Welcome to Exploit Week where we're going to focus on the favorite mechanic of Silumgar and his followers. During the preview weeks, I touched upon how we created exploit and talked briefly about how we had messed around with a similar mechanic during Onslaught design. I think I gave the impression that the two designs were linked—that we created it during Onslaught design, didn't use it, put it away, and then pulled it out of a drawer during Dragons of Tarkir design. The truth is that the creation of the two mechanics actually started from completely different places and only lined up through coincidence. This phenomenon is known as parallel design and happens quite a bit. I was going to use today's column to explain the phenomenon and walk through the types that exist and what lessons we can take away once we're aware of it.

Parallel Design can work in several different ways. Let's walk through them.

Parallel Design #1—Same Input

So who designed the dash mechanic? I did. And Sam Stoddard did. Did we work on it together? No. Interestingly we created it independently at the exact same time. Let me explain. In design, the lead designer of a set will give out homework for the design team to work on in their own time and bring to the next (or sometimes a more future) meeting. The lead designer usually is very specific on what they are looking for. Sometimes, depending a lot on where in the design you are, it will be to fill a specific hole; while in others, it might be a call for designs within a chosen area.

Goblin Heelcutter | Art by Jesper Ejsing

For the first Dragons of Tarkir homework, Mark Gottlieb, the lead designer, wanted us to take stabs at faction mechanics. Remember that Dragons of Tarkir design started about four months into Khans of Tarkir design and many months before Fate Reforged design started, meaning we were on the hook for five faction mechanics and a variant for morph. Fate Reforged would later borrow two (dash and bolster as it worked out). The first homework assignment spelled out what we were looking for. Khans of Tarkir had its faction mechanics and our job was to make the accompanying mechanic play nicely with the Khans version.

This means Sam and I sat down with the same goal: make a mechanic for the black-red faction (leaning toward red) that played nicely with raid. The fact that we came up with the same mechanic isn't a great surprise. We were given the same parameters and followed similar logic. This doesn't mean two designers can't often come up with different designs having the same input, but the fact that we occasionally design in parallel makes a lot of sense.

This type of parallel design is the most prevalent. We work collaboratively, which means a lot of time is spent within the group trying to understand what works well and what doesn't. Similar lines of design thought mirror the group-think that tends to happen with constant group interaction.

My favorite story about this type of parallel design happened during Innistrad design. I started by having my team design top-down cards based on horror tropes. For the first meeting, I didn't even give them any of the tropes, I let them design with whatever they came up with. The first meeting had us sitting around, walking through what each person had designed. Every single person not only designed Wooden Stake but had it do the exact same thing, an Equipment that boosted the power of the equipped creature and killed any Vampires they damaged. Whenever someone asks me who designed Wooden Stake, I answer "Everyone."

Parallel Design #2—Same Idea, Different Time

When Worldwake came out, I wrote a card-by-card design story article ("Worldwake Me Up Before You Go-Go"), in which I talked about the design of a card called Bestial Menace.

Many years earlier, influenced by a card called Cone of Flame (from the set Weatherlight), I created a card called Cone of Creatures which made a 1/1, 2/2, and 3/3 token. I tried putting it into various sets but the three different types of tokens caused confusion and it got pulled each time. I even publicly talked about the card back in 2002 in an article I had written about tokens. ("Tokens of My Affection").

The card finally came out in Worldwake (aided by the now use of token cards in the booster packs), so in my article I explained how for years I had tried to get this card into sets and had given up. I assumed Ken Nagle, the set's lead designer, had picked up the gauntlet and put Cone of Creatures into the set. After I turned the article in I got a note back from my editor, at the time Kelly Digges (who has since moved to editing, and then moved again to the creative team and as one of his jobs oversees "Uncharted Realms"). Kelly said I had gotten it wrong. He had made the card. He had never heard of Cone of Creatures. It turns out that Kelly had walked through a similar design path as I had. He had just done it over ten years later and without any input from me.

This happens constantly. So much so, that I will often see a card in a file that I know I had at one time designed and have to ask, "Did I make this?" I even have the following problem: I'll come up with a card idea; someone else will say, no we've already done that and mention the name of the card along with the set it came out in; and after a beat, I'll say, "Oh yeah, I made that."

