3/21/14 Episode 107: Design Skeletons

All podcast content by Mark Rosewater

Okay, I’m pulling out of my driveway! We all know what that means! It’s time for another Drive to Work.

Okay. So today, I’m going to be talking about something called a “design skeleton.” So what a design skeleton is is something that is from the technical side of the job. I talk a lot about making Magic and I know that there’s—I tend to focus a lot on the intuitive, inspirational, “Oh, I had a dream, and oh my God, bam, I had an idea!”

And I talk a lot about, there’s a lot of thought in the creativity that there’s these like moments of pure clarity where you realize things, and there’s some of that. And I play that up. And there’s a fun thought of creativeness as having this wild sort of right-brain-side sort of thing. But in reality, a lot of the creative process is hard work and structure. Is spending a lot of time carefully crafting and figuring something out.

And today I’m going to talk about a tool that we use to do that. I first started talking about design skeletons in a series I have called “Nuts and Bolts” that I wrote in my column. Where I do one a year sort of talking about how to make your own set. And in this column I introduced the concept of a design skeleton. So I’m going to talk about that today.

So, my metaphor—I use some metaphors today. I never metaphor I didn’t like. So let’s start with the metaphor of building a building. So let’s say you’re an architect. Let’s say it’s your job to figure out, to get the vision for what your building looks like.

Well, you make something called blueprints. Or architectural plans. And what you’re doing is, you’re saying, “Hmm. Here’s my first thoughts and what I think the building’s going to look like.” Now, it is quite possible that when you go to start building, or when the builder starts, building, the architect might have had ideas that while were great on paper, might not actually work due to different limitations.

And so the blueprints are guidelines. And that it’s something that when you want to make a set, you have to figure out the core of what you are doing. So to understand this, let me explain a basic tenet, something that we use to make sets. Something we call a card file.

Now, the idea of a card file is, it is a list of—let’s say I’m making a large set. Large sets currently have 101 commons, 80 uncommons, 53 rares, and 15 mythics. 269 cards. We recently upped from 60 uncommons to 80 uncommons. The reason for this was, we were getting pinched at uncommon, that there was too much we needed to do and not enough space, and by opening up a little bit, uncommon’s really the place we can put build-around-mes for draft that really sort of helps make draft have more variety over time. And anyway, we felt like we needed the extra space. So we added twenty cards.

Anyway, so when you’re making a set, you have a file. But the thing about the file is, you are going to change the cards in the file. That you’re going to make cards, sometimes cards—all sorts of things could change about the card. The card could change, its rules text. Its mana cost. Its power. Its toughness. Its creature type. Lots of things can change.

And in fact, the whole card could be tossed out and replaced with a completely different card. So what we do is, our card file has slots. Meaning that you have 269 slots. 101 common, 80 uncommon, 53 rare, and 15 mythic. You have slots.

And what we do is, we designate each slot. So we have what we call “card codes,” that are used to designate. Now, a card code has three components to it. The first component is rarity, so if it’s a common card it gets a C, uncommon card gets a U, rare card gets an R, mythic rare gets an M. It’s pretty easy.

The second signifier originally was about the frame, it is partly about the frame and partly about the color. When we first started making card codes, they informed the people producing the cards what the frames were. That’s drifted a little bit over time. But as a general sense it talks about the frame.

So if you are a colored card, W for white, blue is U because B is black and L is land, so U is blue. Black is B, red is R, green is G. Artifacts are A, lands are L, multicolor cards are Z. Why Z? I don’t know. We just needed something, a signifier, and we wanted to pick something we felt was just out of the way, wouldn’t be confused for anything. And no one was using Z, so we picked Z.

And the way the card file normally goes is you have white/blue/black/red/green, WUBRG, that’s the first letters in order, pronounced. Or U for blue. And then you have your multicolor cards, and then you have your artifacts, and then you have your lands. That’s the order that we put them in the file. So the idea is, you have a file. So what a design skeleton is, is trying to structure the file so that when you approach it, you have some idea of what you need to do.

Okay.  So the very first thing you do when you’re doing your design skeleton is literally make a slot for every card in your file. In order to do that, first thing you have to do is figure out some basics about your set. So for example, I have 101 commons. So let me explain the 101^st common. That always throws people.

So the way—when we print Magic cards, essentially we have a sheet. We print a giant sheet. That’s the size that the printer prints on. And then what we do is we cut them down to card size. So we print a whole bunch of cards and then cut them down.

