players make educated guesses about things. I'd also argue that's exactly what gameplay is, at its essence. Making a guess based on past experiences, and validating the results of that guess against the current opponent. I might make a guess that, because my opponent is holding a DMR, I should rush in with my SMG to take them down. I turn the corner and get blasted by a shotgun, I lose. My guess was that I could beat a DMR close range with my SMG, and their guess was that I would rush in with my SMG, so they should switch to their shotgun. We make thousands of decisions like that while playing a game. I lost because I didn't account for the human differences.

I have a friend who plays fighting games by playing against the other humans, and focusing less on the actual gameplay. If he notices you typically block when he approaches, he will typically grab you out of it. Finding patterns of human behavior makes him a naturally good fighting game player, even if he doesn't know the full move set and frame data of every character. All he knows is that I have made a mistake, and he is punishing it. Sure, I'll catch him off guard every now and then with something he didn't expect, unreactable dive-kicks, invincibility frames during move startups in Smash, taking advantage of disjointed hitboxes, but I'd still consider him a better player than I am, even if he doesn't know all the nuanced stuff about the game.

My ultimate point is that symmetry and asymmetry can live in harmony with one another. Some people care about the individual nuances in games, and some people care about the differences in human behavior. They both work together to make a game compelling and interesting, and aren't mutually exclusive.

Patrick ODay
profile image
Let me start by paraphrasing the argument: "For competitive players of competitive multiplayer games asymmetric starting conditions are not inherently beneficial and can actually be detrimental to competitive play". Notice the repeated use of "competition"; there are several types of play that don't involve competition with other players and competition against the computer is inherently asymmetric.

The article seems to declare that asymmetry from starting conditions is bad but emergent asymmetry is good (choosing different heroes in Warcraft 3 is the example used). But the article doesn't make clear what divides one from the other. Is the dividing line the fact that the choice is made "before" the match or because the decision can't be undone? Can't my choice of hero in warcraft 3 be seen as much as emergent asymmetry as choice of race in starcraft?

I agree that asymmetry adds production costs, can hide bad game design, and increase the learning curve but that can be said about any feature. Adding RPG elements to non-RPGs has the same issues. But adding different game design elements doesn't unilaterally make the game worse, it makes it different which means it's experimental.

Alfa Etizado
profile image
This is the reason I don't buy the play to win nonsense that Sirlin champions so much and that has unfortunately spread through the Internet like wild fire.

A lot of competitive video games are deeply imperfect because of the asymmetry. On top of that, they have an aesthetic layer that make things further worse.

You have to imagine that the designers don't just have to come up with good actions for the game, they have to come up with actions that look good. Say, someone designs a character, just the looks. That automatically shapes the rest of the design. The character's HP, how hard it hits, how far it hits, how it hits.

When you have that with a dozen of characters, it's clear that the priority for the game wasn't ONLY creating a competitive product. If all they cared about was competition they wouldn't bother with these extra layers that exist purely for flavor.

A ton of competitive games are designed to be fun AND competitive. They're highly imperfect, sometimes good things come from this imperfection, interesting exploits, other times it just creates extremely unfun dynamics, like spamming.

There's no problem in taking the competitive side serious, by all means do. It just bothers me when people shut their eyes to the fact that the game is also designed to be fun. So what you said earlier, about the player "playing designer" that should be accepted, it is part of the game. Ignoring the fun aspect is just as bad as ignoring the competitive aspect of the game.

Play to win ends up as an excuse to use annoying tactics, like spawn camping, that clearly fly in the face of what the game's about.

Steven Stadnicki
profile image
Chris Clogg has gotten at some of this above, but to couch it in 'designer dictionary' terms: asymmetry in a game design serves as a meaningful way of creating _player agency_, and does so in ways that are much more difficult to do in symmetric games (but not impossible; hmm... *goes off to design*). I think this article does a fine job of noting some of the often-underestimated disadvantages of asymmetric design (I especially like the callout of the 'ten different games' aspect of balancing), but (as others have noted) slips up where it asserts the supremacy of symmetric designs. There are strong and legitimate arguments on *both* sides of the coin, and as with so many other design choices, what's left is just that: a choice. Be aware of the impact it will have on your design one way or the other, but there is no 'right answer', only a better and worse answer for specific games (and/or for specific designers.)

