A Different World

Posted in Making Magic on October 11, 2010

By Mark Rosewater

Working in R&D since '95, Mark became Magic head designer in '03. His hobbies: spending time with family, writing about Magic in all mediums, and creating short bios.

Last Wednesday, we started our search for the eight finalists for The Great Designer Search 2 (GDS2). The first test was a ten-question essay test. The next test, to be taken this week, is a fifty question multiple-choice test. Around one hundred applicants are going to advance to the third test, a design test. A big part of this test is that we're asking each applicant to give us an idea for a Magic world. Today's column is partly to help them understand how to build a Magic world and partly to let all of you get some insight into a very important aspect of Magic design.

    Boy (or Girl) Meets World

Let me start with the chicken and egg question of Magic design: which comes first, the design or the flavor? The design spells out the flavor but the flavor inspires the design. Which comes first? The answer is that it can happen either way. Sometimes, we know the flavor we're going after, such as Scars of Mirrodin, and the design is all about making that flavor come alive and sometimes we start with a design idea, such as Zendikar, and build the flavor as we flesh out the design.

Let's explore each of these two approaches:

    Flavor Begets Design

The large fall set in 2011 is "Shake." It began because we had a world we wanted to visit. Brady Dommermuth (Magic Creative Director, a.k.a. the head of the creative team) and I talked about what the world would be like years before the design team was even assembled. We had a concept for a world very much based on flavor and Brady and I bounced back and forth ideas about what kind of world it would be.

When my design team (myself, Jenna Helland, Graeme Hopkins, Tom LaPille and some random guy named Richard Garfield) sat down for the first time, I explained that our design goal was to create cards and mechanics that matched the flavor of the world. This is known as "top-down design." The goal of the designers is to match the feel of the world.

While this might sound attractive, it's the harder of the two approaches. Why? Because flavor is simply more flexible than card design. Champions of Kamigawa block had this problem when it was being designed. The design started with the concept of a world shaped by Japanese mythology and found its design very forced at points because there were too many things that just didn't have a simple mechanical analogy.

The trick to this approach is to not narrow down your flavor too much. Get a general sense of the world and then start designing. Allow the mechanics you create to help narrow down the aspect of the flavor you want to play up. For instance, I keep joking with Magic brand about designing a circus world, so I'll use that as my example (mostly because I'm pretty sure circus world will never be seeing the light of day—probably as it should be, by the way).

So the design team sits down on day one of circus world design. The first thing I would ask my team is what does circus world evoke? What would you expect to see? I take all of their answers and write them down. I am particularly interested in answers that multiple people came to. For instance, after talking with my team for circus world, I find the following things come up most: clowns, animals, trapeze artists, ringmaster.

So what could that mean mechanically? The first observation is that everything we listed was a person or animal or, in Magic terms, a creature. If the heart of your world falls on creatures that means the design is going to be creature focused. Most of the things listed get even more specific, everything named is creature-type focused. You are going to have lots of animals and humans with class based professions. The last interesting point is what are you going to do with clowns? While in the real world, they're humans, perhaps in the world of Magic they are something else. Maybe they're their own race. Perhaps clown wants to be a new creature type. Once you recognize this focus, it's clear that this set wants a tribal component, maybe even as the main focus of the set.

The next thing the design team does is dig a little deeper. Sure the circus has people in it but what else? You start thinking about the tents and the three rings. Perhaps there's a carnival alongside it or a freak show. Dig deeper to get to the crux of the circus. It's about performing. It's about making money. It's about people who might be freaks banding together. It's about people pushing what they can do.

As you talk about each of these ideas, certain ones will start making mechanical connections and certain ones will not. Follow what makes those mechanical connections—don't force what isn't. For example, the idea of a freak show definitely inspires in me the idea of creatures that do things more heightened than what you are used to. Is there any interesting design space in finding cards that push limits?

The trick here is that you have to let the ideas that work, guide you. Performing might be a giant part of what the circus is but if that element leads you nowhere mechanically you have to let it go. Remember that the goal is not to represent the circus as realistically as you can but to use the idea to inspire the best set you can. A well-built circus-inspired world is what you're after, not a world that slavishly copies the circus at the cost of game play. The world has to serve the game play. We're designing a game.

The hardest thing about this approach is the need to occasionally throw away a great idea. You will come up with ideas that cannot be well represented in the game. I often talk about the hardest part of design is letting go of a good thing that isn't servicing the design. Don't' let your mechanics get held hostage to an idea. The key is not to try to make every aspect of your world work. You only need enough that you convey what you want to convey.

Another important point to remember is that your audience will never see what got cut. They won't know what they could have had only what you give them. If that taps into what you need and conveys the sense of your world, what got cut will never be missed. Don't confuse your vantage point (which has the knowledge of all of your cuts) with the audience that has never had it.

