Most weeks, I write about how things in sets came to be. Today, I'm going to flip it over and talk about something I seldom talk about: How do some things not come to be? Why do certain things not make it to print? In particular, I am going to talk about some of the things players expected to be in Dragon's Maze that weren't there.

The reason Dragon's Maze makes a good case study is that it is a set all about designing to completion. We had two large expansions set up each of the ten guilds and Dragon's Maze was made as a capper that added a little more to each guild. The problem, as I explained in my preview week articles (Part 1 & Part 2), is there was a lot to get in. A lot! A lot, a lot!

Two quick caveats before I jump into the explanation: First, I'm going to talk about mechanics and not flavor. I'm the head designer, so I'm focusing today on the design and not the creative. I'm going to talk about why certain cards or swaths of cards didn't end up in the set. Second, I'm not going to talk about execution of cards in the set. Yes, certain things could have been done differently, but my focus today is not how we could have changed the set we ended up with but rather why certain elements were absent.

Missing

Let me start before I begin going through the list by giving a reason that applies to everything I'm going to talk about. "Why wasn't _____________ in the set?" There wasn't enough space!

One of the realities of design (and development) is that you always want more than you can fit into your set. There are tricks to squeeze extra things in, but, in the end, you have to start cutting things. I think most of the players think of design as making things, but a big part of it is actually cutting things. We design way more than will ever fit into a set, no matter how big it is (and remember that Dragon's Maze is the one small set of the block).

The set's design was crammed as tight as a set could be (at a level we'd be willing to see printed, that is). Some stuff just didn't fit in. But couldn't we have gotten ridden of those cards you don't like? No, because those cards were for someone else and they'd be unhappy of we took them out.

With that out of the way, let's get to the list. (Note that they are in no particular order.)

#1: The Nephilim

In Guildpact, the following five cards appeared:

 

There was worry that some players might not like the guilds, so we made a cycle disconnected from the entire guild structure. It was a gold block, so we decided to make something we had never made before—four-color creatures. I was a huge critic of them at the time because I felt that the guild structure would go over well and I didn't see the need for the cycle that, to me, felt out of place. I even argued that we could make four-color cards by having two guilds team up, but as these cards were being designed for the players who might not like the guilds, my idea was voted down.

So we made them and they went over like a lead balloon. They were slammed in our market research. Players hated them. Why? Here are my best guesses:

They didn't fit: The reason I believe the original Ravnica block went over so well was because players loved the world that was built and the overall structure of the guild model. Creative came up with an explanation why the nephilim existed in the world, but it always felt forced. They are not organic to the world or the guilds. They were designed to feel disconnected and they did that job too well.

The designs were subpar: The major problem is that it's nearly impossible to design four-color cards that actually feel like four-color cards. Why? Because there are just too many colors fighting for attention to get a feel. The biggest thing about four-color cards is the missing color, but you can't design about what's missing, as that's the only thing the card isn't allowed to have, color-pie-wise. I do think there is a way to design four-color cards, but it requires some context (such as two guilds working together, interestingly enough).

Their flavor was a mismatch: The flavor of nephilim might be cool in the right environment, but I'm not sure in the middle of the city of guilds was that place.

If the nephilim were so hated, why are people asking for them? Because since the first Ravnica set came out, the Commander format has taken off and the one glaring hole for commanders is four-color commanders. There aren't any. As I always explain, humans are pattern completionists by their nature, so a gaping hole just calls out for some hole filling.

Now, add to this a comment I made in my column on Commander, "On Wedge."

Magic currently has eleven five-color legendary creatures (Atogatog; Child of Alara; Cromat; Horde of Notions; Progenitus; Reaper King; Scion of the Ur-Dragon; Sliver Legion; Sliver Overlord; Sliver Queen; and Karona, False God). It has zero four-color legendary creatures. (I'd errata the Nephilim to be legendary in a heartbeat if I was allowed to do such a thing.) In three-color, it's doing pretty well with the arcs, but the wedges are woefully small in number.

I've bolded the important line. I said I wish we'd made the Nephilim legendary. The reason I said that is because I knew when we returned to Ravnica it was highly unlikely we were going to make more nephilim. Remember, they were hated by the players, hard to design, and a misfit for the world. On top of that, in order to match the player expectation, they would have to be legendary for Commander purposes, and Dragon's Maze was already a small set with ten legendary creatures.

On top of all that, too many elements were fighting for limited space and the nephilim just didn't have a good enough reason to make the cut. The nephilim just didn't fit, in multiple meanings of the word.

