Pattern Recognition #188 - Set Design

Features Opinion Pattern Recognition


18 March 2021


Hello everyone! This is Pattern Recognition, TappedOut.Net's longest running article series as written by myself, berryjon. I am something of an Old Fogey who has been around the block quite a few times where Magic is concerned, as as such, I use this series to talk about the various aspects of this game, be it deck design, card construction, mechanics chat, in-universe characters and history. Or whatever happens to cross my mind this week. Please, feel free to dissent in the comments below the article, add suggestions or just plain correct me! I am a Smart Ass, so I can take it.

We've all made cards of our own. I hosted a series of analytics on cards you people created and submitted to me a couple years ago, and the Card Creation Challenge thread here on TappedOut is the largest and most constantly posted to each day. It's only natural that people as enamored with the game as us seek to experience the joys of creation ourselves.

But Magic is not just a game of cards, or a game of sets, or a game of decks and singles. It is a deeply interconnected game that creates potentially exponential or algorithmic growth in the way things look at each other and it is quiet easy for something to slip through the cracks, be it small or large and make a mess of things. That the game has been, with the past couple of years I hope being an exception and not the new norm, a well balanced and well oiled machine that is pretty stable and can be added to without much issue.

And Wizards, again for the most part because they're only human, makes this look easy. Well, easier than it should be. I've pointed out in the past how the various formats interact with card design to allow for cyclical balance in Standard, where eternal formats can have options long forgotten be available to deal with problems, or be countered by newer cards.

Set design is something that takes 18 to 24 months to complete, from conception through to release. It's a multi-stage process that Wizards has been fairly open about in the past, not seeing it as some sort of secret. I would direct you to the Nuts and Bolts archive of Mark Rosewater's Making Magic for how to design sets. Yes, it's six years old at this point, but the only real major change to the process since then has been the move away from a two-team development process - Design and Development - to a three-team process in Vision, Design and Play.

Now, if all you like it to enjoy the cards in hand and playing with them at the table, on MTGO or Arena, that's fine. There's no shame or anything like that in simply enjoying the end product. You may not find this article to your tastes, as I'll be summarizing and expounding on this process. After all, you don't need to know how an engine works to enjoy driving a car. It may help, but it's not required.

Anyway, when a set is first put down, a lot of ideas get hashed out. Ideas for themes in the set, the nature of the Plane they want to use or explore, mechanics and just random, off the wall suggestions that have people laugh and file them away for the next Silver-border set whenever that comes. This is the Vision step in the current development schema, where they lay out the general outlines of the set and where they want it to go.

In this step, we see the start of the top-down or bottom-up design. A top down vision for a set starts with an idea for the set. Kaldheim, for example, much like all the sets based around real-world mythology including Theros and to an extent, Kamigawa, is a Top Down set. It starts with an idea, and the rest of the set follows suit, molding mechanics and presentation around this idea.

When we talk about Kaldheim - as it is the most recent example - it started with the idea that we were going to be building a Norse set, so research would be done into what it meant to be a part of that mythology and how the cards and mechanics could be created to fit that.

On the opposite end, a Bottom-Up design starts with individual cards and mechanics first, then they are fit together like a jigsaw into a more cohesive whole, such as with Kaladesh and its focus on Energy and Artifacts. The whole rebellion and spirit of innovation thing came afterward to help bind the cards they made together.

Neither method is right, and neither method is wrong. It's just two different ways to achieve the same goal.

With a Vision document in place, as well as some examples of cards and mechanics to show what the set should be directed toward, Design takes over. Design does a lot of the heavy lifting in this process, and this is where a lot of the problems get found and resolved.

Design takes the document handed to them by Vision and begins to work it, weaving ideas, feedback and further creative processes into something resembling a final product. The majority of cards in a set are created at this point, as mechanics are refined, tossed out, accepted, refined again and cards that use them go through the same process.

