Nuts & Bolts #10: Creative Elements

Posted in Making Magic on March 26, 2018

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.

Welcome to another installment of "Nuts & Bolts," my series dedicated to players interested in designing their own Magic cards and/or sets. Once a year, I write another column for this series. For those interested in making your own cards, this series provides hands-on technical advice. For those not interested in making your own cards, it gives an inside look behind the scenes to see how we put Magic sets together.

Here's a recap of what I've written so far.

Nuts & Bolts #1: Card Codes

The first column is the most technical, explaining how we use a system to make sure everyone is always talking about the same card.

Nuts & Bolts #2: Design Skeleton

The second column introduces the most important tool in designing a set, something called a design skeleton. (It makes use of card codes, which is why that article came first.)

Nuts & Bolts #3: Filling In the Design Skeleton

The third column talks about how you begin filling in your design skeleton, starting with the common cards.

Nuts & Bolts #4: Higher Rarities

The fourth column talks about filling in all the other rarities.

Nuts & Bolts #5: Initial Playtesting

The fifth column discusses how to best use playtesting to gather feedback and make improvements on your set.

Nuts & Bolts #6: Iteration

The sixth column talks about the concept of iteration and how you can incrementally improve your set.

Nuts & Bolts #7: Three Stages of Design

The seventh column explains the three different stages of design, walking you through how your priorities shift as the set evolves.

Nuts & Bolts #8: Troubleshooting

The eighth column answers a number of questions about common problems that can happen in the early-to-mid design stages.

Nuts & Bolts #9: Evaluation

The ninth column talks about how you can look at your set as a whole and figure out what fine-tuning it still needs.

This all brings us to today's column, number ten in the series. You've made your design skeleton, filled it, playtested it, iterated it, evolved it, troubleshot it, and evaluated it. Today, I'm going to discuss something that goes on concurrently with all the design work: the creative elements of the set. By that, I am referring to the name, flavor text, card concepts, and worldbuilding (plus art if you have any). At Wizards, there's an entire team dedicated to these elements of the set, but it's still part of the designer's job to interact with this team and to make sure that the design itself is interwoven with flavor. At home, it's usually the job of the designer to do most (or all) of the creative work. Today's column is going to talk about how to incorporate the creative elements into your design.

Once Upon a Time . . .

To start with, usually before you begin a design, you have some idea of the world you're going to be playing around in. We refer to a design as "top-down" if the flavor of the world drives your design (examples being Innistrad, Theros, and Amonkhet) and "bottom-up" if the mechanics drive it (examples being Ravnica, Zendikar, and Khans of Tarkir), so I want to begin by talking about how each element plays into early design. I'm going to break this down into three phases: initial (where you start building your set from scratch), adaptive (where you have to bring in the element you haven't focused on yet), and combining (where you are interweaving the two). For each phase, I'll talk about how top-down and bottom-up designs handle it.

Initial Phase


Usually you start a top-down design by asking: What would the audience expect in a world of this known flavor? Innistrad wanted monsters. Theros wanted gods. Amonkhet wanted archetypal Egyptian things (mummies, pyramids, scarabs, etc.). My technique for the first meeting with my design team is to list everything we can think of. I then ask my designers to use those ideas as inspirations for the earliest cards and mechanics. Innistrad wants monsters? Okay, what kind? Werewolves? How would we design a Werewolf that captures the expectation? It was this path that led us to double-faced cards.

The key here is to use the flavor as the guiding force to figure out your vision for the set. Most of your cards should have playtest names that are evocative and resonant. My technique is to label cards after the trope you're trying to hit. I called the Vampire lord "Count Dracula," not because I thought the final execution would be similar, but because I wanted my fellow designers to understand the purpose the card was serving in the set.

The trick with top-down sets is that the flavor is instrumental in how the set is structured, because you're going to want your structure to reinforce your themes, which are being guided by your source material. Innistrad's cohesion wasn't mechanical, but rather linked together through the flavor. This means for this type of design, you need to start incorporating your flavor from the beginning.


