Last week, I started talking about how the original Ravnica: City of Guilds block introduced a lot of new concepts to how we design sets. Here are the three elements I talked about last week:

  • Equal color weight for two-color pairs
  • Hybrid mana
  • Multicolor factions

But that wasn't all Ravnica introduced, so today's article is a continuation of last week's column.

Color Pie Philosophy for Two-Color Pairs

In part one, I talked about how the guilds introduced mechanical factioning, but the guilds were an innovation on several fronts. The next innovation had to do with how they incorporated the color pie. Frequent readers will know that when Richard Garfield made Magic, there were three genius ideas he created, what I like to call the Golden Trifecta:

  1. The concept of the trading card game (podcast | transcript)
  2. The mana system (podcast | transcript)
  3. The color pie (podcast | transcript)

If you're interested in hearing more details on why I think each was such a great idea, I did a three-part podcast series on them and included the links to those above, as well as links to the transcripts of each if you'd rather read them.

I consider the color pie to be the foundation of the game. The flavor and mechanics are both built on top of it. It provides the ethos to the game and gives it a psychological grounding that makes Magic unique. (Here's a link to a list of all my articles and podcasts about the color pie.)

Up until Ravnica, we tended to approach the color pie only through individual colors. White means this, black means that. We talked a lot about their relationships, how they were allies or enemies. Usually, cards that reflected more than one color mechanically were about one color helping or hurting another. Yes, we made multicolor cards, but they were done in isolation, and we never thought of them philosophically as combinations of the colors, at least not in any bigger way than the individual card flavor.

When Brady came up with the idea of the guilds, he and I were both intrigued by exploring what it meant to be the combination of two colors. Orzhov, for example, was white and black. What did that mean from a flavor standpoint? How did this group function? What was their philosophy? What kind of creatures would be part of the group? What did this group do in the context of the larger world?

And mechanically, what did this mean? Each guild got its own mechanic that only it had. What did that mechanic need to do? What style of play did the combination of these two colors want? How could gameplay match the feel of the guild?

We both knew that the questions were intriguing and ripe for cool content.

As we explored all these questions, we found that we were taking the color pie to the next level. Yes, you can divide Magic up into its five colors, but we could go beyond that. Ravnica divided the colors into ten groups, and by doing so, we were making divisions that had even more definition. Last week, I talked about the emotional power of factions. A lot of that comes from this psychological unpinning. That being two specific colors means something, and it speaks to the kind of player you are. Interestingly, there wasn't just one way to look at the two-color pairs. Yes, Ravnica tended to hit the low-hanging fruit options, but there are numerous ways to look at how the colors combine, and that let us do other two-color combinations. In addition, we've been able to do three-color factions as well. (We have looked at four-color combinations; the restriction there is more related to mechanics rather than flavor.)

The key to all of this was the discovery of the color pie's power, which was even stronger than we'd believed (and I was already a huge color pie fan). Magic has always had a strong element of self-identification, and the ability to get more in depth in how we use color by combining colors has proven to be a potent tool that's now standard to how we approach sets.

Archetype Flavoring

Factions and color pie philosophy are both amazing, but the guilds don't stop there. Another important thing they did was introduce the idea of flavoring archetypes. As Draft has evolved, we've discovered that it's important to help the drafters understand what draft options are available to them. The important tools in this regard were uncommon gold signpost cards that strongly conveyed the two-color archetypes in the set. This technology would come after Ravnica, but the guilds did add an important tool that helps in a similar way.

If you flavor the cards that are supposed to be played together as belonging together in story, you help communicate to the players that they want to use these cards together. This is helpful in a draft because it's another way to get players to instinctively select cards that will work in conjunction with one another. In casual play, it reinforces players' desire to have cards in their decks feel like they belong together. Creating cards that players intuitively put together is an aspect of good game design and establishes a sense of natural decision-making that's fun for players. (This is, by the way, one of twenty lessons I did for a talk I gave at the Game Developers Conference in 2016. I later turned it into a three-part article that you can read here: Part 1, Part 2, and Part 3).