The important part here is iteration. That nothing is set in stone. Each card is an idea, where costs are figured out, where baseline decisions can be made about mechanic distribution and what can go at what rarities. At this point in the process, they would check back with the Vision team in order to make sure they are on track with what Vision was hoping for.

Design tests and measures cards out against both other cards in the set, against the Standard rotation they will be in (assuming Standard is even a concern, such as with supplementary sets), and in, hypothetically speaking - other formats.

This is also the point where reprints become examined, not only because there are holes in the set that need to be filled, and can be filled by pre-existing cards, but also when convergent evolution in design create a card that does what a previous card does to the point where they have to consider if they should be merged into one, or left as a

And given that Commander is apparently the driving force behind Magic right now, not Standard, you can see even that being abrogated. Draft? Standard? Modern? How much work would that be? That's a lot of work, and that's where the third section comes in.

Play Design are the people who don't make cards, but rather they take what they are given by Design and run them through their paces both as a draft format and against other sets currently in development as to see how the Standard rotation would work out. They then give feedback to the Design team about what's working and what isn't. Design doesn't always get context, especially when dealing with cards from other sets in development. This can lead to interesting back-and-forths where Design says something is working as intended, but Play shoots back that it may be working by itself, but against other things, there's this horrible combo that results in instant loss for the player that uses them, so can they please fix this one line?

Not that I'm citing a real event here, just showing how the two sides interact.

Vision lays down the groundwork and the rules. Design builds the set, Play tries to break it. That's in general how the system is supposed to work, and how it should work. Now, of course, we can all name a few of the horribly broken cards over the past few years that should never have seen the light of day....

Look, I'm not calling out names here, Oko, Thief of Crowns. Why would you think that?

Part of the problem is that there are only so many many hours to go into testing. Creative work can be very difficult, even for the best of us. I struggle to put out two thousand words for you guys each week, and I can assure you that this is a far easier project than building a set of over 260 cards into a cohesive whole. Because of this, errors happen, and Wizards does try their best to catch them before they reach the printers, but sometimes, like the Corpse Knight printing where the toughness was accidentally set to three instead of two (I have that, by the way), are out of their control. But interactions like Felidar Guardian and Saheeli Rai, a 2 card infinite combo that can go off on turn 3 or 4. In the same set. Those should have been caught, but Wizards has admitted that a last minute change to the Guardian is what caused the problem, something that should have been caught, but wasn't. So it ate a Standard Ban, one of the few in Kaladesh and Aether Revolt that wasn't the result of Energy or Artifacts.

Wizards has changed how they develop sets over the course of the game, from the wild-west Play Test Groups of the early game being given leave to make their own sets to be published, to the R&D Department to Design and Development to the more current trifold system. It, like the game itself, is a work in progress.

Now, you may be wondering why I've talked about all this, and what it has to do with what I mentioned a couple of weeks ago. Well, the short answer is that one of my running projects has been to design my own Magic Set - because I can - and I'm also one of those types of people who would rather show their work, warts and all, rather than just toss in a few dozen cards that are 'awesome' and call it a day. No, sets have to be designed to integrate into existing sets, be interesting and attractive without breaking the player's opinions of the set and driving them away. New Phyrexia. This will not be a short project, or a quick one. I will be replicating the work of dozens of people over the course of two years by myself. I've already done some preliminary work, but again, it's a matter of showing my work rather than just presenting a finished product.

Who knows, maybe it'll be my resume when Wizards hires again! ;)

Join me next week when I talk about a different subject. What one? I don't know yet.

Until then please consider donating to my Pattern Recognition Patreon. Yeah, I have a job, but more income is always better. I still have plans to do a audio Pattern Recognition at some point, or perhaps a Twitch stream. And you can bribe your way to the front of the line to have your questions, comments and observations answered!

This article is a follow-up to Pattern Recognition #187 - Tuck The next article in this series is Eternal Archive #189 - A Year in Review

berryjon says... #1

No update for this week; 25 Mar 2021.

I have something special lined up for next week.

March 23, 2021 11:43 a.m.

Please login to comment