What, When, Why, How: Set Design - Case Study of MonsterCon Starter Box
What is Set Design?
“Set Design” is a term that is heavily associated with Magic: The Gathering’s design pipeline of Exploratory Design->Vision Design->Set Design->Play Design. There are plenty of Mark Rosewater’s Making Magic articles or Drive to Work podcasts if you want to deep-dive into Magic’s implementation, but the short of it is this: Set design is the process of filling out the broad strokes of the card pool for a card game or its expansions (for the sake of brevity, I’ll be shortening further mentions of “game or expansion” to simply “game”). This usually entails figuring out the overall balance of different factions, card types, kinds of effects, and, of course, rarity distribution. An expansion’s set design, and the best practices for it, are highly specific to the game and the intention of the product release. While Magic has extensive documentation on why and how they do the things they do, I think it’s very important to understand that those rules specifically work for Magic. In this article, I want to help people learn how to adapt and utilize the practice of “set design” for any (card) game they could be designing.
When is Set Design?
Set design, as I will be elaborating on here, is greatly influenced by Magic’s design pipeline. It occurs after a theme for the cardpool is relatively established, but before large swathes of card designs need to be locked in and fine-tuned so manufacturing can begin. Like many design steps, this doesn’t necessarily have to be mutually exclusive with other processes. You can begin set design from the very beginning and you can make set design changes if balancing is proving to need major adjustments. You can, of course, add new themes while establishing set design and so on and so forth.
Why is Set Design?
Understanding set design is incredibly important for figuring out what your cardpool should look like. A lot of the practices I will be explaining will also help you look more objectively at your design decisions and can segue into a deeper understanding… All of which are incredibly helpful qualities that will aid you once you get to fine tuning the balance of individual cards.
In a sense, set design is the middle ground of purely aesthetics-driven choices and purely-numbers driven choices. Balance is key to many best practices. Focusing entirely on one aspect can lead to burn out. However, completely switching from extremes can have consequences such as incompatible practices being very jarring. Set design can be a welcome segue and reprieve from these extremes. If you understand how your themes translates into an overall skeleton for your card pool, you can use that overall skeleton to figure out what can be changed if something feels off. Even if you don’t adopt the practices I’ll be outlining here, my hope is that awareness of them can bring greater intentionality to your workflow.
How is Set Design?
With that lengthy preamble out of the way, we can finally begin tackling set design while wresting control from the context of Magic. For the sake of example, I will be using my own MonsterCon Starter Box product as a case study. However, I will be mentioning how some of these concepts can be stretched and molded to fit the needs of different games and products. If you’d like to see the card gallery that makes up the MonsterCon Starter Box as a specific reference, check out https://www.blobspike.com/monstercon-card-gallery.
Goals and Scope
First and foremost, I consider knowing the goals and scope of a project to be an absolute necessity to even start with. And having a clear vision of those things is also necessary for set design that is both successful and direct. In a hobbyist space, it’s fine to not have a crystal clear vision. The road to success is still possible, it just might need a lot of adjustment down the road as your product becomes more actualized and you begin to grasp what you’re capable of.
To start with, the MonsterCon Starter Box is a product that contains five preconstructed starter decks as well as a package of cards for players to upgrade these decks, diversify strategies, and create entirely new decks that aren’t immediately obvious. One of my goals was to keep the unique card count very tight while maintaining a huge amount of deck diversity. In terms of scope, this is where budget and timeframe is concerned. I can afford to have ~100 artworks made for cards and that’s also about the number of unique cards that I’m more than capable of juggling around for balancing purposes.
To be explicitly clear, scope is something I don’t expect fledgling designers to fully grasp until they’re partway through developing their game. My original goal was a simple set of preconstructed decks, though it evolved because I wasn’t excited to only develop a starter deck environment; A deck construction game with hardly any deck construction possibility is probably not where I want to land. However, I knew there was no way I could afford to develop starter decks alongside a full-sized expansion. So an executive decision was made: The five starter decks would effectively be half of a full-sized expansion and half the cards of a planned expansion would be grafted onto them to create a full set of cards.
You can obviously have different goals and scopes than what I’ve outlined. For example, if you wanted a TCG booster pack opening experience, you’d need to vastly expand your budget and unique card count to improve the odds of novel/exciting pack opening. If you want a draft/sealed experience, you’ll have to weigh your options in terms of novelty and consistency. Regardless of your project, being able to qualify and articulate exactly what you’re designing for will be extremely helpful even outside of set design.