The lesson of starting with flavor is that flavor is used as a guide. You will have more flavor than you could possibly fit into a set. What you are doing is figuring out what aspects of the flavor can best be represented through mechanics. Remember that in the end, a Magic expansion is judged by how it plays. A wonderful, gripping, dynamic world that plays poorly is a bad world. Form has to follow function in design.

I'm not saying that the flavor doesn't matter, because it does—very much so. What I'm saying is that the goal of leading with flavor is finding what parts of your flavor best allow good game play and then focusing on them. Circus world only needs a small part of our list to feel like circus world.

The big advantage of a flavor-led design is twofold. First, the flavor will lead you to places that you wouldn't have naturally discovered resulting in designs that are a little more untraditional. Second, it allows you to be slightly more complex than normal in your mechanics because the players come to them with built-in knowledge. Assuming your mechanics match your flavor, the players can use this knowledge to learn your new mechanics quicker and thus that allows you to raise the bar on complexity while keeping comprehension the same.

Now let's explore the other approach:

    Design Begets Flavor

Design on Zendikar started with me knowing one thing—I felt there was a lot of unused design potential involving lands. The first meeting with my team was all about mining that space. Unlike the last approach, this world building begins by exploring an area of design space. I said to my team (myself, Doug Beyer, Graeme Hopkins, Ken Nagle and Matt Place) I wanted them to come up with every land-based mechanic they could think of whether it concerned land or only went on lands. The team came back with over forty mechanics and we started playing them.

We playtested all the different mechanics generated, slowly weeding out the ones that weren't working. Eventually, that path led us to landfall. I talked to the creative team (and note one of the team members, Doug Beyer, is on the creative team—not an accident) and got them to start thinking about a world where the type of land mechanics we were playing with made sense.

The difference in this approach is that design needed to do some work before creative got involved. We had to get mechanics far enough that we started to get a sense of what the set was doing mechanically. The important thing is that we stuck with just one aspect and figured that out first. This allowed us the freedom, once the world started to be fleshed out, that we could use some of the design space to design to the world.

In the case of Zendikar, that land theme led creative to ask the question: why is the land doing what it's doing? The answer was that the land had to have different properties than normal. That's when they came up with the idea of the land being the antagonist to the heroes in the story. (Also figuring out why the land got this way led to the creation of the Eldrazi for later in the block.) The heroes versus the world led to the idea of an adventure world, one where the inhabitants had to fight the land itself.

Once the design team heard the idea of adventure world, we started from the same point as the first approach. What would you expect to see in an adventure world? We ended up with the idea of maps, traps, and adventurers (a.k.a. "chaps"). Those ideas led to quests, traps, and allies.

The main difference between this approach and the last one is that this approach doesn't bring in the creative until some mechanical identity has been reached. The key though is to limit the mechanical identity to one thing. Otherwise, you end up not having room to adapt the mechanics to the world. Adventure world that doesn't have quests, traps and allies is not meeting the level of flavor/mechanics integration that we like to see.

Another important point I want to talk about in this approach is that part of finding mechanical identity is finding something that has enough openness to allow other things to hook into it. If your initial mechanic is too insular it will become harder later to layer in the mechanics created to fit the world. The key to proper integration is making sure that the mechanics that beget the world and the mechanics begotten by the world can be interwoven. The set's core mechanics cannot feel fractured. I like to think of this style design as papier-mâché, where each layer is put on top of what comes before it to make something that, when combined, is strong enough to hold the shape of the design together.

The advantage of this approach is it creates a very strong structure for gameplay. This extra structure allows the designs to pull off some feats that flavor-based designs cannot. Mechanical-based design can have a much more rigid structure which is necessary when trying to make certain things fit. For example, Shards of Alara block wanted to build to an all-gold third set. By starting from a mechanical vantage, we could plot out what needed to be done to make an all-gold third set possible. A top-down design wouldn't have the rigidness needed to create all the necessary set-up.

Both styles of design are important as they lead to different kinds of blocks. The key is recognizing what kind of world you're building so that you can lead with your strength.

    As the World Turns

So how can you tell if you're supposed to start with flavor or mechanics? To answer this question you have to first identify the core of your set. Every set is, at its center, about one thing. As the lead designer, it is your job to identify what that thing is. Why is it so important that you only have one thing? It's important because when you start building your set you have to know what your priority is.

But what if you have more than one priority? The entire point of having priorities is to understand their order of importance. To use a metaphor to explain this, let's talk TiVo (or any DVR for that matter). Most DVRs have the ability to record seasons of television shows. Whenever "Show X" is on, your DVR will tape it. Sometimes though you tell it to tape more shows than it is capable of taping. How does it solve this problem? By having a "series manager" screen where you list your priority. Show 1 will always tape. Show 2 will always tape except if doing so would keep Show 1 from taping. Show 3 will always tape except when it prevents the taping of Shows 1 and/or 2. And so on.