And so in order to sort of have the right combinations that we need, there’s just certain set sizes that fit with the card sizes. The sheet sizes. And so that’s how we get different set sizes is, in order to do the thing we need to do—it’s very complicated. I learned long ago, in eighteen years I’ve avoided every possible meeting where I had to talk about templating or collation. And I worked very hard to avoid those meetings, they are very important and I’m glad there’s people that do them, but they get really mind-numbing.

But it is—it’s funny, by the way, when I was in school, a long time ago, in math class, my teacher said,
“Oh, this is important, learn your math, you will need this later in life,” and I was sure I was going to be a writer. I said, “I’m going to make my living with words. I’m not going to need numbers.” And I took my math classes, but I’m like… I was super doubtful I was going to need it. And now, now that I’m a game designer, holy moly I use a lot of math.

Like, collation is all math. Like, as-fan and all this stuff we have to do to figure out percentages, and it’s a lot of math. Luckily, I don’t have to do collation full-time, somebody else is doing collation. I’m aware enough of it so I can do my job, which is figuring out set sizes and things.

Anyway, so you have 101 commons. The 101^st, by the way, is actually kind of not a common. It’s kind of not an uncommon. It actually falls in between rarity-wise, that it is half as rare as a common but twice as common as an uncommon. We call it a common, just because it has to be called something. But anyway, the 101^st card shows up a little less.

But anyway, the 101^st card is always an artifact or a land. And then what we do is, at common and uncommon, we always completely balance color. And then at rare and mythic we get close but it’s not exact.

Why do we do that? Well, we’ve had some experience not balancing color, in Torment and Judgment, for example, Torment skewed toward black, Judgment skewed toward white/green, also in Innistrad I skewed the set a little—not as much as Torment, but a little bit toward black because it was so dark. Development ended up undoing that work. Why? Because it really, really skews draft. Not having a balanced color really does weird things to draft and makes draft less fun. And so, at common and uncommon, we balance them completely. Meaning there’s an equal number of black cards, as blue cards, as green cards, as white cards, and such.

So normally what happens is you have to figure out at common, are you doing any land cycles? Because you have to do them in cycles. Are you doing any artifacts? If you do artifacts, other than the 101^st one, they have to come in fives.

So let’s say for example you go, “Okay, I want a cycle of common lands. It fits my theme. I think I want common artifacts, so we’ll say five artifacts plus the sixth, we’ll make the 101^st card an artifact.” Okay. So now, I have eleven cards represented, that means I have ninety cards not represented, I need to do an even split, that’s eighteen cards per color.

So what I do is, I go to my file, so remember. Common is C. White is W. 01—oh, I didn’t mention. The third signifier is number. So it’s rarity, color/frame, number. So the first in the file is CW01, all the way through CW18. That means there’s eighteen slots for white. Then you have CU01 through CU18. You do that, then you have your six artifact slots, your five land slots, now you have your 101 cards.

So that’s the very first thing you do. You literally make slots. Then what you do is you figure out whether or not each slot is a creature or a spell. So the way you figure this out is pretty easy, what we have done is we have figured out how many creatures each color is supposed to have, and made a default.

So right now, white gets the most creatures. Why white? Why not green? So green used to get the most creatures. Green is thought of as the creature color. And for a long time, green got both the most creatures and the biggest creatures.

But some time we realized we wanted to spread out the creature love, and so what we did is we said, okay. White is the color all about the group coming together, the community, the army, like it wants a lot of little creatures. And so we’ll make white the color that has the most creatures, we’ll make green the color that has the biggest creatures. And that way we could differentiate them. White and green already have this problem of overlapping a lot, and so giving a little more spread between them, we thought was good.

Now, the numbers I’m giving you are default numbers, they change from set to set. This is kind of where we start. So white, by default, has 55% creatures. So if you look at the file, that eighteen, 50% is nine, that means, eh, ten, eleven cards we want to be creatures.

Oh, and this is important. Development has a very distinct thing of how they count creatures. They care about creatures from an aggressive standpoint. So anything that makes creatures, for example if you have an instant in white that makes two 1/1 tokens, that is counted as a creature. Might be an instant, or sometimes it’s a sorcery, but the fact that it makes creatures, it functionally is a creature. From a vantage point of counting it as a creature.