#2: Hybrid Cards

Let me start by saying that I'm talking about cards with hybrid mana costs and not other hybrid costs (extort and Savageborn Hydra are in Dragon's Maze, for example). Both Return to Ravnica and Gatecrash had cards with hybrid mana costs—each guild had a vertical cycle, one each in common, uncommon, and rare—why not Dragon's Maze? The answer has to do with the human brain. To use a metaphor, imagine the brain as a tea cup. You only get to pour so much into it before the tea starts spilling out.

In terms of designing Magic, there is a limit to how much stuff you can put into a set before it's just too much for a majority of the players. Dragon's Maze was already pushing what we feel is acceptable for a set. It has eleven keywords and ten guilds, each with its own theme. In Limited, for example, if you divide your cards by color, artifacts, and land, you have seventeen piles—and thirty-two piles if you divide creatures and spells. And that's not even counting the rare split cards that have multiple, but different, two-color cards on them.

The set was already adding split cards to the mix. The design team decided that one "multicolored but not gold" subset of cards was enough, so we chose to give hybrid a rest.

#3: Equipment

Wouldn't it be cool to give the maze runners Equipment to use? In fact, wouldn't it be neat to have guild-aligned Equipment? Yes, it would, and we even designed some. Here's the problem: Every time we make a cycle in Ravnica, we have to make a ten-card cycle.

Dragon's Maze has 156 cards. Eleven of those cards are lands (the Gates and Maze's End—the shocklands were pick-ups—same printing as the previous sets—and don't count toward the set number). That leaves us with 145 cards. That's 14.5 cards per guild. That might, at first blush, sound like a lot of cards, but it's not. The fact is, we had a lot to do and little space to squeeze it all in.

In the end, the Equipment fell victim to the following fact: while players might have wanted some guild Equipment, no one expected it. It wasn't something we had to deliver on. When things are fighting tooth and nail for slots, ten-card cycles that are a luxury more than a necessity just don't make the cut.

#4: More Cards with Guild Mechanics

This category is less of something that was missing and more about something that some players thought would show up in larger quantity. Why weren't there more cards with guild mechanics? Two reasons: First, the mechanics we chose for guild mechanics were often chosen because they were mechanics with a smaller amount of design space. For some mechanics, there just weren't a lot of elegant designs left.

Second, as I explained above, the set was trying to fit a lot in a little space. That meant that there were many compromises that had to be made. One of those compromises was seeing how many mechanics we had to have rather than how many the set could have.

The condensing of mechanic space was also the reason behind certain mechanics not showing up in both colors of the guild in monocolor.

#5: Particular Reprints from Ravnica Block

Lightning Helix
Mortify
Electrolyze
Spell Snare
Prophetic Bolt
Absorb
Condemn
Birds of Paradise
Remand
Telling Time
Silhana Ledgewalker
Ogre Gatecrasher (How could we miss that one?)

The list goes on and on. What happened to that awesome spell from the original Ravnica block that would have been just perfect to reprint? The answer is this: the majority of the power of a set needs to come from new cards. Why? Several reasons:

First, reprints are a known commodity and an important part of Magic is the discovery process. Second, we want the newness to shine. If we use up all the "power points" (not an actual thing, I just want you to be able to envision the idea that there's only so much power to go around in a set) on reprints, that means we have less with which to make our new cards exciting. Third, some of the most exciting cards from the previous visit were exciting because they pushed what was acceptable. Some of them, we don't want back.

Remember, we did bring back some reprints. We just chose to be judicious with how many.

#6: Dragons

I often talk about the importance of meeting expectation. So, Dragon's Maze—where are all the dragons? We do have Dragonshift that makes Dragons, but other than that the set is pretty dragon-light for a set with "Dragon" in its name. What happened? Here's the important thing you have to understand. Much of the time, a set is designed and developed long before it gets a name. We didn't know when Dragon's Maze was being designed that it was going to be called Dragon's Maze,so we didn't put dragons in it.

When the set did get named Dragon's Maze, the fact that the set didn't have dragons was raised. The topic was actually discussed at length. As the name was referring to Niv-Mizzet, we talked about Niv-Mizzet's Maze, but that just didn't sound as good. In the end, it was decided that it was important that the name hit the story point and that hopefully players would get the title was referring to one particular dragon as opposed to dragons in general. In the end, that didn't play out quite as well as we'd hoped. That leads us right into the next category:

#7: Niv-Mizzet

He's the dragon being referenced in the title. Okay, there weren't dragons, but what about the dragon? We talked about this as well, but there were a few problems. First, he was already in Return to Ravnica as a legendary creature. He had to be because he's the leader of the Izzet and all the guild leaders appeared in either Return to Ravnica or Gatecrash.

