Hello and welcome to the second week of Return to Ravnica previews. Last week, I talked all about things we did prior to Return to Ravnica's design, but today it's finally time to actually get to the work we did in the design itself. Today, I'm going to walk you through the five guild keywords and how we came up with them. Also, I have a fun preview card to show off, so let's stop wasting time and get to it.

Art by Chippy

    "I Love It When A Plan Comes Together"

Return to Ravnica was a very different kind of design. Normally, the beginning of the design for a large set is figuring out what the block is all about. We spent the first few months of Zendikar, for example, trying out land mechanics. The idea of an adventure world didn't happen until well into design. Return to Ravnica understood its world Day One.

But what about Scars of Mirrodin? Wasn't that also returning to a previously visited world? It was, but the design for Scars of Mirrodin block was about a war. Yes, the setting was familiar and we used that nostalgia in our design, but Scars of Mirrodin wasn't anything like Mirrodin. Yes, there were some trappings that were similar and I tried to make sure the Mirrans were using mechanics that felt similar to what we saw them use the previous time we visited, but the block was very much about the presence of the Phyrexians, and that was an entirely new addition to Mirrodin.

Return to Ravnica, though, was a return in every sense of the word. We weren't changing what Ravnica was. The block wasn't about large changes being brought to Ravnica. No, it was a simple continuation of what we had done before. We wanted to go back to a beloved world and see it as it was but evolved with time.

This meant that the key factor to making everything work was to latch onto the exact same thing we used last time we were on Ravnica—the guilds. Last we saw, there was a dissension of the Guildpact on Ravnica. I'll let the story folk explain how we got from there to here, but suffice it to say we knew the guilds were going to be the drivers of the design.

This meant that the first part of design was figuring out what new things each guild was going to do. For starters, I made the call that every guild was going to have a new keyword. We talked about having a mix of new and returning keywords, but the problem was that, due to the effect of cramming so much into so little space, along with New World Order complexity issues, each guild could only have one keyword. We felt that if some got new ones while others got old ones, that it would feel unfair to the guilds without new keywords.

Also, as I talked about in this year's State of Design column, one of the biggest challenges was finding a way to combine nostalgia with a sense of novelty. Having new keywords allowed us to do something new while still getting the guild feel we needed. We talked about using returning non-guild keywords (for example, unearth is so Golgari) but I felt strongly that each guild needed something brand new. We did look at a few Future Sight future-shifted mechanics (like delve) but none seemed a perfect fit, so we never had to answer the "Are cards already previewed from the future new?" question.

What follows is how each guild's keyword came about. I put them in the order we figured them out. (At least the order as I remember it. Memories may vary.)


Art by Chris Rahn

Blue loves the intellect. Red loves the emotions. Blue wants to think. Red wants to feel. What happens when you combine these two? You get very passionate thinkers. The Izzet don't see intellect and emotions as opposites. They feel that by combining the two you get something even greater—creativity! The Izzet are inventers. They live to explore uncharted ideas. Any and every uncharted idea.

Mechanically, blue and red are the two colors more weighted toward spells (that is, they have the highest ratio of spells to creatures). Since I know I'll get asked this, here's the order from most creatures to fewest:

White (the color of the army gets the most)
Green (the "creature color" used to get the most, but to separate it from white it now gets the biggest and white gets the most)

Guildpact played into this love of creation by having the Izzet have the most spellcasting (aka the greatest number of instants and sorceries). To help reinforce this flavor, the Izzet got the mechanic replicate.

We knew going into Return to Ravnica that the Izzet were still going to be the most spell-centric guild. This meant we had to find a spell-based mechanic for them, something that would go on instants and/or sorceries. Luckily, Ken already had one in his back pocket. In fact, he had submitted it five years earlier as one of the finalists from the first Great Designer Search. (Ken, by the way, talked a great deal about the design of the Izzet and Golgari mechanic in his feature last week, so I'll only be hitting the high notes here. Please read his article for even more detail.)

Check out out all the drama from The Great Designer Search

Ken called the mechanic dispersion, and it was modeled after the Torment card Radiate.

My comment on the mechanic as the head judge:

Your common cycle used the dispersion mechanic. It's a good mechanic that I could see us using. It has the nice flexibility in that it can be burned early cheaply for a small effect or create a larger effect later for a larger cost.