With a bottom-up set, you don't need to worry about flavor quite as early. It's fine if you have some idea about your creative elements, but you can make use of functional (Small Blast) or silly (This Is Gonna Hurt) playtest names. The adventure world theme of Zendikar didn't happen for many months. The same is true of the city and guild flavor for original Ravnica and the warlords/Sarkhan's home plane flavor of Khans of Tarkir.

If the core of your set is mechanical, what you're going to want to do early is figure out the relationship between the mechanics; you can focus less on the flavor. Zendikar messed around with mechanics interacting with lands and Ravnica made cycles of each two-color combination. Focus your early energies on whatever is the glue of your structure so that you can start building your design skeleton to match.

The trick with bottom-up sets is that the flavor is secondary to the structure and thus doesn't have to be figured out quite as early. That doesn't mean that you can't have ideas for where your set takes place, but you have less obligation to have that guide your design.

Adaptive Phase


Once you have a handle on your setting and theme, you have to start to piece together what your mechanical structure needs to be. I'll use Innistrad again as an example. We knew we wanted monsters, we knew we wanted death to matter, we knew we wanted to evoke a sense of fear, and we knew we wanted dark transformation. The monsters led us down a path where we started listing the types of monsters we wanted: vampires, werewolves, zombies, and spirits, with humans as victims. As we started placing them into colors, we began to see a pattern: the five tribes were falling naturally into ally color combinations. We added a little bit of a tribal component, and all of a sudden, the gameplay started to feel like monsters were an integral part of it.

Death mattering and evoking a sense of fear got us to look at the graveyard and care about whether things have died. That got us to look at graveyard mechanics and to experiment with a new mechanic that checked if something had died that turn. These explorations led directly to bringing back flashback and creating morbid.

The dark transformation started by us focusing on making Werewolves work mechanically, but eventually led us to double-faced cards that allowed us to tell stories where innocent things could transform into darker, scarier things. Once we had the double-faced technology, we started finding all the different horror tropes we could hit with it.

The key here is that we let the flavor drive the goals for the set but then used mechanics as a way to reinforce those themes. Monster tribal, flashback, morbid, and double-faced cards/transform were each a result of us finding a mechanical execution that allowed the gameplay to capture the overall feel that was driving our design.


Once you have a sense of what you want mechanically, you have to start asking yourself, what kind of a world would the mechanics make sense in? I'll use original Zendikar as my example for this section. We started by focusing on land mechanics. We eventually settled on landfall as our main focus, along with various other ways to reference or care about lands. We then had to ask ourselves, what kind of world would allow us to focus on lands? We also knew that caring about lands meant you would have more lands in total in the set, putting another restriction on the setting.

This exploration led us to realize that what we wanted was a world where people versus environment was the driving source of conflict. That meant the land had to play a more active role. Digging into this was what led us to an adventure world where the adventurers had to overcome a hostile environment. The people were trying to survive against a world that was actively trying to kill them.

Once we latched onto the genre, we were free to start exploring the many different subgenres. The subgenres then led us to lists of tropes and resonant clusters, which we collected into lists.

Combining Phase


The point of this phase is to take your two elements, creative and mechanical, and start integrating them. For a top-down set, what this usually means is examining your flavor elements and finding opportunities to tie them together mechanically. With Innistrad for instance, this is the point where we examined each of our monster races and started to give them a mechanical theme. We chose to make Zombies slower and controlling. We made Vampires more aggressive. We integrated the Werewolves with the double-faced cards and figured out a unified trigger that would turn Humans into Werewolves and a different trigger that would turn Werewolves back into Humans.

We also examined all our different top-down card designs and tried to find places where we could incorporate our mechanics. We had a spell that produced a huge number of Zombies to capture the idea of being overwhelmed by a Zombie horde. What if we added flashback to that spell so you could make a giant horde of Zombies not once but twice?

The trick in this phase is trying to mix and match flavor and mechanics in such a way that the audience won't be able to tell where you started from. I bring up top-down and bottom-up because they are radically different from the designer's side of things, but when a set is done well, the audience shouldn't necessarily be able to distinguish how it was made.


