Get updates via e-mail:

« Addiction, Diablo 3, and Portal 2 | Main | Puzzle Strike is Coming »

SCG4 Update

It's been over a year since I said anything about "Sirlin Card Game #4," the customizable but not-collectable game that I've been kicking around for over a decade now. Over the last year it's solidified quite a bit and gotten a lot of polish, too. If you happen to be going to PAX Prime this year, you can find me at the Game Salute booth to maybe get a chance to play it.

I thought I'd share a bit about a design problem that I kind of accidentally stumbled into an answer for. That's how things happen sometimes. First, the things that have explicitly been goals all along:

1) Make a game that would be interesting to play for 10+ years without any new cards being released.
2) No mana-screw.
3) Inject some "characters" into the game.

I won't say much about points 2 or 3. Regarding "no mana-screw," hopefully it doesn't take much explanation to see why you wouldn't want to randomly be locked out of even *playing* in a supposed strategy game. Point 3 is a matter of preference, and I just think it feels better to have characters and personalities to connect with in a game. This one was hard to figure out, but the current implementation really adds to actual gameplay as well as feel, so it worked out well.

It's #1 that's the real big one. If we can't rely on new cards every 3 months, it means the game actually has to hold up past that point on its own merits. If we can't rely on the metagame constantly changing, it means the game itself will need to have enough depth to support years of play at a very high level. In order to make that possible, the codenamed SCG4 gives you access to a much larger set of effects than you'd normally have during a game of any other customizable card game. More than you can use in any single game session, on purpose. And furthermore, much finer control over when you draw those effects. These two things together mean that you have much more *versatility* in how you play any given game. You can pursue pretty different strategies even without changing decks, and you can change which strategy you are pursuing during the course of a game--in response to how your opponent is changing his strategy.

Customizing Out the Fun

So that's all just great, isn't it. But a while ago, I forget how long, maybe a year or year-and-half ago there was a problem in the back of mind with this. The decks I was building for playtests were fun and all, but I was looking for that fun. I generally included about three different sets of strategies in these decks, and that was very good for gameplay. But what if a player who was playing to win built a much more boring and shallow deck? What if someone made a deck with only 1 strategy, but it was 20% more effective than any of my individual 3 strategies would be? We could debate which is actually a smarter idea, but if there is any chance that the more boring and shallow version is more capable of winning, that's going to really suck for the game.

Before going on, we should take a look at the more general problem that exists in all CCGs: unfair matchups are not only common, but often considered a good idea. If you can develop some deck that has really strong matchups vs several decks, but really weak matches vs only a few, you did a great job as a deckbuilder. You might win the tournament even, but you will have possibly played all unfair matchups, one way or another. For a more concrete example, a friend of mine told me about a Magic tournament he entered where he expected the (red) Goblins deck to be *most* of the field. He built a super hate deck directly against Goblins that included 12 maindeck protection from red cards, just for starters. He gave little thought to beating non-Goblins decks, though probably he had a sideboard to help as much as he could against those.

He told a pro player at the event that his deck was almost 10-0 vs goblins. The pro player said uh no it isn't, so they played several games. Eventually, the pro conceded that he didn't see a way that goblins could win at all, ever, vs that deck because it was just so extreme. My friend got 5th, but only due to an unlucky draw at the end. The bracket had: goblins, goblins, goblins, goblins, u/w control, and him, and he happened to face u/w control. He placed high, and he could have won the entire thing. What's most notable here is that 100% of his matches had bad gameplay. In every case, when he sat down to the table, one player or the other had overwhelming advantage.

Losing Before You Even Sit Down at the Table

Let me use the word "gameplay" to mean the part where you sit down at the table and play cards until someone loses. You could say "gameplay" also includes deckbuilding and metagame choices, but let's not, because then I'd just need some other word for when you sit down and play cards. The part where you sit down and play cards--the "gameplay"--really should be as generally fair as we can make it. I don't see it as a virtue that 8-2 or worse matchups are frequent things. It's clearly a bad property when fighting games have lots of highly unfair matchups, and it's something we work hard to fix there, rather than applaud.