Meanwhile, defenders, cards that have no ability to attack, don’t count against creatures. And so there’s a little bit of, we say 55% creatures, we’re talking about things that matter aggressively as creatures. Now, we do count things that have a tap activated ability, or even things that have zero power if they’re capable of attacking. Because the game has ways to pump those things up.

So anyway, white’s default is 55%, and I stress the word default, that as each set rolls along, we match it with what we need to do. And so it might have a little more, it might have a little less, although the cards will always go in the following order. So white is number one at creatures, as far as number of creatures, green is number two, it has about 52%, black is number three at about 50%, red is number four at about 48%, and blue is number five at about 45%. So it goes in order, white, green, black, red, blue. Blue is the most spell-oriented, red is second most spell-oriented, black’s in the middle, and then white and green lean toward the creature side.

And like I said, even though we’ll change them up, usually they’ll stay in that order. Even when like sets are a little heavier on creatures or a little lighter on creatures. They still tend to stay in that order.

Okay, so once you do that, the next step is figuring out what size creature you have. And so what do at first is, we break creatures into three general sizes. A small creature is any creature whose power and toughness adds up to four or less. A [medium] creature is any creature whose power adds up from five to eight, and a large creature is any creature whose power and toughness add up to nine or more.

And so now you go through the file, and you figure out for each of your creatures, whether they’re small, medium, or large. Once again, we allocate at certain rarities—like for example, white tends to have mostly small. A little bit of medium. Traditionally, it doesn’t get a large at common. Every once in a blue moon it does, but as a default it does not.

Where on the flip side, green always has a large creature at common. Every once in a while it will have two large creatures at common. Different creatures we allocate.

So let me (???) a little aside here. I spend a lot of time and energy talking on this podcast about how I try to make each set different. I talk about pushing the pendulum, I talk about how we keep wanting to make each set be this new thing to explore. But the thing I don’t talk much about, which I will today, is very, very much what the design skeleton is about, is how we try to make every set the same.

Because one of the things that’s important to understand is, I talk a lot about how Magic is many games and to many people there’s many ways to play it. But the flipside of that is, at its core, Magic is one game. It has one running set of how it works.

And that when you sit down to play a game of Magic, we want to make sure that it is a game of Magic. It is possible to push it too far. You could come and say, “Okay, this set, we’re not going to have creatures! And we’re going to… all damage is going to be permanent.” And we could constantly change things and just change the fundamentals of how it’s going to work.

But the thing though, is our job is to make things comfortable. So one of the things, like in Hollywood. If you’re trying to pitch a new product in Hollywood, one of the things that’s important is you always want to pitch your product as a combination of two known quantities that will come together to form an unknown quantity.

So the idea essentially is, like let’s say you have this awesome idea for a movie about a guy who gets taken by aliens and has to fight his way out. Okay. So you want to go pitch this to Hollywood. But the problem is that—in a vacuum, I’m like, “Oh, that’s too weird. I don't know if we’re comfortable with that.”

But when you say is, “Okay, I have an awesome film, it is Die Hard meets Independence Day.” What you’ve done is said, oh, well, the trappings of the story, it’s about a guy who surrounded by bad guys, who has to fight his way out. Oh, that’s Die Hard. Die Hard. It’s similar to Die Hard. And you say oh, well, aliens, well like, oh, Independence Day, that was super, super popular. And that’s all about aliens.

And so what you do is, you try to take the unknown quality and put the known (???). And so it’s like, “Oh, well Die Hard was popular. People liked that kind of movie. Well, Independence Day  was popular. People liked that kind of movie.” And you go, “Oh, I see. Hmm.” But you are mixing them. No one’s ever mixed Die Hard with aliens before, that might be interesting.”

And in some ways, when I pitch a Magic set, I mean essentially, what we do is, “It’s Gothic horror meets Magic.” “It’s Greek mythology meets Magic.” “It’s the guilds meet Magic.” “It’s an adventure world meets Magic.” That what we do every year is we try to take some known quantity, and another known quantity which is Magic, and the fun of it is the interactions. Like, oh, I know Magic. Oh, I know Greek mythology. Oh, how is Magic going to do Greek mythology? How are those things going to blend together?

And that one of the things that is very important is that we are always trying to—I mean I talked about this obviously on my communication theory, is that comfort is important. That people have to come and know what they’re getting into. And so a lot of what we need to do every year is make sure there’s consistency.