Once we know our world, in a bottom-up design, we want to dedicate some mechanical space to things inspired by the world. For Zendikar, that meant finding some elements of adventure world flavor and building some mechanics around it. We explored a lot of ideas but ended up with what we, at the time, called "traps, maps, and chaps." Traps became the trap mechanic, which allowed you to cast cards at a reduced cost if the opponent performed a certain game action. Maps became quests, an enchantment mechanic where you were asked to do the same feat multiple times and were rewarded when you did it enough times. Chaps became Allies, a tribal theme that allowed creatures with the Ally creature type to work together, much like an adventuring party.

You'll notice with both top-down and bottom-up that we save some design space to allow elements of the opposite component to be built in later. Top-down sets leave room for flavor to be layered on top of mechanics, while bottom-up sets leave mechanical room to design based on the flavor once it's decided.

Now that we've walked through early design, let me give some practical tips about when and where to incorporate creative elements.


If you're doing a top-down design, you're going to want your design names to help shape the feel of the set. Early on, it's fine to reference the trope or inspiration directly in the name. The key is ensuring the people playing with the card get the flavor you're trying to capture. For top-down, the early playtest names are less important. You need playtest names so you can refer to cards, but the names can be on the more practical or sillier side if you wish.

Once you move out of the initial design, around the time you're going to want to start having outsiders playtest your sets, you're going to need to start coming up with more realistic names. Top-down sets are going to need to shift away from trope references to actual names you'll want to use on your cards. Bottom-up sets are going to need to get a sense of what each card represents.

The best way to think of this phase is by comparing it to something we do at Wizards. We need to have an artist illustrate the card. Before we can do that, we need to do what we call card concepting. What exactly does that spell represent? Let's say, for example, it's a direct-damage spell that can hit a creature or player for 3. What exactly is the spell doing? Is it using fire? Or lava? Or heat? Or earth? Or sound? What exactly is the spell doing physically with the magic? If it's a creature, this is where we figure out what creature type it is (if the card has tribal concerns, usually the design team has already indicated that it needs to be a certain tribe) and what kind of role it plays.

While many of you might not be putting art on your cards, going through the exercise of what your card would show if it had art is a good way to approach creating real names. Here are some tips on names:

Avoid real Magic card names – If you call a spell Lightning Bolt, many players will read the title and stop reading. They know what it does. Even if they understand that it's a different spell, it will still cause confusion and lead to players comparing your card with the named version, which will warp your playtesting notes.

Don't use words associated with other colors – Because the color philosophy has become so engrained in how we name cards, putting the wrong word on a card will throw people. Maybe you intend your Fire Rhino to be green, but the mere presence of the word "Fire" will make players assume it's red.

Make sure the name implies what the spell does – The card name has a function, and that's to help the player remember what it does. Having a name that implies something different than what the card is doing will cause players to misplay it.

Name the card so it matches the type of spell it is – Creatures should sound like creatures. Instants and sorceries are usually named with verbs to imply action. Artifacts are physical objects. Enchantments are intangible concepts or magic. Lands are a place. Planeswalkers should be given character names. You want to match expectations with card names, not fight them.

Avoid Magic lingo that contradicts your card's function – If you call a card "Mill," players will assume it forces the opponent to put cards from their library to their graveyard even if it doesn't do that. Magic has spent 25 years building up a vocabulary and assigning meaning to various terms. Fighting that vocabulary will result in players jumping to incorrect conclusions.

Try to use names that actually fit on a real Magic card – While it's funny in early playtest to have very long names, if those are used on "real" cards it makes them feel less real.

Make sure your card names sound different from one another – This is a problem Magic has had over the years, with Quick Sliver and Clickslither being the most famous mistake. Having your cards sound too similar just leads to players playing one card assuming it's another.

When making up proper names, make sure people can pronounce the names – The trick I use is to show the name to someone unfamiliar with it in a vacuum and see if they can pronounce it. If they can't do it instantly and easily, I change the name.