But what can you DO about this? (Sideboards barely count as a good answer. They do literally nothing for game 1 of a match.) Deckbuilding is fun and captures the imagination, and that's what we're running up against here: deckbuilding is allowing unfair matchups to exist and to be common, even. If we limit deckbuilding, that sounds less fun. And so I didn't even really try to solve this problem, I just kind of gave up on it. And then something happened. Two ideas looked a lot alike, it was a clue.

When All Decks Really Interact...

In pursing that goal #1 of making the game interesting to play for years and years, we have to care a lot about your interaction with the opponent. You really need a lot more interaction than you get in most CCGs. There just has to be more to it if you're hanging your hat on the depth of gameplay of a single deck giving you YEARS of strategy space to play in. So what about cases where what you're trying to do is so different from what your opponent is trying to do that you hardly interact at all? I have joked amongst playtesters that we "force you to have fun" by making it not really possible to do that. You pretty much have to interact. And if we theorized about this or something, we might think oh that really limits what you can make! You can't make some solitaire thing that has no interaction at all. In actually playing it though, it feels the opposite of limited. Because the general game system has a lot more decisions going on than in other CCGs, "I feel so limited," doesn't come to mind, at least not to me. If anything, you have a wealth of choices and oh by the way, you can't go off in the corner and completely ignore what your opponent is doing.

Another way of putting this is that there's an illusion and reality that are at odds here. If we allow you make solitaire decks, it feels like that's more choice. In reality though, it's allowing choices that hurt the quality of the game overall as an interesting strategy game that can last years. It's like wanting freedom in your country, and saying part of that freedom is to murder people indiscriminately. In that case "more freedom" is a somewhat misleading label.

Limiting Deckbuilding to Create MORE Viable Decks

Back to the whole deckbuilding thing: it's exactly the same there. What if players want "more freedom" to build decks that cause the game to overall have a lot worse strategy? Wait...why are we allowing that? Deckbuilding is fun and exciting, but it has that same illusion, unfortunately. When you have more and more cards and more and more freedom to make anything, the illusion is that you get more and more choices. But what is more common is that you get more and morely likely to degenerate into just a very few choices, or one choice. Just imagine a CCG with 550 cards that you can combine however, and how many tournament viable decks there are going to be in that game. Yomi, a fixed deck game, has 550 cards that compose 10 such decks, but you'd be lucky to have even 4 if it were customizable. And there's just no way those 4 would end up having all 4-6, 5-5, and 6-4 matchups against each other. So you'd have the illusion of way more choice, but actually end up with fewer viable choices, and more unfair matchups plaguing the few choices you have.

So I realized that when I was including three different strategies in these decks, I was really on to something. This evolved to be a more and more central part of the game. Interwoven with the "heroes" you control, and part of the back-and-forth strategy where you and your opponent can each shift around what you're doing as you play. What seemed years ago like a bad limit to place on deckbuilding has now become (accidentally?) one of the best features of all. Yeah there are limits, there are chunks of your deck that have to have certain kinds of things. So how has that turned out? Has it made me and other playtesters sad?

The answer is that it's resulted in so many viable decks that we are overwhelmed. It's possible to make over 815 different decks with all the cards that exist today, and every one of those differs by at least 33% from every other one. And here is the most incredible part of that. I don't know which of those 815 is the least powerful, but whichever one it is, I do know that it's at least as able to win (and probably a lot more able to win) than the worst character in the average fighting game. So when I say this huge number of decks, I'm not talking about useless stuff like "all lands" or "all 1/1 creatures with no way to play them." I mean those are all real decks that can do coherent things and win in the hands of a skilled player. Somehow the "limits" on deckbuilding have produced more decks than we even know what to do with--every one of them playable.

In Closing

The Yomi expansion will be the next game I release, and that's going to be a while, so the game mentioned in this post will be even a while after that. I honestly don't know how I'll pay for the hundreds of thousands of dollars of art such a game needs, but one way or another you will get to play this game. I'm determined to eventually release it because I think it's incredible and I just don't know anything like it. Probably the higher the sales of the Yomi expansion, the sooner I'll be able to finish "SCG4."