I think it was the first meeting (my memory could be faulty, but I like to remember it as if it was because it happened almost immediately) that Ken suggested using dispersion as the Izzet mechanic. I remembered it from The Great Designer Search and even remembered that I liked it. Ken was very excited about it so we made some cards for it on the spot. The mechanic went in and was never removed.

As Ken talks about in his feature article, the big issues we had with overload (the printed name of the mechanic) was finding the right effects to put it on. In general, we wanted the overload effect to be positive, so we ended up restricting what the effect could do—positive things could only target your stuff while negative things could only target opponents' stuff. This would allow you to use overload without having to weigh whether the negative impact to your opponent outweighed the negative impact to you.

Because Izzet was spell-centric, we tried like last time to make sure there were a number of other cards that liked being in a deck of instants and sorceries. Our goal was to make sure that if you took all the cards with the watermark of a guild from both Ravnica and Return to Ravnica blocks, the cards would play well together. These instant and sorcery enablers help hold Izzet together.


| Art by Christopher Moeller

White likes the group. Green likes the group. Put the two together and you have a guild that values the group over everything else. Selesnya understands that the greatest strength comes from the bonds between people. A group is greater than the sum of its parts. With unity comes power.

This means that the Selesnya mechanic wants to be creature-based. Selesnya is all about growing its power through growing its ranks, so it wants a mechanic that either rewards you for having a lot of creatures or helps you get more. Convoke, the Selesnya mechanic from Ravnica block, helped you do both, allowing you to use your creatures to get more creatures—as well as a few spells—cheaper.

I was very interested in finding a mechanic to help make more creatures. Scars of Mirrodin had just come out a few months earlier and it was clear that proliferate was a big hit. In one of the early meetings we were talking about the Selesnya mechanic, when the following exchange happened (as always, this conversation takes some dramatic license):

Me: How about we just do proliferate for tokens?
Ken: What would that mean?
Me: Whenever you, let's call it propagate, you get a copy of each type of token you have on the battlefield.
Ken: Explain in greater detail.
Me: Okay, I have two 1/1 Squirrel tokens.
Ken: We're not going to have Squirrel tokens.
Me: My example takes place in an alternate dimension where we can have Squirrel tokens. I call it the Odyssey dimension.
Ken: Fine.
Me: I have two 1/1 Squirrel tokens, three 2/2 Bear tokens, and one 3/3 Elephant token.
Ken: Bear and Elephant?
Me: It's the Odyssey dimension. 4/4 would be a Beast and 6/6 a Wurm.
Ken: Continue.
Me: If I propagate, I would get one 1/1 Squirrel, one 2/2 Bear, and one 3/3 Elephant.
Ken: Okay, let's try it.

So we did. Playtesting showed propagate (that was the playtest name) had three problems:

  1. It was broken.
  2. It was really broken.
  3. It encouraged weird deck building, where you were trying to get as many different types of tokens into your deck as you could.

For the second playtest, we decided to make one small change. Instead of getting one of every token type, you just got to pick one token and copy it. This fixed both problems. It played well and propagate (later renamed populate) stayed as is for the rest of design.

Populate did have one huge design issue, but before I get to that I should address an issue with populate that I've seen online. Cloning is a blue thing. What's Selesnya doing with it? It doesn't even have blue in it. The answer is that sometimes the necessity of Magic wording to be precise tends to use words that are a little misleading. Populate is technically cloning but the copying comes with a huge limitation—you can only copy tokens. This is limiting not only because tokens are a tiny percentage of creatures in Magic but also because tokens are very limited in what they can do.

For starters, the vast majority of tokens are vanilla creatures (that is, they have no text—just power and toughness). The rest of the tokens, with only a few exceptions in the history of the game, are French vanilla (they only have creature keywords), almost all of which only have one keyword, and for the vast majority that keyword is flying. So, yes, green and white get to technically copy creatures, but mostly what they're doing is making tokens, which is right up green and white's alley.

Back to the design issue at hand. So we made populate. What problems does it cause? Populate is what I call a Reese's Peanut Butter Cup mechanic. (For my international readers, a Reese's Peanut Butter Cup is an American candy that combines milk chocolate with peanut butter.) In order for it to do its thing, you need to have both chocolate and peanut butter. In populate's case, you need to have both a card with populate and a card that makes tokens.