Be stingy with "25 cent" words – That's the R&D slang for words that the average player doesn't know. A few at high rarities are okay, but it's an element that works best in small volume.

Creature Type

While the creature type can be mechanically relevant, most of the time, it's more about adding flavor to the card. Think carefully about what you'd like each creature to be. Combined with names and flavor text, this can lead to cards that liven the player's imaginations.

Some tips on creature types:

Don't always default to the most commonly used creature types – It's easy to make every red creature a Goblin, but that will lead your set to feel less rich.

Humanoid creatures can have both a race and a class – The race is what they are and the class is what job they perform. It's not essential that all humanoids have both a race and a class, though. Usually when we leave one off it's the class.

You can have multiple creature types – A lot of fun can come from combining different things into one creature. Just be careful of the card type line length. Your creature types all have to fit. Legendary creatures are the ones that most frequently run into space issues.

Be creative – You're making up new cards; you can also make up new creature types.

Flavor Text

Besides names and creature types, the one other place to show off your world and characters is in flavor text. If you feel uncomfortable writing it, chances are you know someone who would enjoy creating some.

As far as flavor text, here's my advice:

It's not mandatory – You can have a successful set without flavor text. If you're intimidated by writing flavor text, and you can't find someone else to help, it's okay to leave it off.

More is less – If you're unsure whether or not flavor text will fit, leave it off. Wordy cards are intimidating, so there's no reason to compound that problem by adding flavor text where it's unneeded.

Use the flavor text to aid comprehension of the spell – Sometimes the name and concept are a little removed from the mechanics of the card. Flavor text can do a good job of reinforcing the connection between the two.

Use flavor text to build your world – There are not a lot of ways to let people know about your world. We have the luxury of art and an entire worldbuilding team, which you probably won't have, so use flavor text as a way to convey important information about your world.

Use flavor text to build your characters – Quotes from a character do a much better job of helping players grasp who that character is than explanatory text.

Remember that the best flavor text works in isolation – What makes flavor text fun is that it's bite-sized and doesn't require additional information to comprehend. Keep that in mind when creating it.

Last Thought

Before we wrap up today, there's one last thing I want to stress. Your mechanics are a creative element. How they play will shape how the audience perceives your flavor. Yes, you should spend time on names, creature types, and flavor text (and art if you're able), but don't forget to make sure your mechanics themselves are integrated with the flavor so that the act of playing your set makes the player feel something. Magic's best successes have come when the mechanics and flavor are so intertwined that they feel like a singular thing rather than two components.

Have Fun Storming the Castle!

That's all I have for today. I hope this helps get your homemade set one step closer to completion. As always, I'm eager to hear your thoughts on both today's article and the "Nuts & Bolts" series as a whole. You can email me or contact me through any of my social media accounts (Twitter, Tumblr, Google+, and Instagram).

Join me next week when Dominaria previews begin.

Until then, may your set feel as good as it plays.


#521: GDS3 Trial 2, Part 1
#521: GDS3 Trial 2, Part 1

In this podcast, I go over the answers for the first 25 questions in the Trial 2 multiple-choice test for the Great Designer Search 3.


#522: GDS3 Trial 2, Part 2
#522: GDS3 Trial 2, Part 2

In this podcast, I go over the answers for the second 25 questions in the Trial 2 multiple-choice test for the Great Designer Search 3.

Latest Making Magic Articles


Doubling Down by, Mark Rosewater

Hello everyone, and welcome to Battlebond Week. Normally during a preview week, I introduce the design team, talk about the design of the set, and show off a preview card. But I didn't wo...

Learn More


Odds & Ends: Dominaria, Part 1 by, Mark Rosewater

Every set, I dedicate a column or two to my mailbag to answer whatever unanswered questions exist about the newest set. It's time for me to answer some questions about Dominaria. Here's ...

Learn More



Making Magic Archive

Consult the archives for more articles!

See All

We use cookies on this site to personalize content and ads, provide social media features and analyze web traffic. By clicking YES, you are consenting for us to set cookies. (Learn more about cookies)

No, I want to find out more