So for example, I just talked about the creatures. Now I get to the spells. So the first thing you’ll do in the spells is sometimes you’ll break out whether you want enchantments or instants or sorceries. Usually instants and sorceries are the same thing for the design skeleton at first. Eventually you might break them up, but in the beginning it’s just kind of “Oh, this is an instant or sorcery, this might be an enchantment.” Normally at common enchantments are auras.

The other thing that you will do is you will signify established spells that you need. For example, green at common is going to do a Giant Growth effect. It’s going to do some sort of land fetching. Odds are it probably will do some kind of life gain. It will do some kind of Naturalize effect.

Blue’s going to do a hard counterspell and a soft counterspell. Hard counterspell means it can counter anything, like Counterspell—a card that just says “Counter target spell.” A soft counterspell is a spell which only affects a narrow band, like “Counter target enchantment.” Or “Counter target creature.” Or it’s conditional, meaning “Counter target spell unless your opponent pays two.” Well, it’s not always going to counter the spell. But anyway, blue will always have a hard counter and soft counter. It will have a bounce spell. It will have card drawing. It will have card filtering.

Black will have creature kill. Usually there’s a straight-up Terror effect and then there’s an effect that’s a little smaller.

And the idea is, there’s certain things that we always do, that we make sure to put those in early on. That you signify, “Oh, this is a Giant Growth.” Now, what you do early on is you tend to default. Like for example, the Giant Growth usually is allocated to an instant early on. Why? Because most of our Giant Growths are instant. But it’s possible, let’s say you’re doing a set in which there’s a strong aura theme, oh, well maybe instead of a Giant Growth I make a +2/+2 or +3/+3 aura with flash on it. That functions a lot like Giant Growth, I can surprise you and it’s bigger, oh, but in aura world it’s permanently bigger. Rather than just being a temporary effect.

So one of the things you do is you sort of start establishing what you want, and remember, the design skeleton is a living, breathing thing. In that it’s constantly evolving. You notice there’s a theme in Magic that everything’s constantly evolving. That’s kind of the nature of the game.

And so what you are doing is, so I’ll use a different metaphor today, I used my blueprint metaphor, now I’ll use my storyboard metaphor. So in Hollywood, let’s say you’re going to direct a film. You’re going to make a film.

So what you do before you ever get to touching a camera is you get an artist, or multiple artists, and you do what’s called a storyboard. What a storyboard is, is you figure out all the shots you want to do, and you draw them. You have an artist actually physically draw them.

By the way, guys, I am sitting in traffic, as I look at my clock here. So you guys have an extra-long podcast today. Which is good, because there’s a lot that goes into a design skeleton. So luckily I picked a topic that can handle some traffic.

See, every once in a while, I pick a topic, and then I’m like, “Don’t be traffic, don’t be traffic, I have maybe like thirty minutes on this.” If there’s traffic I’m doomed. Doomed! But luckily today, I’m not doomed. I’ve written a whole bunch on this. So lots about design skeletons.

Okay. So anyway, the idea is, as you create stuff, it will evolve with what you’re doing. That—ohh, storyboard. Sorry. So if you’re a director and you’re making a film, you start doing storyboards. You get an artist to draw what you expect to see with the camera. And the reason you do storyboards is so you can map out what’s going to happen.

Because when you get to actually filming the film, you want to be able to do things when you see opportunities. For example, let’s say I get to the film and there’s a gorgeous sunset. I’m like, “I gotta capture the sunset.” So you might shift what you’re doing. You might change it around a little bit to try to capture that.

And remember, the cool thing about doing design is, you are not locked into anything. As you progress, you start committing to things. I mean, I guess—with time you get locked into things. But early on design, you have a lot of freedom. And that just because you set out to do something doesn’t mean that when you playtest, if you find it’s not working, or you find something even better, that you can’t capture that.

I mean, I talk about this all the time. That design or most creative works is iteration. Which means you’re going to keep doing—essentially the way you do a Magic set is, make a card file. Play it. Learn from that. Make corrections. Play again. Continue. Lather, rinse, repeat. That you’re basically constantly changing the files and playing with it. And seeing what you learned from it. Then making changes based on what you learn from your playtests, then playtesting again.

And what the thing that the skeleton does for you is it gives you the structure you need so you  can understand what it is you are doing. Because one of the things about building a set is, and this is one of the hardest things, is there’s a lot of moving pieces. There’s a lot going on. That it’s not like when you build a set there’s just one thing to focus on. There’s twenty things to focus on. Maybe there’s a hundred things to focus on. There’s tons of things. A lot of moving pieces are happening.