Setting Up the Set Design
Once you’ve got an idea of what your game is going to have, it’s time to get a bit mathy. An important aspect of set design is laying out the broad strokes of what you need to make your game in a practical way at a level of quality you’re satisfied with. For a card game, this usually means laying out how many cards you need to make and what kinds of cards will fill those slots. This is something that will be highly dependent on how exactly your game is structured. The two common factors that will really drive a lot of your design are how factions are implemented and how the game’s deckbuilding works.
MonsterCon does not necessarily have colors or factions in the way most card games would have. While this attribute is usually something that creates deckbuilding restrictions, even if they are soft restrictions, MonsterCon allows any card to go into any deck. However, MonsterCon does use the concept of colors as a design tool, though the main terminology I utilize when actually designing sets are “themes”. The main point I’m trying to bring up though is: For each MonsterCon expansion, I am planning to have five “main themes” in a similar fashion to how Magic has five colors, how Digimon has six colors, or how Railgrind!! has four styles. To maintain an even representation of different strategies and aesthetics, it’s good practice to create an even amount of cards for each faction/color/theme.
In terms of deckbuilding rules, MonsterCon requires a 40 card Main Deck, 1 Leader card, and 1 Domain card. The maximum playset size is 3. That means a deck that maxes out the copies of every card will have 14 unique Main Deck cards (13 playsets and 1 straggler one-of) and 2 unique non-Main Deck cards for a total of 16 unique cards for each deck at minimum.
With that in mind, we can start to quantify how many cards we need to design (at minimum). 5 themes with 16 unique cards each means a total of 80 unique cards. However, you might recall that (1) this product includes preconstructed starter decks, so more immediate novelty and unique cards per deck could only be a plus and (2) I want there to be a huge amount of deck diversity, not just the absolute bare minimum.
First, adding a few more unique cards to the starter decks and making a few cards not max out on playset size will introduce a lot more variety even if players do not engage with deckbuilding. I ended up with about 4 more unique cards per deck. Next, this is something that is more of a “tell by the vibes” thing, especially if you have different deckbuilding rules, but having somewhere between 5-6 viable alternative choices to place into your deck is the minimum I need to feel like there are options. Last, I think having an alternate Leader and Domain per theme is a must for a deckbuilding leader game.
As a result, we’ve gone from 16 unique cards per theme to 28 unique cards per theme. This means a total of 140 unique cards, with 20 of them being Leader or Domain cards. From here, we can finally start what people might finally begin to think of as “game design”!
Designing the Set Design
The working knowledge of exactly what your game needs is similar to a lump of raw material. However, we still need to refine and chisel out the material so it can be decorated on by individual card designs and balance passes. This is something that requires a fairly advanced understanding of your game to do effectively. And, in an almost contradictory manner, you actually can’t put the finishing touches on a set design until well after the set design stage. Let me explain.
I’ve written in length about “figuring out what your game needs and aligning that with what you want” in the previous sections, but you can’t possibly know what the best iteration of the game will be until you test, test, test. However, you can still make well informed design decisions that will give direction to future stages of design and minimize outliers (or maximize them, if you so choose). This is where the bulk of set design work actually is.
For context, MonsterCon is a creature battler with a mana resource system that allows players to gain about 1 additional maximum mana each turn. Different games will have different properties, such as a different resource system or not even have one at all! Whatever the case, it’s important to understand what an average deck is made up of, what you want to incentivize players to utilize, and know what cards each major demographic of player will pay attention to. This is quite a lot to suddenly drop on you and each certainly deserves its own write-up and deep dive, which is why I opted to instead graft a case study of MonsterCon to oil up the gears.
1. What is an average deck of MonsterCon made up of?
There is usually an average of 25-30 total Monsters and the rest are made up of Spells of various kinds, though around 5 are Fast Spells. The mana curve of the cards generally follows a classic bell curve, though it tends to swell at a lower number than most other games with a mana resource. However, the higher level answer to this is: Decks can broadly either be very Main Deck focused with Leader/Domain choices to enable those strategies or vice versa.
With this in mind, I now have a pretty good amount of data. For each theme, I can leverage a little over half of the card designs for Monsters and leave the rest for non-Monsters. Additionally, I now have a lot more context for what cards I can design for each Mana Cost.