Reader Comments (60)

This is why I hate solo-queue League of Legends. 95% of games are vastly imbalanced in favor of one side by A: the champion picks, B: the jungler setting up shop in one lane, or C: certain players just being off their game or unfamiliar with their champion. It's fun when games go even, and all the mechanics work pretty well in a team v team context, but it sucks bad to have games more-or-less be a coin flip based on everything but the phase of the moon :x

August 13, 2012 | Unregistered CommenterRCIX

This sounds awesome, Sirlin!

I can't wait to see it!

August 13, 2012 | Unregistered Commenterajfirecracker

can i guess the mechanic: cards have new or changed effects based on what cards your opponent is using. I'm always interested to learn how you solve the problems you choose to take on. You seem to be one of a few designers who tries to do new and interesting things, so I hope you make enough money to continue. good luck with this.

August 13, 2012 | Unregistered Commentermas

Uhh... so, how do you know all those 815 decks are viable if there's more of them than you can play? I don't mean this in a snarky way, but as a serious question. Is there some property to the deckbuilding system where if you try to make a really skewed deck, its efficiency goes way down, so the only reasonable decks have relatively predictable power levels and ways of interacting with each other? Or what?

August 13, 2012 | Unregistered Commenterpkt-zer0

The more I read about this game, the more I want to play it. Too bad that I can't be at PAX Prime. :(

After trying out over a dozen CCGs over the years, I would put "no mana-screw" and "no unbeatable match-ups" at the very top of a list of elements that a good CCG should have. It is just sad to come up with an interesting deck idea, only to realize that this deck will be beaten mercilessly be the deck that currently sees the most play. Now, if no match-up were unwinnable or if the playing field were so diverse that you wouldn't run into your worst match-up that often, this would make trying out that cool new deck more viable. David, if you have found a way to help on both accounts, then kudos to you!

Being able to switch strategies during a game instead of being locked into a single one from the beginning of the game sounds pretty cool, too.

August 13, 2012 | Unregistered CommenterStar Slayer

David, this game sounds great. I think you hit the nail right on the head with regards to the problem inherent in games with totally customizable decks: most of the best decks just aren't fun. Up until recently I played a fair amount of magic the gathering, but found myself sticking to draft formats because constructed games were not super thrilling. It seemed that all the best decks were just about setting up a game state where your opponent could just not do anything to change the course of the game anymore. Once such a deck is possible, the game becomes much more about good draws and favorable match-ups than it does about smart strategic gameplay, which just isn't fun. That isn't to say that there isn't a great deal of skill in side-boarding properly or taking good mulligans or whatever, but all of that stuff takes place before you've even tapped your first land, so it isn't really gameplay.

Anyway, I'm looking forward to this game whenever it comes out and I'm sure it'll be a blast. I also can't wait for the Yomi expansion. If you're planning on doing another kickstarter like you did for Puzzle Strike you can count on at least one purchase of both the complete first and second sets.

August 13, 2012 | Unregistered CommenterChristian

So I'm unclear on how you are finally deciding to handle this problem. Are you creating a rule-set for deck building that is based on composition requirements? Are you creating a skeleton deck and then players add cards as they choose? Is it a healthy mix and match of both?

August 13, 2012 | Unregistered Commenterskeis

Thanks everyone, I appreciate the supportive comments.

pkt: yeah that's definitely a fair question. The answer is that I know about the effectiveness of the "chunks" involved in making a deck. So the worst possible deck to build would be the one where the chunks have the least possible synergy with each other. But even then, I know that the player will have coherent stuff within that chunk. They will for sure have access to some heroes, they will for sure have access to some spells that make some sort of sense with that hero, they for sure have access to some 'creatures'. Furthermore I even know something about their 'mana curve.' I know that they have access to low cost things early and higher cost things later. And I know that if they don't have enough resources to pay for stuff later in the game, it was almost for sure because of their own decisions.