And what you need to do in order to see all those moving pieces is you need to track them so that you can understand what’s happening. And the thing that’s really, really nice about the design skeleton is, sometimes it allows you to see where your problems are.

So let me give you a real—like a concrete example. I think this was from Odyssey. Okay, so I had my design skeleton, and I realized that I had not yet done the [Rampant] Growth effect, but I ran out of noncreature spells. And so what I realized is I said, “Okay. I know I need to do a Rampant Growth, and I don’t have a spell left.” So I said, “Okay, well what if I made a creature, a cheap creature that you could sacrifice to go get a basic land?” 

And I’m like, “Okay, well that allows me, you know, you could use it as a creature for as long as you needed, but as soon as you need the land, you can go get the land.” And that was—the reason I sort of was able to solve that problem was, I instantly saw my—I had allocated things with my design skeleton so I understood what things were doing.

Like, one of the things is, it’s very easy, when you look at a card in a vacuum, to not remember what that card’s purpose in the whole set is. And a design skeleton helps identify things, like, “Oh, yes, yes. That’s part of this. That’s part of this cycle.”

Oh, that’s another thing. Let me talk about cycles real quick. So cycles are one of the most useful tools for a designer. In fact, one of these days I should probably do a whole podcast on cycles. In fact, I will. I (???) I will do that.

But anyway, suffice it to say, cycles are very important that when you’re doing your design skeleton, that you want to make room in place for the cycles. It also—when you signify the cycles, it’s very easy to lose track of cycles. Like, what happens, it’s very common, is you’ll make a bunch of cards, they’ll be a cycle, they’ll do something, and later on you’re trying to fix some problem, and without thinking you change a card because in a vacuum it just seems like whatever card, and later you’re like, “Ohh, that was part of a cycle!” And so it’s important that you label things and you have some understanding of where things are so that when you’re putting your things together, you can see the moving pieces.

Like I said. One of the biggest problems in design is trying to track all the moving pieces. In fact, one of the things I’ve gotten very, very good at, and like I said, I’ve put in—I’m working on my I think twentieth lead design right now? One of the things that you just get good at with time, with experience, is that I now have an intuitive sense of whether a set’s missing something. That I can look at a set and sort of say, “Oh, here’s what’s missing.”

And it’s funny, because my designers would love to get this ability, so they’re always like, “How do you see that? How do you…” and I’m like, “It’s just experience. You just do it enough that you get a general sense of things.”

Okay. So you have the skeleton. And basically, by the way, the way we build sets is, what we’ll do is we’ll start by making commons. So why commons? The reason you start at commons is—well, first and foremost, when you open a booster pack of Magic, on average, you will have one land, ten commons, three uncommons, one rare, which 1/8 of the time will be a mythic rare.

That means that 2/3 of your experience of every booster pack is common. That’s huge. 66.6% of all experiences of booster packs—and that’s me counting the lands. Actually, it’s even higher if you don’t count the land.

So anyway, it’s a huge part of your experience. And if common doesn’t hold your set, if your set doesn’t live and breathe at common, your audience won’t know what it is. I have a dictum I say to my designers, which is, if your theme isn’t at common, it’s not your theme.

And what that means is, let’s say I do an awesome thing. Like, the classic example of this mistake was in Champions of Kamigawa, where we wanted to have a legendary theme, and so we made all our rare creatures legendary. And we made a few uncommon legendary creatures, which is something we never, ever do.

The problem was, how many packs did I have to open to understand that? For starters, I only get one rare per pack. If it wasn’t a creature, I didn’t even see a legendary thing in my pack. I mean, maybe I got an uncommon. But I could open a pack and just see nothing.

Maybe on my second pack I do get a creature at rare. And it is legendary. But there’s legendary creatures. “Oh, I got a legendary creature.” It’s not until I open five or six—like, I have to start getting a preponderance of something to go, “Oh, this is not normal.” And the problem is, not everybody’s going to buy six or seven packs of your product, or not necessarily Day One.

I don’t want someone buying my product and opening up multiple boosters and not knowing what the product’s about. That is bad. In fact, I want them opening one booster and knowing what the product’s about. I want the theme to hit you so quickly that it only takes one booster to have a sense of what the set is about.