The critical thinkers might recognize something though. “Isn’t this something that I can specifically encourage through system and card designs?” Yes. In fact, it’s something a card game designer has to be very aware of. You can have a platonic ideal of how a deck should be constructed but doesn’t necessarily mean that players will conform to it. You’ll have to figure out whether you want players to exactly conform, whether you want there to be a standard with some deviation, or whether it’s a straight-up free-for-all. Regardless, this should give you a vague idea of how much representation each kind of card should have.
This is incredibly important if you wish to design effectively. For example, your set design has 80% monsters, but decks are made up of only 10% monsters. The remaining 20% of your set that is not Monsters will see dramatic overrepresentation and crunch design space and variety. However, when these ratios generally match, you’ll that find players will often be making deckbuilding choices based on trade-offs because there’s simply too many choices for each bracket of cards.
In a MonsterCon example, about 10% of the unique cards in the starter box are universal or conditional removal. This generally lines up with many decks having 10-15% cards that act as removal. You’ll notice that set sizes are much larger than deck sizes (100+ unique cards vs. 14 unique cards). As a result, deckbuilding sees a much larger variety. If a player sticks with an expected ratio, there’s a good chance there will be novelty in which removal they choose to keep. If they want more removal, it’s a fairly viable option. And some players may opt to not include some at all!
2. What do I want to incentivize players to utilize?
A core design philosophy I have for MonsterCon is wanting players to mainly focus on Monster combat and care about field presence.
While the core conceit of the previous question brought light unto the amount of representation different kinds of cards can have, this question should elucidate what they should actually do and what they should be designed towards.
Since I want Monster combat to matter, I want more representation of card designs that satisfy that. Maybe when I’m coming up with Leaders and Domains, I design their effects to support an ecosystem that incentivizes Monsters trying to remain on the field. However, if I wanted a different result, I would change things up. For example, if I wanted there to be less Monsters in decks, I would simply produce less Monsters and I’d make each Monster less viable. Hopefully, this illustrates what tools you have in your arsenal to actually begin to shape your game’s set (design).
Depending on your game and your goals, it might not be quite so simple. You’ll have to figure out the balance of what your game needs with what your players expect. For example, you know that your game could use a broader spectrum of the mana curve for the sake of longevity but players only want to play the cheapest, most efficient cards. One possible way to tackle that in set design is simply design less low-cost cards. Or maybe you recognize that longevity isn’t actually a practical goal and make decisions to maximize player enjoyment in the moment.
3. What cards does each major demographic of player pay attention to?
This question is actually too multifaceted to have a succinct answer. The classic player psychographics are Timmy, Johnny, and Spike. Each one of these psychographics care about different cards a different amount. You can’t please all of them all at once with every card, and they certainly don’t actually want that. While it might seem like a platonic ideal to release 5,000 new cards that are equally and highly engaging to all people every week, it’s practically impossible and, also, human beings can’t actually process that without extreme fatigue or bouncing off completely. However, diversifying the appeal of your cards is still a great practice. Not only does it encourage curiosity, it allows you to be more specific and intentional with your designs.
This means that when I’m looking at each color/theme, I want to make sure each demographic is catered to in some capacity. People looking for splashy and exciting cards (Timmy) should be able to find one in each color. In fact, once you get in the weeds of it, you’ll learn how to design those splashy cards for each color’s Timmy. For example, one Timmy might like explosive burst DMG whereas another Timmy likes a massive cascade of Spells.
While the classic psychographics are the most well-known in TCG spaces, there are several others such as age groups, fans of certain genres, or even geographical demographics. This starts to get more into market research and personal goals, so I’ll digress. Only you (and probably your design team if you have one) can know what demographics you want to engage.
With all of that said, each one of the answers to these questions will give you incredible insights in how you should structure your sets and figure out what designs you should put in. Should you have universal spot removal? If so, at what power level? Should there be powerful, game-ending mega threat monsters? How early should they come out? Are there alternative game modes you could or should incentivize? Can you fit that in with the standard gameplay you’ve already designed?
I want to briefly talk about Magic again, if only to tell you when you should take more or less inspiration with how they do things. Their expansions are mostly designed for limited gameplay as well as commander. Meaning they have plenty of “pack chaff” and “commander cards” sprinkled within their booster sets. If your game does not have draft gameplay (with the exact gameplay systems they have) or an alternative format like commander, you should be extremely selective with what you take from them. They have a breadth of cards in each color to satisfy (1) guaranteeing you see an average number of cards for each color and mana value, (2) incentivize a certain play style in their limited gameplay, (3) seed in certain cards for constructed formats that will have monetary value, and (4) have a budget, including a team of professionals working day-in, day-out, that you cannot even fathom. Most designers, especially hobbyists, probably should not use their model exactly. Instead, adapt certain aspects. For example, having a specialized form of removal for each one of your factions while also tuning how efficient it is depending on whether that effect is in your game’s “color pie”.
