#515: Building New Worlds
This is a companion podcast to one I did about designing sets that revisit worlds we've been to before. In this podcast, I talk about designing for a new world.
Posted in Making Magic on March 5, 2018
Four weeks ago, I walked you through the first trial (the essay test) of the Great Designer Search 3 (GDS3). Two and three weeks ago, I walked you through the second trial (the multiple-choice test). That means today it's time to walk you through the third and final trial—the design test. I'm going to show you what the test was, then I'll walk you through how I would tackle it.
This time, 7,800 people signed up expressing interested in GDS3; 3,085 passed the first trial and were invited to take the second one. Ninety-four candidates, each with a score of 73 or better out of 75, advanced to the third trial. (Three got perfect scores.)
Let's begin by showing you the design test.
For Trial 3, the design test, you will need to design ten cards that meet the following criteria:
All the cards will be two-color and each of the ten two-color combinations (listed below) need to be represented.
Each of the following five card types (creature, enchantment, instant, planeswalker and sorcery) needs to be represented twice, and never on the same color.
Each rarity (common, uncommon, rare, and mythic rare) must be represented on at least two cards.
Submit your cards in order of quality of design from what you consider your best design (first) to what you consider your worst design (last).
Cards must be submitted in the following format:
Card Type – Subtype
Power/Toughness [IF CREATURE; SKIP OTHERWISE]
Two examples solely for the purpose of card formatting (one creature, one spell):
Lab Experiment (common)
Creature – Bird Elephant Mutant
That Should Hurt (uncommon)
CARDNAME deals 3 damage to target creature. That creature's controller discards two cards.
Here are the abbreviations for how mana is written:
W = white
U = blue
B = black
R = red
G = green
Number = number of generic mana (i.e. 2 for two generic mana)
Here is the mana order for two-color spells (you must have one design of each):
You have until 11:59 pm PT on Sunday, February 4, 2018, to submit this form. Good luck!
I then sent out some follow-up instructions to the 94 contestants taking the trial:
Before I explain how one would tackle a test like this, I want to take a moment to explain why the test is made the way it is. The goal of a design test is for us to see several things:
The way the test works is that we give the players a bunch of flexibility, but then put in limitations that will start restricting them as they design. This will allow them to have a lot of freedom early, to show off something they've previously made, for example, but then force them to demonstrate that they can design to constraints. As you will see, this test does a lot to let us see how they tackle design problems and how good they are when forced to design in a corner. I designed the previous two Great Designer Search design tests, but this one was made by Erik Lauer. We limited them to ten cards because we have a lot of tests to grade, and ten cards felt like enough to show what they could do while still being few enough that we could grade 94 tests.
For those interested, the way it works is I graded all the design tests. I then read all the essays of the ones that did well on the design test. I used the essay tests to narrow down the field to a small enough group that I could then get numerous other R&D people to read them, including all the judges and the person who would manage the winner. From that we got to our eight finalists. (You'll get to meet them all on March 9.)
A quick caveat before I begin demonstrating how I'd do the test: the cards I'm going to be mentioning were made up for this article; nothing previously existed or was taken from any design I am currently working on, so don't look for any hints of Magic's future from these card designs.
Back to the design test. So how exactly does one handle a test like this? The first thing you have to do is handle what I call "the grid." Any design has a structure to it, and the structure will dictate choices you have to make, so one of the earliest things you want to do is map out the choices so that you understand what has to be designed. With a normal Magic set, this requires making a design skeleton. (I talk about this in depth in one of my "Nuts & Bolts" columns.) For this trial, it means mapping out the card types for each color combination.
Let's start by listing our ten card slots (note that this is the order in which we list the ten color pairs, along with the order we list the two colors in their mana costs):
There are five card types (creature, enchantment, instant, planeswalker, and sorcery) that have to each appear on two card types and can't appear in the same color. That means, for instance, if we choose to make the white-blue card a creature, white-black, red-white, and green-white can't be creatures. As we start selecting things, it will begin impacting what other things must be. I, therefore, would start by mapping out what card type each color combination will be.
The place I would start is by seeing if I have any design already in my pocket that I'm very proud of. Have I made a cool two-color design? Odds are, if I've been designing enough cards in my spare time, the answer is yes. Depending on what you have and how many different card types they cover, we can start filling in our grid. Once again, I'm making up all these card designs on the spot, but let's assume I already had a few ideas to use as a starting point. I'm going to pretend that I had a cool black-green creature, a red-green enchantment, and a white-blue instant.
To help myself out, I'm going to make a chart.
Now I'm going to fill in the chart based on the three cards I "already had."