The flavored archetypes have another big bonus: they help make the world feel more cohesive. The nature of a trading card game is that players will mix and match cards. This will lead to a lot of quirky creative pairings. By creatively connecting mechanical themes, you increase the times that a deck will have a creative throughline—that is, the deck will have components that give it a larger cohesive theme.

This is a component that took a while to sink in. After Ravnica, we made a big point to flavor factions, but it took us a while to realize that we can flavor archetypes even when it isn't a faction set. Wilds of Eldraine is a good example of this. It's not a faction set, but we did flavor each of the two-color archetypes to a fairy tale, at least our version of it. The feedback from the players was that it added a lot to the set, both in how it felt and how it played. Flavoring our archetypes is now becoming standard practice.

Watermarks and Iconography

We're not done with the influence of the guilds just yet. Here's a question that came up midway through design: do we want cards to signify what guild they're from? Sure, it would be easy to know what guild a multicolor card is from, but any monocolor card would have multiple options. Is that something we cared to indicate? Yes. But how?

Unique frames felt a bit much, and we didn't want to use rules text because we felt it might prevent us from making some cards we wanted. The solution ended up being a tool we'd only used on promo cards up to that point—something referred to as a watermark. A watermark is an image that can be put lightly in the text box underneath the text.

A quick aside because I get asked about this a lot: Watermarks are not allowed to be mechanically relevant. The reason is that all cards that have the same English name must be mechanically identical, so things that might change between printings of a card, such as artist, expansion symbol, and watermarks, can't be mechanically referenced by the rules. If you're eager to do that, though, Unstable did have it as a small theme. (Un- sets are much more cavalier when it comes to the rules.) Couldn't we just promise to always use watermarks on certain cards? No. That would mean we couldn't reprint cards into a set that has watermarks and might end up with watermarks in a set that contradict the other watermarks in the set.

Because a watermark is usually a shape, we had to create symbols for each of the ten guilds. Those symbols were then used in many places. They showed up in card art. Prerelease Packs each themed after a guild had stickers and a pin using the guild's symbol. We made merchandise with the symbols, tee-shirts being particularly popular.

Ravnica taught us that watermarks and symbols have a lot of power when used carefully. They also allow for the enfranchised players to find symbols that feel more personalized to them. We now think of them as a valuable tool that can be used to enhance a set. Not every set wants or needs them, and we must be careful not to overuse them, but when used properly, they can be quite impactful.

Pushing Boundaries

One of the earliest playtests for Ravnica had all ten two-color pairs, including hybrid cards for all of them. After the playtest, R&D member Henry Stern came to me. He said, "Mark, I was a semi-finalist for two back-to-back World Championships. I'm one of the better Magic players in the world. This was too much for me. I think I had to put my cards in something like 35 piles. There's no way the file can stay like this."

When you tackle the design of a new set, you encounter all sorts of obstacles. Most ask, "What do I need to do to offset this issue?" A few, though, ask a different question: "Is this even possible?" Just because you come up with an idea doesn't inherently mean the idea is possible. Sometimes in design, you progress by realizing that something you wanted to do can't be done, and then you find a different way to accomplish your larger goal. After that playtest, I started questioning if what I wanted was feasible.

Here was the core of the challenge. I wanted to make a set focused on two-color play, and I wanted all ten two-color pairs to appear equally, but having ten two-color pairs was just too much for a single set. The low-hanging fruit would be to do either ally or enemy colors in one set and then do the other five in another set. That's basically what the Invasion block did. Invasion and Planeshift used only ally-color cards, and Apocalypse used enemy-color cards. But one of my main goals was to make it not feel like Invasion. This was the first time we were repeating a block theme, so I wanted to make something that felt different than what we'd done before. Was there a way to do that?

Before I walk through how I answered that question, let me start with a brief history of block design. Ice Age, the block we talk about as technically being the first block, wasn't designed to be a block. Alliances wasn't designed to be connected to Ice Age. That was added in development. Mirage and Visions had been designed together and were later broken into a large and a small set. Weatherlight was done internally by a completely different team and didn't have a lot to do with Mirage and Visions. Tempest, Urza's Saga, and Mercadian Masques were designed as full blocks, but the connective tissue was mostly that they shared two-keyword mechanics (and with Mercadian Masques, they weren't even keyworded). The way we designed them was to make the first set, choosing exactly two mechanics that were expandable, and then let the small sets figure out what they wanted to do when they got there.