A Set Ending to Design
To summarize the major points of set design:
Know how much representation each kind of card receives from your players.
Adjust the representation of each kind of card to meet your desired gameplay.
Pay attention to whether your decisions are leading to designs that appeal to your desired demographics.
To end things off, let me list off the working model of a set design that we’ve made for MonsterCon in this article: We’ve set up MonsterCon to have 28 unique cards per theme, with 2 being Leaders and 2 being Domains. Out of the 24 Main Deck cards, we want a little over half, about 14, to be Monsters. Which means 10 should be the various kinds of Spell types depending on the theme’s needs. The Mana Costs of each card, and, by extension, each card’s impact, should generally fit a standard bell curve distribution. To better mold each theme’s identity, we can adjust what we expect in each variable. If a theme calls for a larger focus on Spells, the card representation around that theme would lower the number of Monsters in that theme and create more Spells. If a theme is about having high Mana Cost pay-offs, the bell curve distribution of the Mana Costs in that theme would lean more towards the higher costs.
At this point, I’ve probably reached the practical end of what I can write about set design as it applies generally. From here on, each game will have set design practices that are extremely specific to their gameplay. MonsterCon requires a great deal of focus on Leaders and Domains. Magic requires a fair bit of focus on lands, color fixing, and ramping. Yu-Gi-Oh requires great attention to combo setups, enablers, and pay-offs. Your Game… Well, I can’t even begin to imagine what unique conditions you’ll have to address.
If you’ve read this far, I hope that this has illuminated the benefits of this workflow as well as put to words what you might have subconsciously been doing this whole time; Maybe this could serve as inspiration for more accurate and intentioned design choices in your own project!
BONUS ROUND! Min-Maxing Your Set Design
MonsterCon fans might have noticed this but I’ll mention it anyway. The total unique card count of the starter box isn’t 140. It’s actually 102! But why and how did I get to this point?
The number one main reason is simple: Budget and scope. Having to work within practical means is something all projects will have to manage with (outside of obvious exceptions such as pet projects of the richest humans in the world).
Regardless, how was I able to trim my calculated 140 cards to 102 while not giving up on my goals and still being very satisfied with the quality of the product? This answer is also simple, though not in implementation: Doing more with less. This is something only possible with experience and intentionality. It’s not something I can advise on generally, only something I can speak with authority specifically only in MonsterCon. With that in mind, I hope this brief case study could give you inspiration on how to apply min-maxing to your own set design.
1. Embrace generic/neutral cards.
Part of making sure the starter decks were balanced included having a set of basic cards and effects that players would see no matter which deck they chose to play. Additionally, there are simply some effects that I think each theme should have access to without needing some kind of specialized twist. As a result, I moved around a couple of cards. Of the 20 Main Deck cards you’ll see in a starter deck, 9 of them are generic. Suddenly, rather than needing to make 100 unique Main Deck cards, I’ve cut it down to (11*5 + 9), or 64 unique cards. With 1 Leader and 1 Domain for each theme, that brings our total to 74 unique cards.
Of course, there are a few “generic” cards designed for the package of extra cards. These few cards were designed to fill some gaps in the Mana Curve. I ended up making 3 other generic cards for a current total of 77 cards.
2. Hybridize some cards.
This is a similar practice to making generic cards. However, rather than designing the card to fit any and all themes, this is more about making cards that shine best in two or three themes. There’s many ways of doing this. For example, I’ve designed one Monster for each theme that suits the needs of one other theme. As a result, five cards can fulfill the role of ten cards, lowering our grand total to 72 unique cards.
I want to again stress that hybridizing cards can be applied in many different ways. For example, although there are five themes in the MonsterCon Starter Box, I can design a few cards within the theme to care about the theme in a slightly different way… Or designing towards a sub-theme if you will. And if I also design each sub-theme to work with one other sub-theme to create a cohesive deck, then I have effectively tripled deck diversity without needing to add any unique cards! Of course, the practical application of this isn’t so hyperbolically simple. However, this method of min-maxing card versatility without entirely opening itself to goodstuff piles does introduce a great deal of deckbuilding decisions and diversity.