Another classic example of this type of design was the miracle mechanic from Avacyn Restored. During Tempest design (my very first design lead) I got the idea of cards that did something when you drew them, but couldn't figure out a way to execute it and finally gave up on the idea. Fifteen years later, Brian Tinsman, unaware of my attempt during Tempest design, attempted the same mechanic but figured out a way to actually make it work. (Being fifteen years later also helped a bit as Magic has expanded what it's willing to do in black border).

The key to this type of parallel design is that some ideas are just good, strong, solid designs that are obvious enough that multiple designers will stumble over it in time.

Parallel Design #3—Repeat Processing

The Jeskai needed a mechanic in Khans of Tarkir. As we normally do, we set a bunch of parameters. Blue and red are the two colors with the highest percentage of spells, so possibly something that could go on instants and sorceries or care about instants and sorceries. The Jeskai had the flavor of Shaolin monks so we wanted them to have some ability to fight. Jeskai was the faction of cunning so we knew we wanted a mechanic that could be tricky. Taking in all of the restrictions, I made a suggestion for a new mechanic.

The mechanic boosted creatures +1/+1 for any spell you cast. Too strong. It then boosted creatures for any instant or sorcery you cast. Too weak. How about any noncreature spell? For a while, we even toyed with Khans of Tarkir caring about noncreature spells while Dragons of Tarkir would care about creature spells. Anyway, we fiddled with the mechanic until we finally got it to the prowess mechanic.

Jeskai Sage | Art by Craig J Spearing

Then one day in the Pit, Jon Loucks was playtesting Khans of Tarkir and he commented how he was happy we used his energize mechanic that he had submitted in the second Great Designer Search. The interesting thing was that we didn't start with his mechanic but had evolved into it. In fact, we went a step further, fixing a problem I had noted when I judged the mechanic during the GDS2.

The reason this category is different from the last was that at least one or more of the people making the newer cards had interaction with the older ones, meaning there was some influence even if subconsciously. I, for example, have worked on so many sets that I can't possibly remember everything but as I work on mechanics or individual cards I will often realize that I am covering ground we have explored before.

In many ways this category is less parallel than the others because they're correlative, but these type of situations play out a lot like the last category—where new things get created that only later are recognized to be something we've tried before.

Parallel Design #4—Disparate Approaches

As this is Exploit Week, it seemed only right to talk about the creation of the exploit mechanic. Let me start by walking through the two different designs that led to exploit. The first happened back in 2000, while we were working on Onslaught. The design of Onslaught block had a lot of twists and turns (one day I'll write an article about it; I've already recorded a podcast about it which you can listen to here), but the important point for this story is that I was pushing hard for the block to have a tribal theme.

This was back before the tribal theme was a proven thing, so I was making the case for why I believed players would enjoy it. Part of making the case was designing enough cards and mechanics that cared to make it relevant. I started by exploring all the obvious avenues. I made creature lords, cards that targeted certain creature types, and cards that needed a threshold of that creature type to work. I was making a lot of cool designs, but I found that the effects I was generating were getting a bit on the complex side.

For example, the spells either had to count the number of a particular creature type, had to target just that creature type, or just tended to do things that took a lot of words. I was eager to find a mechanic that would allow me to do simple basic effects, but in a way that mattered in a tribal environment. At one point I came up with a novel approach—a split card where half of the card was an instant or sorcery, and the other half was a creature of one of the supported creature types. The idea was that I could have a Zombie or a zombie-like spell. There was just one small problem—split cards can't support permanents.

Lifebane Zombie | Art by Min Yum

So I did what we always do when an idea doesn't work within the rules, I tried to figure out a different way to do it. Okay, what I wanted was a card that could be either a creature or a spell. It could be a modal spell and one of the modes could make a token, but that would greatly cut down the amount of design the mechanic could have. I then explored creature cards that I could spend mana on and discard to generate an effect. We'd messed around in this space with the channel mechanic in Saviors of Kamigawa and I wasn't happy with how it played out there.

Then it dawned on me—what if when the creature entered the battlefield, you could sacrifice it to generate the effect? That created the modal sense I wanted, but I wasn't done yet as I was trying to make a modal mechanic. What if, instead of sacrificing that creature, I could sacrifice a specific creature type of which that creature belonged? Here's an example: (converted to modern templating)

Plague-filled Zombie


Creature — Zombie


When CARDNAME enters the battlefield, you may sacrifice a zombie. If you do, target non-black creature gets -4/-4 until end of turn.