George Booth
profile image
You lost me when you called MUGEN a gag game. It's a fighting game engine. What people put into the game is their own prerogative, and it's possible to have a completely balanced game should the user decide to put the time in. Then you have things like Salty Bet, where it's AI fighting, and the game itself becomes about betting on the outcome with funny money.

We have a tendency to overbalance things in this day and age. Everything has to be fair and square for the competitive scene, despite the imbalances of previous games which led to the popularity that the current installments enjoy. The best example of overbalancing I have is MVC3 and UMVC3 (earlier MVC titles weren't known for their stellar balancing).

Generally, I find myself playing MVC3 with my friends, despite the fact that we have both versions. There is more fun to be had because the balancing was not thrown out of whack by the competitive input (too much). Sure, Hulk with X factor can wipe out an entire team, but we're talking about The Incredible Hulk! The beast that rips Wolverine in half without even thinking about it, then hops about his day! Not to mention X factor letting you milk out some super epic low-life fights at the end of a match.

From a competitive standpoint, it doesn't make sense. However, from a gameplay perspective it does. Games aren't about making things fair for the player; they're about challenging the player and making them earn their victory with the tools provided. Whether it's in the competitive field of play or in a single player action game.

There's also something to be said for taking the harder route development-wise.

Dan Felder
profile image
Keith, I applaud your counter-intuitive takedown of an accepted element of game design. Deep analysis and addressed logically rather than emotionally.

I have a slight reservation with your analysis though and was wondering if you could elaborate on it. You mention that symmetrical games can have vastly different playstyles and thus fulfill the desire for a different feel or style of gameplay. However, can you explain the meaningful difference between...

A) One faction with the tools for 3 possible strategies it can embrace.

B) Three factions, each with the tools for 1 possible strategy that it can embrace.

Just because the asymmetry isn't created during a character-select screen doesn't mean that all the same problems don't apply once the game gets rolling. In fact, the one-faction model is often going to be significantly harder to balance than the three-faction model, because it has more interactions between its tools to account for.

I'm not discounting the article of course, but it seems that one of your examples of how asymetry isn't necessary brings in a design decision that is, for all intents and purposes, asymetry as well - just not chosen during character selection.

Keith Burgun
profile image
You don't want to front-load your game with the most important decision right at the beginning. So if you had situation A, I assume that it has to COMMIT to a strategy right at the start, right? My problem with asymmetry is that you're committing to some kind of strategy before the game has even *begun*. Like what information are you even basing this HUGE strategy choice on? Your whims?

Dan Felder
profile image
Dang, just saw this when I have to run out the door. All I can offer is a quick shallow response to be elaborated on later.

Since you brought up Go, I"ll bring up chess. The opening is one of the most interesting and engaging elements of chess strategy. Gambits vs. Defenses vs. Hypermodern strategy and much more. The decision is effectively made at the beginning of the game with very little input from the opponent (the only input has the same feel of 'counter-picking' that you deride). I'm running now, can't offer an analysis of this (or even if it's a positive element), but wanted to bring up the example to get your feedback.

Dan Felder
profile image
Alright, I have a few more moments now. I've thought a lot on the subject, but I think that a single example is going to articulate a core reason that asymmetrical design can be helpful.

In one of my projects, an asymmetrical tabletop game where players battle one another via decks of cards - after choosing a specific, unalterable deck for that represents specific character, we used the fact that each deck could be developed separately and kept self-contained.

Each deck contained 1 super-card that was the character's trademark power and was more than 3x as powerful as a normal card (but not so powerful as to decide games on its own). These cards were a lot of fun to use. Because the game was meant to be extremely accessible, it had no cost system for playing cards - just a limit on how many cards could be played each turn.

This way we could make lots of super-cards and include them in the game without ruining the balance - because we could guarantee that each deck only had one. You can't get that element across as elegantly via a symmetrical game. The asymmetry opened additional design space while allowing it to be tightly controlled in execution.

If you want to design 200 tools for players to work with - it's significantly harder to balance all those interactions. A way to manage it while also giving players a sense of additional theming and identity is to use asymmetry.