The Invasion block was Bill Rose's first as head designer. This was the start of block themes. Invasion was about multicolor, Odyssey was about the graveyard, Onslaught had a typal theme, Mirrodin was about artifacts, and Champions of Kamigawa was a top-down set built around the flavor of Japanese mythology. We weren't restricted to two mechanics, and small sets were allowed, even encouraged, to have new mechanics. But the model of making them was the same. Design the first set, then we would figure out how to design the small sets when we got there.

This led to what we called "the third set problem," where the third set often found itself in trouble because it ended up having restrictions that limited what it could do. The classic example of this was Fifth Dawn. By the time we got to its design, we realized that the block was full of broken mechanics and almost every mechanic we'd used in Mirrodin or Darksteel was either completely off limits or had to be done in such limited volume that the set couldn't be built around it.

When I became head designer, I vowed that I wanted to do more block planning. Namely, I believed that the role of the first large set was to create definitions for each set in the block. I was inspired by the Invasion block. We'd set out with all ten two-color pairs in the block but pushed back the enemy stuff until the third block to give us more space to make additional ally-color cards. I liked how it gave the third set a real definition, something that seemed the antithesis of the "third set problem."

I'd spent a lot of time coming up with different block design models, one of which I called the pie method. The idea behind it was to think of a cool design that you cut into pieces, giving different pieces to each set in the block. Any one set wouldn't have the whole pie. You'd have to play all three sets to get the whole pie.

Back to how I problem solved for the Ravnica issue. There are ten two-color combinations. Ten is too much to fit into any one set. What if we applied the pie method to it? What if each set only had some of the two-color combinations? Structurally, I thought I could do it. My bigger problem was the justification issue. Why does green-white, for instance, only show up in one of three sets?

My solution would come when Brady Dommermuth came up with the idea for the guilds. If each guild was its own little subset, complete with its own mechanic, that could be the justification. The guilds would become the structure that defined the block, with each set being responsible for just some of them. I was excited. I believed we could build it and that the players would accept it. The challenge, however, was convincing the rest of R&D.

For a little context, I was proposing something we'd never done, something we hadn't even done anything like. Every set we'd done with multicolor had a balance of either every ally-color combination, every enemy-color combination, or both. The idea that the first set would have four color pairs and the other six simply wouldn't appear at all was radical. I got my design team on board. I made sure my boss, Randy Buehler, was okay with it. I even got the head developer, Brian Schneider, to give me a thumbs up. In the end, I had enough people to agree to it that the set happened, but there was some tension along the way.

An interesting thing about Magic design or, one could argue, all creative endeavors is that when you're successful, it encourages others to do the same, or as I like to say, "Success breeds repetition." Ravnica pushed the boundaries for what a Magic set could be and how design can be stretched between sets. In doing so, it set the standard for what design was allowed to try. It made it easier to be bolder because pushing boundaries became less of a scary thing and more of an accepted thing, something we should aim for.

Ravnica was my first outing as head designer, and I had a lot of ideas I wanted to try. My success with the Ravnica block did a lot to help pave the way for all that followed. I like to think that it became the ideal of what Magic design could be. Looking back, I'm very proud of all that the Ravnica block accomplished. I feel it inspired everyone involved in its creation to bring their A-game and make a true home run of a set, one whose influence would shape all of Magic design to come.

City Tour

That's all the time I have for today. I hope you enjoyed this look at the original Ravnica block and all the influences it's had on future Magic design. If you have any thoughts on this article, last week's article, or the Ravnica block itself, you can email me or contact me through any of my social media accounts (X [formerly Twitter], Tumblr, Instagram, and TikTok).

I'll be off for the holidays over the next two weeks, but I'll be back in three weeks with an article talking about why Magic design sets so many restrictions (only cast this as a sorcery, only play this on your turn, only play this once per turn, etc.).

Until then, may you have happy holidays and get a chance to play a lot of Magic.