The idea was that this mechanic allowed you to convert any Zombie into a kill spell. We also toyed around with the idea that you could sacrifice any number of Zombies and, for each one, generate the effect. So what happened to this mechanic? Onslaught design didn't need it. Every design always makes more than is required, and ideas get culled based on the needs of the set.

Let's now flash forward from 2000 to 2013. Dragons of Tarkir needed a blue-black mechanic that played nicely with delve. To accomplish this, we focused on mechanics that got cards into the graveyard. We looked at discard and self-milling, but in the end we liked some form of sacrifice. The big question was "Why did this mechanic require you to sacrifice creatures?" We explored different answers, but the most compelling was that it generated a spell-like effect. Sacrificing a creature is a big deal but, if you were getting what seemed like a card out of it, it might be worth the trade.

In the end, we put the mechanic on creatures so that if you didn't have any other creatures you could always sacrifice it. We made the creature a smidgen bigger on purpose, to encourage you to sacrifice other creatures. We also designed smaller creatures that gave you benefits for sacrificing them, most often by having death triggers. Unlike with Onslaught, this time the mechanic made it to print.

This last category is the strangest as it creates similar results from different inputs. As the exploit example shows, each side got to the same mechanic but through a completely unique path. Onslaught got to the mechanic through a tribal need. Dragons of Tarkir got there through a need for a mechanic that sacrificed. They literally came from opposite ends. Onslaught said, "Oh, we could also make it a creature," while Dragons of Tarkir said, "Oh, we could also make it a spell."

These disparate designs happen a lot more often than you might think. My best guess on how it happens is that certain concepts just lend themselves well to Magic game play and, no matter how you approach them, you end up discovering similar design space. Or in other words, as you start exploring certain aspects of design, no matter how you start, they gravitate in the same direction through iteration.

So What Does This Mean?

The reason it's important to understand parallel design is the following:

#1—Design is More About Finding Elegance than Novelty

The more I've done Magic design, the more I've come to understand that good design isn't about discovering what hasn't been done, but rather figuring out where the elegant part of the game lies. The metaphor I like to use is that of sculpting. Design is often faced with giant blocks of stone. Our job is to find the statue within the stone. We're not creating the statue as much as we are uncovering it. Twenty-two years in, there's so much craft to Magic design that it's more about understanding and following the rhythm of the game than it is about discovering uncharted territories.

#2—To Create the Future, You Must Understand the Past

One of the biggest resources I have to offer the design team is my twenty years of experience. It's rare that we travel down a new area of design that we haven't at least dipped our toe into. And sometimes, we've spent a lot of time and energy mapping out a design vein even if we never ended up using it. Often, when we are exploring an area, I have my design team research what we did when we tackled the theme previously. Magic's long history is a valuable tool that we would be foolish not to use.

#3—The Key to Finding Answers is Defining Your Parameters

Parallel design reinforces the idea that good design will push in a similar direction. What this means is that the key to finding answers in design is better understanding what it is you need. Adopting this concept has changed how we problem-solve in design. Rather than jump to the brainstorming, we spend more time clearly defining the parameters.

#4—The Quality of an Idea is Dependent on the Environment it Exists In

Another important take-away from parallel design is that an idea doesn't exist in a vacuum. In order to judge it, you have to see the environment it's being proposed in. There are mechanics that can shine in one block that would be miserable in another. Context is important. Synergy is important. Flavor is important. A mechanic doesn't get put into a set because it's a great mechanic, it gets put it because it's a great mechanic in that set.

#5—There Is No One Right Way in Design

The final lesson from parallel design is that the path to a good design is not a singular thing. Take any good idea and there are a multiple ways to get there. The key is not dwelling on where you start but rather on where you end.

Parallel Parking

I hope today's column allowed you to take a look at a different aspect of design, one that is much less visible from the finished product. As always, I am eager to hear any feedback on today's column. You can write to me through my email or through any of my social media (Twitter, Tumblr, Google+, and Instagram). 

Join me next week when I go back to the very beginning.

Until then, may you see the parallel design in your own life.

"Drive to Work #218—Innistrad Cards, Part 3"

This is part 3 of a 5-part series talking about card-by-card design stories from Innistrad.

"Drive to Work #219—Green-White"

This is part 5 of a 10-part series on the two-color pairs, and their philosophies and mechanical overlap. Today is the final of the allied-color pairs, Green-White.