Couldn't we give him a second, new card? There are some problems with that. First off, we don't often print the same legendary character on two different cards, especially in the same block, but when we do we always make sure there's some fundamental shift in the character. Usually, that means they change colors, but Niv-Mizzet had to stay blue-red as he's the leader of Izzet. Sometimes, the character goes through a big transformation—for example, Mikaeus dies and becomes a Zombie. There wasn't a huge transformation Niv-Mizzet went through.

Couldn't he become a Planeswalker? (I know that's where a number of players thought we were going.) The problem was he didn't in the story. Magic already has a super smart, powerful dragon Planeswalker in Nicol Bolas and the creative team didn't see a need for a second one.

Let's assume, though, that none of that mattered. Let's say we just wanted to make him a new card. Where do we put it? Ral Zarek is in the set at mythic rare. There were only ten mythic rare slots and we wanted one per guild. Putting Niv-Mizzet in would mean that Izzet would get two and one guild would get none. That didn't feel right.

How about making him a rare card? Well, rare already had a cycle of legendary creatures: the maze runners. Having Niv-Mizzet at rare would mean Izzet gets two legendary creatures (and remember they already got the Planeswalker) while the other guilds all got one.

Sometimes, in design, things just don't fit, so you have to learn to live without them.

#8: Selesnya X Spell

In Return to Ravnica, Izzet got Epic Experiment, Rakdos got Rakdos's Return, and Azorius got Sphinx's Revelation. In Gatecrash, Boros got Aurelia's Fury, Gruul got Clan Defiance, Orzhov got Immortal Servitude, Dimir got Mind Grind, and Simic got Biomass Mutation and Nimbus Swimmer. That means Golgari and Selesnya didn't get X spells.

Then along comes Dragon's Maze. Golgari got Gaze of Granite (also Orzhov got Debt to the Deathless, Dimir got Reap Intellect, and Gruul got Savageborn Hydra), but where's Selesnya's X spell. Come on R&D, how'd you forget to finish the cycle?

The answer is simple. We didn't even realize there was a cycle. How could we not notice that nine out of ten guilds had X spells? It's easy to miss when you piece them together one at a time over the course of a year. You all get to see the block in its conclusive whole, but R&D doesn't have that luxury. When we work on the set, it's constantly shifting, and sometimes things slowly line up that we just don't notice. Guild X spells was one of those things.

#9: Three-Colored Spells/Mixing and Matching of Guild Mechanics

Remember Shards of Alara block? We introduced five three-color shards, each with its own mechanic (some named, some not). Then, over the course of the year, we slowly started combining those mechanics with one another on the same card. Why didn't Return to Ravnica block do that?

The answer is that Shards of Alara block was about five distinct worlds all collapsing down to one shared world. The combining of the mechanics played into the story and the world. The same is not true for Return to Ravnica block. This block is about the definition of the two-color guilds. Three-color spells or spells that mix and match the guild mechanics simply don't fit.

It's easy to want to experiment for the sake of experimentation, but staying true to what the block is about is how we keep each world feeling right.

#10: Kiora and Fblthp

Other than Niv-Mizzet, these were the two characters whose absence was most asked about. Let me explain why each of them wasn't in Dragon's Maze.

Kiora Atua was one of two Planeswalker characters created especially for Duels of the Planeswalkers. Ral Zarek was the other. We decided that we would find a home for each in card form when the time was right. We weren't going to force them in but rather be patient and wait for the proper moment. That moment for Ral Zarek was on Ravnica. It's his home and he is as Izzet a Planeswalker as they come. For Kiora, Ravnica was an odd fit. She doesn't match Simic at all and a city plane seemed an odd place for a seafaring, Leviathan-loving Planeswalker to visit. You all will get a chance to meet Kiora when the time is right.

Fblthp is on the opposite end of the spectrum. He wasn't made with any future planned. He was just a cute creature that showed up on a card in Return to Ravnica and then on a second card in Gatecrash. We weren't trying to create a pattern. In fact, we had no idea how beloved the character would become. By the time we realized that Fblthp should be on a card in Dragon's Maze, it was far too late to put him in. For what it's worth, if I was willing to use our time machine willy nilly, I'd go back and put him on a card.

"Bueller... Bueller... Bueller..."

That's all the time we have for today. I hope my look at what wasn't in Dragon's Maze (and why) was insightful. I'm curious to hear from all of you. Was there anything else you expected to see? Was there something we did include that you wish we hadn't? Any other comments on what I said today? Well, I'd love to hear your comments. Write me an email, respond to the thread (which I do read every week, by the way), or send me a message through my social media (Twitter, Tumblr, and Google+).

Join me next week when I open some Gates.

Until then, may absence sometimes make the heart grow fonder.

Drive to Work #33—Future Sight, Part 2

It's time for Part 2 of my look at Future Sight design, and as a special surprise I'm carpooling with Matt Cavotta, one of the designers of Future Sight.