mas: Hmm interesting, but then each card would need to have a lot of stuff written on it. That's a wall I hit up against for a long time. I had several game systems that were along the lines of what you're saying, more stuff on cards. I had another branch of systems that was more like "somehow have more cards in the first place." It kind of reminds me of fighting games that want to have a whole lot of moves but want to have as few buttons as possible to be simple. But then you combine those two ideas and you get a lot of complexity in commands for moves, and overlapping commands. It appears simple to have only 3 or 4 buttons, but adding some more buttons makes it simpler in a different way, by making any given button have less stuff that it does. I guess I made what's analogous to a fighting game with more buttons than usual here.

August 13, 2012 | Registered CommenterSirlin

Sirlin is being kind of vague, but it sounds unintentional, I guess he can not approve this is if it's too specific...

I think a lot of the comments are mostly answered by saying that the limitation Sirlin is talking about for deckbuilding is basically that cards come in sets of, like, 20. You don't construct decks by choosing individual cards, you construct decks by choosing a set number of these pre-existing blocks and combining them.

So, he's able determine exactly how many decks exist, and how they would generally play, by just knowing how many chunks of cards can be combined to make a deck, and what each chunk of cards is designed to accomplish.

August 13, 2012 | Unregistered CommenterClaytus

Well, yes though you might argue a chunk is 13 cards or 25 or 6, or some other number depending on what we're counting and I didn't want to get into that. I really just wanted to mention the general idea of limiting deckbuilding in order to make waaaaaay more viable decks, and how it sounds at first glance like unlimited deckbuilding would do that, but it sort of ends up doing the opposite in every CCG out there. As for all the specific workings, I'd rather wait and go over them in more depth at a later time.

August 13, 2012 | Registered CommenterSirlin

I was just about to post a thread asking about the slowly upcoming SCG. I'm extremely excited! I just wanted to also add that I appreciate the degree of thought and dedication you put into your games.

August 13, 2012 | Unregistered CommenterSeth

I remember way back you were talking about having two separate game modes; one where you could just build whatever, and one where you would use preconstructed decks. I assume this new idea replaces the latter mode? Will the game still support a mode where you can just build whatever you want? Even if it leads to worse gameplay, I still think it's fun to explore the potential of such a format.

August 13, 2012 | Unregistered CommenterBen

Thanks Seth!

Ben: there's really only one mode now, everything has grown up around it and nothing else would make sense at this point. Well, there are other "modes" in the sense of 2v2 and also a free-for-all that actually works well. All those use the same method of making a deck though. After the first release of the game, there will be some extra stuff that allows even more kinds of customization without ruining everything. So if you really want to be able to turn up the customization knob even more (as if 800+ decks wasn't enough), I think you'll be satisfied.

August 13, 2012 | Registered CommenterSirlin

How do you plan on getting rid of mana screw? More importantly, how do you get rid of it without either introducing another resource that needs to be kept track of (think Yu-Gi-Oh and tribute summons), making the power curve completely flat (probably your ideal but almost impossible considering your other goals) or making cards better than one another (considering your claim of "815 playable decks", not obvious at this point in playtesting).

When I think of games without mana screw, I generally think of three games off the top of my head: Duelmasters/Kaijudo (which allows any card in your hand to be used as mana), X-men (which increased mana linearly, with little/no interaction from the players) and Dragonball Z (which had no mana system I can think of off the top of my head, though admittedly my experience with it is very limited). Based on what you've said so far, I'm assuming the model is closer to the Dragonball model, but when you choose a character you have to have some of that character's cards in your deck (10 with a 30 card deck?), making it closer to the Magi-Nation model (though I've played even less of that, I remember each character started with certain monsters). Dragonball Z was hurt by the third point (a non-flat power curve), probably because it needed to promote itself (it had a long restricted list, and 50-75 card lists were often a third restricted cards), which is a problem this game shouldn't be worrying about (since it's only one set, and will probably be sold in pieces like the Yomi decks--not to mention it'll probably be highlander regardless).