Also, your point about a player having to learn dozens of matchups being a barrier to the casual player is flawed. This creates a sense of, "easy to learn, difficult to master". The player starts by learning a smaller amount of information and, if casual, is often content to focus on how to execute that faction's options. For players seeking mastery, they're given an extraordinary opportunity to learn counter-tactics.

TLDR; In a way, you're looking at asymmetry backwards. Asymmetry is useful to make certain kinds of games that are already good systems *easier* to balance and make life easier for new players. When assymetry helps manage existing tools in tight little packages, it's helpful. When it's just adding more content for the sake of more content - it's a problem.

Super Duper Garrett Cooper
profile image
I'm not convinced that a choice of race or character at the beginning of the game is any different than a choice of strategy once the game starts. This asymmetric choice is merely the first strategic choice the character makes; in symmetric games, the player only starts to make strategic choices after the game starts.

I definitely agree that asymmetry makes a game much harder to balance, but that's the price one pays for depth.

Keith Burgun
profile image
Choice of strategy can be changed after the game begins.

Jake Forbes
profile image
Thanks for the interesting discussion here and on your blog.

Puerto Rico is an interesting example for the symmetry discussion, as while the game is more or less symmetrical over the course of many games, it's highly asymmetrical at the turn level and in terms of starting order. You could say that the appeal of Puerto Rico and other variable role or worker placement games is that it gives the fun of asymmetry in a more accelerated way since each turn is imbalanced in interesting ways -- it's the "horrible metagame" sped up.

Ultimately, asymmetry is an aesthetic that is neither the solution to nor the bane of good design. For those of us who find this type of discussion interesting, what we're ultimately looking for is games that become more interesting the longer you play them (or at least games that don't take a strategic nosedive while our enthusiasm is still high), and the asymmetrical aesthetic can be a great solution, if one that takes more effort and patches to get right.

Bob Fox
profile image
I'm sorry but the author shows no understanding here. He tries to use league of legends of an example of 'balancing 100+ champions' without understanding 99% of those champions are just reskins of a limited few archetypes of old champions with renamed and reskinned skills with only the most minor of differences.

Kevin Maxon
profile image
1) As others have pointed out, it's hard to know when a game is symmetrical. If you count army selection as part of the game, Starcraft gives players symmetrical choices. Their first choice just happens to be really important.

2) RE: "playing designer," it's faulty to assume that players will, within the confines of a system, make decisions logically. If they're playing to win, they will likely use character selection competitively. But many people play playfully, and making non-optimal decisions is part of that.

Ron Dippold
profile image
Well, I thought it was an interesting polemic... but cognitive dissonance and user challenge are still useful tools, and many users still seem to appreciate the extra cognitive load even though I'd agree that the majority just want minimal obstacles to slackjawed reward. But I don't want to make games for those people, which is why I didn't jump ship to Zynga at the time. Consider me still bunked.

Shane Murphy
profile image
From what I have read, most of the things I have to say have been expressed either here or on your board, Keith, but I found this article so provocative that I'd like to put in my two cents.

To start, thanks for writing this. I don't agree with your stance in general, but you make some compelling points.

There are a few places where your argument breaks down.

First, you need to define what you mean by "ideal" or "good" game design, or better yet use more specific terms less laden with evaluative baggage. It's really hard to assess anything you say without having established premises for your argument.

Second, while asymmetry is not technically necessary to create a complete game (you can play Ryu vs. Ryu ad infinitum) it does allow for a fuller exploration of the game's mechanics. Sure, you can have a symmetrical RTS with emergent strategies and asymmetries just waiting to be unearthed from a mine of near-infinite possible inputs, but imagine a fighting game like that. Imagine a single character having every single normal attack, and every single special attack in Street Fighter. Imagine what an unwieldy control scheme would be necessary to play that game. In Street Fighter, asymmetric character selection allows a for a deeper experience of the mechanics while still rendering the game, y'know, playable.