Token cards stand alone, meaning you can make cards that make tokens and players who care not one iota about populate can still play them. Populate, on the other hand, is useless without tokens. The solutions to this problem were as such:

  1. Make more token-making cards than populate cards. Also, up the "as-fan" (Ramp;D speak for how often they show up in boosters with rarity skews accounted for) of token cards by sticking them at lower rarities.
  2. In general, make more token-making cards than normal.
  3. Whenever possible, try to put both effects on the same card. This way there is a built-in use for the populate effect on the card itself. Skew these types of populate cards at lower rarities since they can stand alone.
  4. Skew populate cards that only populate at higher rarities and at higher costs to decrease players being able to cast populate spells without having anything happen.

Populate was definitely a thorn in design and development's side because it requires crafting a careful environment for it to thrive. We didn't want to do that with too many guilds, but everyone liked populate enough that we decided it would be okay for one guild.

Before I move on to the next guild I would like to show off my preview card for today, because it happens to be a Selesnya card. In fact, I am going to show off the guild leader of Selesnya—Trostani, Selesnya's Voice.

Trostani, Selesnya's Voice

Because populate was influenced by proliferate, I was eager to see if we could create cards with repeatable populate. The problem was that getting a creature is a much bigger deal than getting an extra counter (well, most of the time at least). In the end, we managed to find room on one high-profile card—the guild leader of Selesnya.

Have fun—Trostani has a lot of fun design potential.


Slitherhead | Art by Greg Staples

Black appreciates the value of death. Green appreciates the value of life. The Golgari appreciate both. It's just one big cycle, after all. The Golgari are what I like to think of as extreme recyclers. Things dying is not an end of anything, it's just the start of a new way to use it.

As such, the Golgari are all about the graveyard as a resource. First time around, they got the dredge mechanic.

For those who might not know your Magic history, dredge went on to be the most broken mechanic of the Ravnica block. It's so powerful that it shows up in every format it can be played in, using decks that look little like traditional Magic decks. The trick for Return to Ravnica was to make a mechanic that used the graveyard but did not abuse the graveyard.

Ken talked about this mechanic's design a great deal in his feature. The reason the mechanic happened third isn't because the rough idea wasn't around early (the idea that players can salvage their dead creatures for value), but that it took a while to figure out how exactly to correctly make the mechanic work.

Sluiceway ScorpionDreg Mangler

For a quick recap, here's what happened. The earliest version of the mechanic, called digestible, allowed you to exile a creature. If you did, you could grant to another creature a power boost equal to that creature's power, a toughness boost equal to that creature's toughness, and any abilities that creature had, all until end of turn. I believe the reason we started with the weak version of the mechanic was the fear created by not wanting to make another dredge.

Playtesting showed that the temporary effect wasn't strong enough, so we tried a permanent version. Instead of granting a temporary power/toughness boost, we now put +1/+1 counters on the card. This, by the way, forced us to make all scavenge cards (that's the final name) with square stats—Ramp;D speak for power and toughness being the same.

The first attempt of the +1/+1 counter version also permanently granted the abilities of the creatures but that ended up having some issues:

  1. +1/+1 counters do a good job of reminding you what you need to know. Three counters? The creature has +3/+3. Having other information tied to those counters—this creature has deathtouch or this creature flies—proved to be mentally taxing.
  2. The complexity issue along with rules issues made it hard to make scavenge cards with abilities other than keywords.
  3. The increased power level made us have to cost the cards even worse and scavenge already had the problem that the mechanic looked far worse than it was.

After some deliberation, we decided to drop the abilities from the scavenge keyword. I know whenever I say things like this, I get letters from people with things like "Aw. Why'd you do that? I could have handled it!" My response is this: The goal of a game design is not including everything that will fit but rather only including the things that need to fit. The ability-laden version of scavenge was wordier, had more rules issues, was more confusing, was harder to track, and didn't add enough play value to offset all the things I just listed. The leaner version was a better mechanic. That's why it was chosen.

The above three mechanics were the ones handed over in the design file. The next two would happen during devign. (Devign is the period in between design and development—two months for a large set—where design still has control of the file but development makes comments to make sure the file is ready for development.)


Art by Kev Walker

Black loves itself. Red loves doing what it wants to do. Neither has any issue with doing whatever it needs to do to make that happen. Put the two together and you have a very hedonistic, dangerous group. One that does, at least, know how to have a good time.

The mechanic for Rakdos in Dissension was hellbent.

Hellbent played into the full-throttle play style of Rakdos but never quite captured the feel of the guild. It made the guild feel a little more reckless than focused on its hedonism. This time around we wanted to find a mechanic that had a little more "your pain, my gain."