Before going any further, let’s find the missing 30 cards in the set design: One alternative Leader and Domain for each theme makes up 10 of that. Additionally, 4 main deck cards for each theme makes up 20 meaning there’s now a total of 102 cards. A great deal of them utilize “hybridization” to fulfill the “5-6 viable alternative cards in addition to the minimum necessary” criteria I set up earlier, even for sub-theme decks. However, we can still go much further.
3. Introduce head-turners.
At this point, the MonsterCon Starter Box has more than fulfilled the basic goals I outlined earlier all within the scope of the production budget I’ve set forth. However, there are still a few parts of the set design that could use a little bit more spice. This is where I want to bring up the concept of “head-turners”; A few select cards that specifically recontextualize other cards.
Generic and hybrid cards exist to slot into multiple themes seamlessly, sort of like how liquid can move around to fill an ice tray. However, “head-turners” as a concept exist to not fill the ice tray. Instead, they aim to key players into realizing “hey, what if I actually freeze the ice tray to create ice cubes?” or even further beyond “what if I processed the ice into shavings for a dessert or used it in a drink?” Silly analogy aside, the benefits can be laid out as such. Generics and hybrids state that rather than have 10 cards that support 1 theme, you could have 10 cards that support 2 themes. Head-turners say that rather than needing to adjust those cards further and maybe add new cards, one new card OR changing one card could create a third theme.
As far as I can tell, there’s no scientific formula on how to create a head-turner, but I don’t think that should stop you from figuring out where you can place one and seeking one out. One consistent factor I’ve noticed in my more successful head-turners is they flip some core rules of a theme on their head with an exciting pay-off. My best examples are AIN-065 Leonidas, Hero Among Kinfolk and AIN-084 Shamble Mountains.
Field presence and getting Monsters onto the board is hugely important to MonsterCon’s general gameplay loop. The main theme that wants to swarm the field with Monsters does so while also trying to protect them. These two cards flip the script: Leonidas wants to use them to fuel aggression whereas Shamble Mountains wants to use them as sacrifice fodder.
Similarly with hybrid cards, I don’t think that the presence of these two cards will actually create two entirely new themes off their backs alone. But they create a great deal of variety in how players can approach deckbuilding and novelty in what they can expect to face, especially for how little space they take up in the set design!
4. Planting seeds.
Lastly, I want to talk about the concept of “planting seeds”. You may have heard or read this phrase as “seeding in strategies/themes/concepts” from other designers. In short: this is the concept of introducing a theme, playstyle, mechanics, archetypes, etc. that is more or less undersupported. But why would you do this?
The main reason is to create intrigue. Other reasons include simply having enough space without being detrimental to the already existing set design, wanting to spread out cards across different expansions to entice customers to buy multiple expansions, and testing out the waters for customer feedback. Here are a few seeds I’ve planted. They work well enough within the context of the set that they aren’t an outlier, but they’re seeds nonetheless. I’ll leave you all to guess what exactly I have planned for them and the strategy that they’re intended to be part of.
In essence, this is a similar concept to head-turners, but they have a much less impact. To be cute, they might even be considered “head-scratchers” instead. But why bring up this concept specifically when so much of this article is about tight design that wants to min-max card design slots? I don’t want this to come off as too contradictory, but I think there has to be some whimsy and silliness, even when deep in the design think tank. While these don’t produce the huge boom of variety and diversity that generics, hybrids, and head-turners do, the seeds of future potential are a good reminder for any set designer to relax. You don’t have to make sure every theme, deck, or faction is completely maxed out on its potential from the get-go. It’s fine for some strategies to only be partly realized. It’s fine if the final set design actually didn’t have enough cards to support that theme you wanted.
I saved this for last because I know as designers and gamers, our instinct is to create the perfectly ultimate great moth set design that covers every base completely. However, it’s not ultimately practical and maybe not actually the best choice. There’s a great joy to be found, for both the designer and gamer to be clear, in fans of a specific deck receiving support in a later expansion. There’s a great joy to be found in appreciating the negative space of what isn’t there so those gaps can be filled in a personal way. Make more with less, but also know that less sometimes says more.
With that, I’m really truly done with what I have to say on this topic… At least when speaking broadly. Each game has its own set design circumstances that requires its own deep-dive and analysis. While I’m not going to personally analyze and do a write-up for every game and every expansion, I hope that this article at least provided tools and examples of how to use those tools for your own set design.
Ciao,
blobspike