So obviously, there’s a wrapper—I mean, there’s things to help sell the theme beyond the cards. But I want—let’s assume you’re ignoring the booster wrap, I want you to have a sense from the fifteen cards you open what the set is about. That means your set has to live and breathe at common.

So, the very first thing we do is try to figure out how to make commons work. So normally what we’ll do is try to make design skeletons with commons, we’ll build the commons, we’ll playtest the commons. And then, with time we slowly layer in the uncommons, the rares, the mythics. Those come with time.

And the design skeleton, for example—you only need to map out common before you make common. You don’t necessarily need to have the whole thing done before you start at common. In my article, Nuts and Bolts, it’s easier to have you do the whole thing to explain it, but in actual practicality, there’s a lot of things that happen concurrently in Magic design.

I mean, it’s not like A then B then C, it’s kind of like A and B and C are kind of happening at the same time, so while you’re doing A, keep an eye on B, and an eye on C, and D might start soon, and… I mean, one of the things about doing design, and this is why having the blueprint, having the design skeleton is so important is, design is chaotic. Chaotic. Because you have all these different things you are trying to understand and figure out. And they’re happening in conjunction with each other.

So let me give another important tip. When playtesting, you want to make sure that you’re introducing one new thing at a time. Now, I’ll admit, we break that rule all the time, but I think that’s just—we have enough experience that we can watch multiple things and understand what’s going on. Sorry, I’m yawning today. I even got sleep last night.

And so one of the things that is very important is, because things are going on concurrently, you need to ground yourself to understand what is happening. And when you are playtesting, it really helps to have one thing to focus on, so that you can understand what that one thing is doing. If you have too many new things added in, then you just can’t tell when you playtest.

Like I said, experience—I can throw a bunch of things in, and I have enough experience that I can sort of separate what’s going on. But once again, that takes a lot of time. If you’re doing this for the first time, add one thing in. At a time.

And that it’s also—when you’re playtesting, by the way, there’s nothing wrong with trying a bunch of versions of something. When we were doing the land set, Zendikar, and we were trying out land mechanics, we would throw a whole bunch of land mechanics in the set at once, because we just wanted to experience different things. And it’s not so important that—even if you know you’re not going to use them all, you’re just trying to experience things.

That’s another important thing of early playtesting, which is the role of early playtesting is experiencing things, it’s not having a tight-wound environment yet. Don’t worry necessarily about how everything interconnects. Worry about if each individual card taken in isolation is something that’s fun and enjoyable to play.

Let’s see how we’re doing, time-wise… ooh. I should be at work. But I’m not. Although it’s moving along now. So I shouldn’t be super late today. It is funny though, my podcast made me rethink traffic. I now see traffic and I’m like, “Ehh, I’ll talk a little longer.” It lowers my blood pressure while driving. You guys are good for my health.

I’m trying to think, any other key thing on the design skeleton. The one thing that I will stress to people is, detail is fine in making a design skeleton. But be aware, there’s a difference between what you need and what you want. It is very easy when making a design skeleton to put in things you would like to have rather than things you need. The design skeleton is not about want. The design skeleton is about need. What do you have to have to structurally make the thing work?

Now, it’s fine when you’re making your set to do things you want to do, but the design skeleton is trying—the point of the design skeleton is what is necessary. Not—I mean, I’m saying this a little bit incorrect. You can have things in your design skeleton that are things you want. But you have to be careful that you understand what are the things you need from the things you want. And so in the design skeleton, I like to prioritize things you need.

So I often tell my designers, when they start  making their design skeletons, early on, talk more about what you need. If you really want to put some conditional stuff in there, put it in italics or something. Do something that when you look at it, you know it’s secondary to what you’re trying to do. Because a lot of what the design skeleton is trying to help you with is figuring out how to prioritize things, and how things need to come together.

I mean, the trick is, because game design or Magic design is about so many competing forces all happening at once, that a lot of what you need to understand is prioritize. And that’s a big part of making a set, is figuring out what do you need, what does the set need.

And one of the reasons I talked about tossing out things that are good but don’t fit your set, is they’re just taking up mind-space that is needed elsewhere. That if I have a card that’s this awesome card that does neat, wonderful things, but it’s fighting the rest of my set, it’s just going to cause me problems. And I need to take it out and free up space so that I can do more things that are helping my set do what my set wants to do.