Third, asymmetry allows the designer to package complex arrays of strategic options into digestible, meaningful chunks. In a symmetric game, the player might have to muddle around for ages before determining the precise combination of preambulatory moves that would allow him to optimally proceed with a chosen strategy. Heck, it might take ages of muddling just to figure out what his strategic options are. Asymmetry lowers the barrier to entry by saying, "Hey player, this is a grappler. He has to work hard to get close to his opponent, but if he does, he dominates. This is one way to play this game." Keep in mind that even after the player has selected a character, the number options available for getting in and/or dominating is staggeringly large. Large enough that two Zangief players, for example, can play completely differently and still meet with success. So asymmetry grants accessibility at negligible cost to depth.

Tangentially, the theme/fantasy of the game can be used to communicate the mechanical differences between these choices. One look at Zangief tells the player that he won't work the same way as Ryu. While this doesn't necessarily provide perfect information to the player, it is an unobtrusive way for the game to teach players about itself.

Finally, I will be petty. You say that the content creation mandated by asymmetry causes "individual elements [to] lose contrast." Can you really levy that as a criticism against asymmetric games when the amount of contrast you're advocating is, um, none?

Thanks again for the article!

Randen Dunlap
profile image
Obviously lots of dialogue back and forth here, with some validity on both ends of the spectrum.... but there's one thing that stood out to me...

It forces the player to “play designer”

While I agree that TOO much of this can cause paralysis from excessive choice, I think there's value here.. when done correctly, I believe players enjoy and value "Playing Designer".

I know it's been said already, but design is about knowing the right tools for the job. "Different strokes for different folks", there is no "hard and fast rule" or "One tool set to bind them" type of thing in design, and personally I believe that's what is the most beautiful about the discipline.

Lastly, life itself is asymmetrical, so I believe when done correctly, asymmetrical systems naturally make sense and resonate with players. This discussion should be more about the strengths and weaknesses of both, and how we can strategically use them to our benefit, and not on how to build a case against one or the other to disprove its usefulness.

"Only the Sith deal in absolutes..."


Duncan Hendrick
profile image
Great read, thanks! You brought up Magic: the Gathering very briefly, which is probably my most frequently played game. I love it, but I do realize it has tons of game design flaws. I'd be interested to hear your opinions on the game if you've ever given it serious thought.

But more specifically in regard to asymmetry and Magic (which has a lot of it), I have a few questions. Is it possible to make a TCG without asymmetry, or are TCGs asymmetrical by definition? If it is by definition, does that mean a TCG is then, by definition, a non-ideal game design?

Also, Magic (and other TCGs) have created a way to turn the asymmetry into an additional game in the form of drafting cards. By the base definition of game design it's all there. The hidden information of what other cards are in other packs and what others are picking leads to a lot of ambiguous decisions that have major consequences. And this is all before the "actual game" even begins! Is this a reasonable answer to a problem with asymmetry, or do you think it's another form of applying makeup over a flawed system?

I could go on, but I'll save you here. It's all very interesting so thank you for getting my brain going.

Keith Burgun
profile image
>If it is by definition, does that mean a TCG is then, by definition, a non-ideal game design?

There are already like 5 other reasons why TCGs/CCGs are far from ideal game design patterns, but yeah, I'd say so.

Shern Chong
profile image
Keith, I have a few games I'd like to know from you what you discern as asymmetric or otherwise, before I continue asking.

- Company of Heroes + Opposing Force
- Homeworld 1
- Savage Heroes of Newerth
- Giants Citizen Kabuto
- Battlefield 1942
- Bushido Blade 1 & 2
- Streets of Rage
- Alien vs Predator (coin op)
- D&D Shadow Over Mystara

Thanks Keith. Your inference of those examples will help me know if I understand your article, and your meanings of asymmetry.

Rich Chatwin
profile image
Some games I've played in the past that were good:

1. Total Annihilation - symmetrical
2. Command & Conquer (original) - asymmetrical
3. Quake 3 deathmatch -symmetrical
4. Day of Defeat - asymmetrical
5. Starcraft 2 - asymmetrical
6. Aliens vs Predator - asymmetrical
7. Counterstrike - symmetrical
8. Team Fortress- asymmetrical
9. Natural Selection 1/2 - asymmetrical
10. Shogun 1/2 - symmetrical

In summary, there are compelling arguments for both sides (all the above, I believe, are fairly highly regarded). But ultimately, if it works, it works, right?