I bring this up because your designs have to work like the "series manager" screen. You have to know what comes first. Then if anything comes in conflict with it, your main priority needs to win. Only by keeping a consistent focus can you ensure that you are making the choices you need with your set.

A common problem of less-experienced designers is that they allow too many different things to pull their focus. They make something they really like and squeeze it in even if it doesn't really fit in with what the rest of the set is doing. Remember, the right thing in the wrong place is actually the wrong thing. Part of shaping a world is figuring out what fits in with your vision. Things that don't fit, no matter how good they are in a vacuum, have to go.

    Out of This World

This topic is far bigger than a single column, but I will try to give you some more pointers before I wrap it up. Things to keep in mind:

    1. Your Job As Designer Is To Convey The World Not Through Flavor Text But Through Rules Text

Design has to work with the creative team but world building isn't something design just hands off. The key to building a world from a design standpoint is making the play of the game itself convey the flavor. One of the reasons, I'm very happy with Scars of Mirrodin is that the mechanics themselves convey the conflict of the environment. You get a sense of the Mirrans and the Phyrexians by playing with the cards that represent them.

Before I move on I do want to stress that there is one flavor element that I urge designers to spend a little time on—names. A lot of presenting design to others is conveying flavor. As you don't have the luxury of the creative team naming the cards yet, you have to take some of the burden to make sure that the names play up the flavor you are trying to create mechanically on each card. You would be surprised how easy it is to miss top-down concepts without a name to guide you. Always remember that the playtest names are a tool for the designers to help explain to others the feel of the card designs.

    2. Think Of Your World As A Jigsaw Puzzle With All The Mechanical Elements Intertwining

Once upon a time, a Magic block had two new keywords and called it a day. The two mechanics seldom had anything to do with one another. That was then. Today, block design is very much about creating a dynamic environment that allows many options, especially in Limited play. To accomplish this, Design has to create synergy between its mechanics and other cards.

The easiest way to do this is to pull out all the elements of your set and write them down on paper. Start mixing and matching items on the list. Ask how they play together? When you find two elements that fight one another, ask yourself—which one is more important? Then keep that one and cut the other. Note that sometimes the connection won't be mechanical but flavorful.

    3. Playtest, Playtest, Playtest

Would you ever buy a car without test-driving it? Cards and mechanics are no different. It's very easy to come up with reasons why something will or won't work, but theory only gets you so far. There have been many mechanics that I was dubious about but that grabbed me when I played with them. On the opposite end of the spectrum, there have many mechanics I loved until I played them.

A lot of people think designers just sit around and think up ideas all the time. Getting the idea is just the first step of the creative process. (Roger von Oech has a wonderful book on this topic called A Kick In The Seat Of The Pants—it's the sequel to my favorite book, A Whack on the Side of the Head.) Once you get an idea, you have to execute on it and then test it. Then, and only then, can you judge it.

    4. Most Of What You Create Will Not Work And Get Thrown Away

There is a lot of disposal in design. As I have said numerous times, I am the most prolific Magic designer in the history of the game having led the design for fifteen sets over fifteen years, and even I am lucky if over 1% of my work ever sees the light of day. To quote Pablo Picasso, "Every act of creation is first an act of destruction."

Design is dirty work. If you want to do it right, you're going to have to get your hands dirty. This is hard to do because designers (and artists) become emotionally attached to the work they do. Killing cards often feels personal, but it's what the job entails. If an idea isn't doing what it needs to do, no matter how good it might be—it's got to go.

Always be aware that inertia will make you want to keep things the way they are. Mediocre cards will sneak by because, well, they begin feeling comfortable to you.

    The Real World

Today's lesson is a hard one because there is no one right way to build a world. I've created numerous worlds and I'm not sure if I've ever done it the same way twice. Each world has its own needs and, as such, gets built according to its own standards.

The key, though, is understanding the need for focus and discipline. To have vision you have to know what's important and then make the necessary decisions to support it. In addition, never forget that you're there to make a game. No idea will work if it doesn't play well.

I hope this gave you all a few things to think about. For those of you that make it to the design test portion of the GDS2, I wish you all the best and I'm anxious to see what you come up with. To everyone else, I hope today's column has given you a better insight into what we in Design do here day in and day out.

Join me next week when I'll craft some metal.

Until then, may you find world peace.

Latest Making Magic Articles


June 27, 2022

A Double Take, Part 2 by, Mark Rosewater

Last week, I started telling some card-by-card design stories about cards reprinted in Double Masters 2022, but there were a lot of stories to tell, so I'm continuing today. Body Double ...

Learn More


June 20, 2022

A Double Take, Part 1 by, Mark Rosewater

Before the release of Double Masters 2022, I thought it would be fun to look back at how some of the cards were made. This set has cards from all over Magic's history, so over the next tw...

Learn More



Making Magic Archive

Consult the archives for more articles!

See All