There’s always  other sets, there’s always—if you make an awesome card, it will find a home. If it’s something that deserves to be printed, it will eventually get printed. But that doesn’t mean it goes in the first available place it can be put. That’s how bad design happens, because you force things in, because of your attachment to the thing, and not because of the role the thing plays in the larger design that you are doing.

The other thing about design skeletons is, they’re not written in ink. They are written in pencil. By that, I mean you can change them. If you try something and it doesn’t work, change it. If you’re having some problem, it can change. One of the things I’ve noticed once I tell people to start using design skeletons is, they write it, and then they assume it’s a locked thing. And then they get all in trouble and they don’t know how to solve the problem.

And I’m like, “No no no, if you run into a problem, a design skeleton is a guideline.” Much like if the architect’s building a building, and there’s a boulder on the side of the hill that they can’t dig out, “Oh, well, maybe we’re changing around the house a little bit to accommodate this thing that we can’t change.”

And design is the same way. Card design’s the same way. Which is, sometimes you discover, “Oh, I can’t do something.” It breaks a rule, or it causes some interaction that’s problematic. And I can’t do that. Well, go back to your design skeleton and say, “Okay, what are my non-negotiables? What are things that I need to have, and how can I work around knowing that I can’t have this thing I assumed I was going to have?”

The other thing to remember about a design skeleton is that the design skeleton is a tool, which means as much as you make  use of the tool is how useful of a tool it will be. Different people make design skeletons of different amounts. I’ve had designers who are super, super, super detailed, that they write every possible thing they can think of. That almost when you look at their skeleton, it’s almost like they’ve made the card file. It’s so—“This has to be an instant and it costs three mana, and it’s green, and it has to be a Naturalize effect, and it has to make use of the new keyword.” And like, “Well, you’ve kind of designed your card right there.”

Other people, myself, for example, like I don’t actually even use a design skeleton anymore. Because I’ve internalized most of it. I mean, I like to think I do use a design skeleton, but it’s in my head. But it’s very, very loose. And it’s super flexible. A lot of what I do is sort of I’ve learned to look at a set and get a sense of what it needs and what it doesn’t need. That I’ve sort of internalized this process. It takes a long time, so I will stress, use a design skeleton. I’ve been doing this for eighteen years, so I’m able to take a few shortcuts that I think most designers would be—I would not advise them taking.

So, I know I’m almost to work. The real point of today is this: there’s a lot of art to what I do. There’s a lot of art to card design. But there’s also a lot of craft. And that it’s fun to talk about the exciting, intuitive, things pop in your head moment. But it’s also important to understand that there’s a lot of work. There’s a lot of elbow grease. That it’s not a lot of sitting around and just thinking, and “Oh, that’s an idea.” A lot of time it’s plotting and planning and mapping things out and structuring things. And that there’s a lot of technical craft that goes into making card sets.

And the design skeleton’s just a sign that like you need to pay attention to that. If you do not pay attention to that, it will come to bite you. Because there’s too many things going on and too many things you have to care about, that without a real strong sense of a tool to guide you, you will get lost, and you will forget things, and the other thing that happens a lot is, because there’s so many things you’re trying to fit in, and while 269 cards might seem like a lot, when you’re actually trying to do a bunch of complex things, it gets tiny quick.

And we talk about getting pinched. Where at some point in the file you have too many things you need and not enough card slots to do them. And that’s a real problem that you run into every single design. That you get pinched somewhere, and then it’s a matter of figuring out, “Can I shift those abilities, can I change them in rarity, can I change them in color, is there a way to overlap two things that don’t normally overlap?” Like my example before with the creature and the Rampant Growth. Is there a way to have two things that need to coexist, but overlap them in slots? That’s very important.

So anyway, design is—like I said, today was a little more of a by-the-books podcast, where I’m sort of talking about the nitty-gritty of what we do. But the nitty-gritty is important, and that—and by the way, the nitty-gritty is fun. I very much enjoy structuring things and I love puzzles, and I think a lot of design is trying to crack puzzles.

And a lot of the puzzle solving is all the groundwork you do with your skeleton to figure out—like a lot of solving puzzles is figuring out all the parameters. And so I think the design skeleton is doing a lot of the homework. There’s a way to think of it. If you want to do design, you’ve got to do your homework. Design skeleton is your homework. Luckily it’s very fun homework.

So anyway, I love talking about Magic and technical design, but even more, I love making Magic. So it’s time for me to go. I’ll talk to you guys next time.