Wednesday, May 16, 2018

GDS3: The Design Test

Jeremy Geist has offered us some insights into how he approached the Design Test portion of the GDS. Click through to find how he designed his ten cards.

By Jeremy Geist

The Great Designer Search 3 is finally being revealed to the public! I wanted to take the chance to expound on some of my cards, either because I think there’s an interesting story behind them or just to explain my thinking.

This first entry talks about my submission to the design test. (The rules of the design test are here.) There were lots of ways to approach it; Mark Rosewater’s suggested approach was to decide on card types for every color combination ahead of time and try to think of good fits for them later, while first place scorer Jay Treat created two separate files and submitted one. I also saw someone on the subreddit talk about how they made five cards for each color combination and picked out the ones they liked the most.

My approach was a cruder than any of those three. I had a couple of ideas that I liked, so I stuck them in and slowly built my file around them. Once I came up with a card I liked, it was pretty much locked in and I stopped thinking about replacements for it. My habit of stopping once I have a solution that’s “good enough” has been a weakness of mine as a designer, and one that I think participating in GDS3 has helped me address.

The design test was really clever because it challenged participants on multiple axes in an elegant way. There’s room for you to show off your cool, innovative designs, but once those are in you need to fill holes that have increasingly narrow requirements. The 1 through 10 ranking also checks how good you are at self-evaluation, which is an important skill.

(Editor's Note: Jeremy's cards and the judges notes on them can be found here

Card-by-card breakdown 

1. Pick Your Poison

When I need to brainstorm, I take long walks. Pick Your Poison was the result of one of these. I was having trouble coming up with a green/black sorcery that I liked: I had a few ideas jotted down, but most of them were two effects stapled together and didn’t do anything interesting or innovative.

While I was on my walk, I thought about what an interesting structure for a card would be, and I came up with the idea of a card that acted like a store in an RPG. You got a certain amount of “currency” and could buy various effects with it. Then the structure I submitted occurred to me and I thought, “Wait, can you do this? I’m pretty sure there’s nothing stopping me from doing this!” I was so pleased with it I ranked it first immediately. Besides looking splashy, it also plays with weighted modes, something Magic hasn’t really done before.

PYP isn’t a perfect card by any means. With the 1/2/4 structure it’s pretty much a four-choice modal spell, and as Eli Shiffrin pointed out, targets are chosen before the spell resolves so you can’t put your counters on your snakes. But what I would actually change in retrospect is the color combination. I’ll talk about this when I get to number 7.

2. Bewitch

Bewitch arose from a series of removal Auras centered around black. My first pass looked like this:

Mummify (uncommon)
Enchant creature
Enchantment – Aura
Enchant creature
Enchanted creature can’t attack or block and its activated abilities can’t be activated.
2WB, sacrifice Mummify: Destroy enchanted creature, then return it to the battlefield under your control. It’s a black Zombie in addition to its other creature types.

I then realized that destroying a creature and immediately getting it back was basically a “gain control” effect, which is in blue. So I switched into a blue/black Aura that looked like this:

Geralf’s Preparations
Enchant creature
Enchanted creature gets -5/-0.
5UB, sacrifice Geralf’s Preparations: Destroy enchanted creature, then return it to the battlefield under your control. It’s a black Zombie in addition to its other creature types.

But I was still unhappy with it. I thought about it for hours, trying to crack the precise combination of blue and black Aura effects that could work together well, when the final version struck me like lightning. I immediately jammed it into the number 1 slot, where it stayed until Pick Your Poison exceeded it.

3. Mama Bear

I didn’t intend on making a hybrid card with Mama Bear; the original version was something like a 3/3 for 1GW. Midway through the week, I received an e-mail stating that deciduous mechanics were all right to use, and I realized that hybrid mana would allow me to make the stats better for common. The ability works fine in green or white IMO, though certainly white has more claim to being the small creature color.

There aren’t really any bears in white, but we weren’t being judged on names or typing so I went with the concept that resonated the most.

If I were to go back and change this card, I’d definitely follow Melissa DeTora’s suggestion of putting the counter on the small creature.

4. Pam, Shapeshifting Planeswalker


Shortly after the round ended, Mark Rosewater tweeted that a lot of submissions had way more text than could fit on a card. My mind immediately went to Pam. I had tested text length on my commons, but her first ability (designed so it didn’t have to target, which is mandatory for +1 abilities) pushed the card into something like eleven lines of miniscule text. Theoretically, if I made the +1 ability a 0, changed it to “becomes a copy of target…” and gave her a different +1, the card would probably be fine.