The first mechanic we made was called paincast. I can't tell you what it did (although feel free to guess from the name). The reason I can't tell you is because while it didn't stay the guild keyword, it did stay in the set. When you get a chance to see the guild leader Rakdos (yes, the guild is named after the demon who leads it) you'll get to see what paincast is all about.

Paincast fell by the wayside because of... Gatecrash. What? You see, while Ken and company were hard at work on bringing the first five guilds to life, I and my company (yes, I was in both companies) were busy making the second set of five guilds. For a while, all five of the Gatecrash guild keywords were creature-centric, most of which had a big impact on creature combat.

It was noted that we needed to have fewer creature combat centric mechanics (I'll talk with you during Gatecrash preview weeks how we went about making that change.) At the same time, Aaron Forsythe (Magic Director of Ramp;D) looked at the Return to Ravnica guilds and noticed that no one had any keywords that really mattered in creature combat. Izzet had a spell mechanic. Selesnya was making creatures. Golgari had a graveyard mechanic. None of these meant anything during combat. Aaron decreed that one of the Return to Ravnica guilds should have a combat mechanic.

Rakdos was the clear favorite, not only because it seemed like the Return to Ravnica guild that most liked to fight but also because paincast was proving to be a tough mechanic to work in. Like populate, paincast demanded a lot of cards be designed around it, but we didn't feel it had the potential upside of populate. Some on the design team had asked that we just put it on a few cards and have it lose keyword status. Making a new combat mechanic would kill two birds with one stone.

Dead RevelerGrim Roustabout

The unleash mechanic (what the mechanic was finally called) came about because Aaron decided he needed to do more than just tell us something had to be done. He designed some creature combat mechanics for Rakdos. One of them I liked. It was very simple. A creature with the ability could choose to enter the battlefield with a +1/+1 counter. If it did it gained "This creature attacks each turn if able."

I suggested one small tweak. I liked that the creature could choose offense or defense but I thought the forced attack could create a lot of bad feelings. In addition, it allowed the creature to block the turn it was played (provided it didn't have haste) which also felt wrong. I suggested changing the forced attack to "cannot block." This both made it feel not defensive and created fewer feel-bad moments.

The team agreed with my recommendation and unleash went in the file all the way to print.


White likes order. Blue likes knowledge. Put them together and you have one super uptight guild whose greatest weapon is bureaucracy. If you control and understand the rules, you can make the world work the way you want it to work.

Every time I've worked on a guild set, there's always been one troublesome guild that we had a hard time finding a mechanic for. (Oddly, never the same one.) For Return to Ravnica, that troublesome guild was Azorius. Going into Dissension, control decks were very powerful, so the Dissension design team was asked to downplay Azorius's control nature. As I explained above, that is Azorius's nature. We ended up playing up fliers in Azorius but never quite got the guild to the right feel.

The mechanic in Dissension was forecast.

For both power and logistics concerns, forecast was not pushed and Azorius never got a chance to shine.

Going into Return to Ravnica, Ken was on a mission to do Azorius right. We didn't have the same constraints so we were able to play up Azorius's controlling nature more.

The earliest mechanical space we explored had two major issues. One, it didn't play well with what the other guilds were doing and, two, it stepped on the toes of what we wanted to do in "Friends" block. (You'll see a year from now when I do next year's fall preview columns.)

We tried a number of different mechanics but nothing really stuck. I don't even remember what mechanic we went into design with. It was that memorable.

One of the things we'll do in design if there is some piece of the set that is proving problematic is to create a sub-team. That team usually has two to four people and it meets for a week or two to focus on the problem at hand. The Azorius had enough issues that a sub-team was created.

The sub-team was led by Mark Globus and included Ken and developers Dave Humpherys and Billy Moreno. After a bunch of meetings, the sub-team came up with the detain mechanic (created by Mark Globus, I believe). The concern at the time was that detain wasn't very splashy. It played well and felt very Azorius, but there was worry.

My response was that the role of a guild mechanic is to make something that played well, fit the guild identity, and appealed to players who liked that guild. Yes, detain wasn't for everyone, but I felt confident that it would be liked by Azorius players. The design team agreed and detain was added to the set.

    The City That Never Sleeps

That's all the time I have for today. There's still much more to say about the set's design, for which I'm glad there's a Part 3 coming up. Join me next week when I talk about a whole bunch of other aspects of Return to Ravnica's design.

Until then, may you find the guild that suits your personality.