The biggest problem I'm seeing is that you're only catering to Spikes (using Mark Rosewater's Timmy/Johnny/Spike terminology). If the power curve is flat and all cards have to be playable, what can Timmy experience besides gaining incremental advantage? If all the decks/deck archetypes are known before the beta test is over (and published in a book like you did with Puzzle Strike), can Johnny express something new--and if he does, does that mean that the game has been broken since it went outside of the box? So far Yomi is probably the most-Spike game I've ever seen, Puzzle Strike is a haven for Johnny-Spike, and Flash Duel is a Spike-based game while having other modes for the other psychographics (Johnny likes the BOO character mode, and Timmy loves Deathstrike Dragon).

There's probably more concerns I can think of right now, but this post is rambling and wall-of-texty as is, so I'll stop now. This post isn't supposed taken in hate (and I bet most of the questions can't be answered because of the need to keep things secret), just as a post by a fan who's interested in CCG design.

August 13, 2012 | Unregistered CommenterCheater Hater

Regarding "Johnny," if you have a game where you can make extreme decks in some direction or another is possible, such that you can avoid interaction with the opponent and cause 8-2 or worse matchups, then naming "Johnny" after it and glorifying that problem would be a good PR way to go. Or if you just mean doing cool combos that are not really making the game unfair, then I don't see why we'd have any lack of those. Puzzle Strike has several, as you said. Puzzle Strike's gold chips, as well as a few others, are specifically there for "Timmy" as well.

I'm catering to people who don't want rares/random packs, who don't want a game changing constantly for the sake of change, and who want to have reasonably fair matchups that are lot more interesting than any given matchup in an existing CCG. It's really misleading to cram all that into a "spike" box that MTG created, and that is not accurate anyway. I could make up three types of players and say MTG only caters to one of those, so that MTG seems really limited in comparison. But doing so is more of a PR battle about how to frame things than any kind of real insight.

By the way, it isn't hard to know from playtesting that my statement about the 800+ decks is true. That even the weakest deck is at least as strong as the weakest character in the average fighting game. I'd be surprised if you could find even a single playtester who has played the game who would disagree. If we know how good various "chunks" are, the weakest deck is at least as good as the sum of each separate chunk. You're probably thrown off because that kind of statement in MTG is totally meaningless. There, the weakest deck literally cannot play the game (it's all land or something). The second weakest also can't play the game. I think even the ten thousandth from weakest can't play either. We're in a different ballpark here where I know that all 800+ have have at least the basic tools necessary to win, plus in-chunk synergies.

Or to put it another way, the weakest team in the Marvel vs. Capcom 3 fighting game would at least have three coherent characters on that team. Assume I have played all 50 of those characters, but not all 19,600 teams of characters. I'd have a pretty good idea about how bad the weakest team could even possibly be. Compare this to an MTG of fighting games where the "weakest character" is some incoherent mess of moves that don't work together at all (every single move can be customized in this example) and that have overlapping inputs so most of them literally can't be triggered. That might give some intuition about how one could know anything about the minimum strength of an MvC3 team compared to a random pile of cards.

I don't think the comment section is a good way to learn more about the game though. My intent was to share one thing learned from design work, not explain the whole game. A better way to learn about it is to wait for a set of articles that explicitly lays out the various systems and how they work. I should really focus on Yomi now though. Just know the other game is coming too.

August 13, 2012 | Registered CommenterSirlin

Actually, I think it's pretty accurate to say that your games cater primarily to "Spike." Most of the design decisions in your games are made under the assumption that all players are playing with the primary intention of winning, and will (generally) take whatever option gives them the best chance of doing so. From the perspective of a serious tournament player, it seems almost absurd that somebody could prioritize something other than winning in a competitive game. But a lot of Magic players will knowingly choose weaker cards/decks that they like over stronger ones that don't appeal to them as much.

The "Johnny" player archetype is one who prioritizes doing something original and "different." He'd rather play a B-tier deck of his own design than the A-tier deck that everybody else is playing. When you reduce the degrees of freedom with which he can build his deck, you also necessarily reduce the degrees of freedom by which he can separate himself from other players.