Planeswalkers are really hard to design.

5. Sign In Someone Else’s Blood

My main goal with this one was to demonstrate that I could design cards that were important to have in sets at common but that weren’t splashy or exciting. The “D+ level sacrifice to draw stuff” spell is a mainstay of Limited formats, and I figured this was a way to show off my breadth while creating an effect simple enough to be a two-colored common.

I didn’t realize, when I designed this, that black has the second-most creatures with high toughness, so this is closer to just drawing you two cards than I thought at the time. That said, I’m still pretty happy with the elegance of it, and zapping your 5/5 for three damage means it can’t block that turn.

This card also has my favorite name of my submissions this round.

6. Whimsical Djinn

Whimsical Djinn was the first card I put on here. It was originally a mono-blue shapeshifter that I was going to submit to Hipsters of the Coast’s practice challenge, but I messed up on the deadline for that and decided to submit it to the real thing instead. I had it more strongly themed as a djinn who gives you three wishes and returns to its lamp, then I discovered Djinn of Wishes and felt stupid.

Melissa DeTora correctly pointed out that the current options are way too strong, so I think in the end it would have been better to make the Djinn at rare and tone down the abilities by a whole lot. Making busted mythic rares is a recurring weakness of mine in GDS – I have the most trouble with the mythic rarity in general.

I’m not sure if the card wouldn’t just be better if it picked an ability at random and didn’t use the Demonic Pact stuff, but I like that you have more information about what’s going to happen the longer you have him out, and it allows you to construct more of a strategy.

Incidentally, there are three abilities because it allows you to roll a d6 to decide which one happens, no matter how many have already been chosen.

7. Mix//Match

Mix//Match was the last design I put on here, and I was so jazzed I thought of anything at all I put it at a generous number 7. This is clearly my worst design in the design test; part of it is because I painted myself into a corner by having to do a red/white sorcery, possibly the worst color combination to do sorceries with.

If I had thought a little clearer, I could have made this use Pick Your Poison’s weighted modal ability, because that could work in any color combo, and used BG for some other thing, since those colors are much more sorcery-friendly.  (The name for the R/W modal spell would have been Select Your Fighter.)

8. Sorin, The Prequel

It turns out that Ryan Siegel-Stechler and I turned in planeswalkers with two exactly identical abilities, which is hilarious.

9. Remember Your Charlemagne

I filled out the slot for a white/blue instant almost immediately. But it wasn’t this card, it was:

Geist’s Mercy (common)
Return target creature to its owner’s hand. If you are that creature’s controller, draw a card.

I had to change it after I created Sign In Someone Else’s Blood, since I didn’t want to have the same rarity among card types and because it was also a card draw spell. But I still think Geist’s Mercy was a much better design.

Ipaulsen on Goblin Artisans pointed out that Remember Your Charlemagne should have granted flying, and I’m kicking myself for not doing that.

10. No Humans Allowed

I came very close to calling this one Werewolf Bar Mitzvah.

I’m happier with this card than the three before it, but this was my riskiest design so I ended up ranking it tenth. This was also one of the last cards I came up with; other things I tried in the R/G enchantment slot were related to fighting or destroying noncreature permanents. One of them made an Elemental Boar token, and I probably would have gone for it if my other enchantment wasn’t also at rare.


  1. The test structure is complicated, but it was a really intriguing microcosm to capture the kinds of things you have to do to fill a set.its funny because it doesn't really read like that, it feels like a random challenge, but the seemingly random limitations all add up to create a process that looks like it feels like building a mini set.

    I think that while you view it as a weakness, it's hugely important to be able to shelve a design at some point, sometimes even if you don't think it's perfect, because what's perfect is a very vague and nebulous concept that's hard to quantify and achieve. And I say that as a self described "perfectionist" as well, as I believe there is a "perfect" iteration in most things. Even in real design you have to do it eventually, and I'm sure many cards that make it to print the designers would have liked further iteration on. I'd actually consider it a good skill and one I envy if you can manage to put down a design at some point and say "okay, this is done. I have to move on to the next one."

  2. I was really hoping Pam was a McElroy reference!!! :3

    1. My Jefferson Tallpipe planeswalker was quickly scrapped.

  3. Those are really nice. I love the concept of "No humans allowed", even if I agree it's not that plausible to print. And the "choose modes that add up to 4" is a great idea, I'd like to see other spells with that.

    I like the Djinn too.

  4. Nice piece, Jeremy.
    I particularly appreciate the detail of being able to roll a d6 for the Djinn at all stages.