I agree that "extreme decks that avoid interaction with the opponent and lead to 8-2 matchups" are terrible for the game, but I think that equating "Johnny" to "degenerate combo decks" is unfair. For one thing, most of the crazy original decks that Johnny wants to build are actually below A-tier, not above it. Also, if the deck really is that dominant, then all of the "Spikes" will start playing it, which will make Johnny *less* interested in playing it himself.

Now, the place where I'll diverge from Cheater Hater's argument is that I actually *appreciate* that you're focusing primarily on "Spike." My experience is that competitive players make up a small enough percentage of many games' audiences that developers tend to include features to attract and retain casual players. Unfortunately, these "features" include mana screw (which helps weaker players occasionally beat stronger players), leveling up via grinding (which allows players to feel like they're getting better at the game without having to actually improve), and intentionally imbalanced characters (which lets players feel "smart" for picking the "right" one). I'm extremely happy that there's a place I can play games without having to put up with any of these Spike-unfriendly design decisions, and I hope that you continue pushing forward in this vein.

I'm particularly interested in seeing where you go with SCG4. As a longtime Magic player who loves the game but hates the business model, I'm excited at the possibility of a competitively balanced CCG with a buy-once model. I hope to hear more about it as it gets further along, and I'm looking forward to it being (eventually) available on

I'm also looking forward to getting my hands on the Yomi expansion. Good luck with that!

August 13, 2012 | Unregistered Commenterawall

I was getting this sneaking suspicion that you were using the same basic solution to some of these problems as I am using in a card game I'm working on, then I read stuff about 'heroes' and 'creatures' and whatnot and was like, "phew, now I can get excited about SCG4 again!"

Also, random request, maybe reasonable maybe not: Can we have a codename or something to refer to this by? MtG codenames their sets (like one of their blocks was "Snap, Crackle, Pop") and sometimes that ends up being kind of a dumb name, but it's always easier than "Spring 2013 small set" or whatever. SCG4 doesn't really roll of the tongue (although maybe that's just me).

August 13, 2012 | Unregistered Commenterajfirecracker

Interesting article! I am wondering how "specific" you intend to get with the deckmaking rules.

I am thinking of Spectromancer ( which is vaguely like MtG with all the "deckbuilding" done according to a very complex randomization process. The structure of the deckbuilding process lets you learn things about specific cards your opponent might have. For example, it lets you say things like "my opponent just played Wall of Reflection so I know Nature Ritual is definitely not in his deck".

Are you envisioning the constraints on deckbuilding to be precise enough that players are thinking things like this during play? Or just in more general terms like "oh, I've identified all of the chunks in his deck so I see the synergies which he will be working with despite not knowing his specific cards" ?

August 13, 2012 | Unregistered CommenterDrTall

You've come to a similar conclusion as Guild Wars 2: Restricting choice in deck-building allows for more non-degenerate gameplay. I think this was the underlying idea of "talent trees" in games, yet most designers have no idea, they just blindly copy it, and end up with completely broken and non-sensible arrangements for skills.

Time for some reverse engineering! If you say there are 815 decks, and we assume 600 cards total, at 60 cards per deck, and you need to select three chunks (because you said "33% difference at least"), that gives us 30 chunks of 20 cards. I postulate that there are 10 / 10 / 9 chunks (which share the 500 cards, that's the briliance), that's a total selection of 810 if you pick one of each. 815 is a weird number, as it does not really have enough divisors (only the two primes 5 and 163) that make sense.

Totally want to play it.

August 14, 2012 | Unregistered CommenterKdansky

awall, I do not think mana screw is a feature, really. I know some guys will defend it as such (eg Rosewater), but really I think that's just apologetics. Every player hates not being able to play or do things, not just experts. For MTG in particular it's just too ingrained in the system to eliminate, though they print many cards to help deal with it.

August 14, 2012 | Unregistered CommenterLofobal
Comment in the forums
You can post about this article at