Once I fill in a color pair, it does two things: One, it puts a check on one card type and eliminates the other card types as a choice for that color combination, making Xs horizontally. Two, it starts eliminating other options for the second card of the card type, because once we pick a combination we are eliminating every other combination that shares a color with it. Thus, we add Xs vertically. Mapping this all out makes our first decision for us. There is only one other option for an instant, black-red, so we fill that in.
Our next choke points are creatures and enchantments. Our first creature is black-green; the only two options left are blue-red and red-white. Our first enchantment is red-green; the only two options left are blue-black and white-black. I decide to make a blue-red creature and a white-black enchantment.
We now have four color combinations that aren't spoken for yet: blue-black, green-white, red-white, and green-blue. Two have to be planeswalkers and two have to be sorceries, but neither can overlap in color, meaning we have only one way to divide the combinations. Green-white and red-white have to be on separate card types, as do blue-black and green-blue. Green-blue can't overlap with green-white, so that means blue-black goes with green-white and red-white goes with green-blue. My one choice is which side goes to which card type. I've chosen to have red-white and green-blue be planeswalkers and blue-black and green-white be sorceries.
Here's how it all breaks down:
We're not done yet, though, as there's still another layer to add to this. The test requires us to make use of all four rarities (common, uncommon, rare, and mythic rare). Each one has to appear on at least two cards. Although the test doesn't specify that you can't use one rarity four times, I would lean toward using two twice and the other two three times.
One of the ways to assign rarities is to just wait and see what cards you make early and tick them off as you use them, but I believe that knowing ahead of time what you're aiming for will be a help rather than a hindrance. As I like to say, "restrictions breed creativity." Having a more focused target will make it easier to design toward that target. Therefore, I would pre-assign the rarity to each card, knowing that if in the act of designing I stumble upon something cool in another rarity I have some freedom to swap.
Here's how I'm going to assign my rarities. The two planeswalker cards need to be mythic rare. I would like to show my ability to design a non-planeswalker mythic rare, so another slot will be mythic rare, in the color not hit by the two planeswalkers. Also, I'm going to make each pair of card types, other than planeswalker, in different rarities. Finally, to help spread my designs apart, I'm going to divvy up my rarities in different colors, so I'm not trying to make two red common spells. Here's what I come up with:
Obviously, I wasn't able to hit all my goals perfectly. There's a little overlap in color, but meeting my other goals makes that impossible not to do a little. Planning allows me to have a good grasp on what I'm aiming for. I can't stress enough how important it is to figure this out first, because it's very easy with this test to paint yourself into a corner and have to throw away designs.
The next step is to do a little research and see what's been done in this space before. Remember, the goal of this design test is to stand out, so what that means is you don't want to recreate something that's already been designed. I'm just going to do this with one of the cards, but if I were taking the test, I would do this with all ten.
Let's look at common white-blue instants. (Yeah, I know I said above I had a cool white-blue card, but I don't, as I have zero cards walking into this test. The contestants should have some cards pre-designed that they can use. Note that we required that those designs were solo designs made only by them.)
It turns in all of Magic's history, there are only seven white-blue common instants, two of which use hybrid mana. Personally, on this assignment, I'd steer clear of hybrid mana. I know there's some pressure to stand out and do something different, but hybrid design space is pretty narrow, so finding something that hasn't been done before is going to be very tricky, especially at common. I'm in the camp that you stand out best by making cool cards, not pushing the boundaries of the test. When I assign homework for a set, I'm not looking for my designers to match the letter of the assignment but not the spirit.
Three of the seven cards are basically counterspells, two of which gain you life. Two are bounce spells. One is a creature-pumping spell. The last is a flickering spell.
My first gut instinct is to see if we can find some common design space that avoids what's been done before. Luckily for all the designers, I made a great tool to help with this last summer: the mechanical color pie article.
What abilities are primary in white and primary in blue?
Let's start with white:
The first thing we can do is eliminate everything that isn't an effect you can do on an instant, isn't an effect that you can do at common, or is an effect already done by one of the seven cards above.
Now we do the same for blue:
This is what we have after we remove the things inappropriate for instants or commons and abilities used on any of the seven cards above:
From these lists, I'm going to cull down to what seems the most interesting to work with.
Now we have some primary effects in each color that haven't been done on a white-blue common instant before. Before I start trying to mix and match the abilities, I would quickly go look at the white-blue uncommons that have been made, because it's possible to come up with a common idea that's very close to something done in uncommon. It turns out there are twelve such cards, a few with effects to watch out for.
I don't have time to walk you through the entire test. I do want to point out that not every multicolor card wants to be two effects stapled together. (I wrote an article on multicolor design called "Midas Touch" that walks through the various types of multicolor cards one can create.) We're making a common card, though, so it limits the kind of design we're going to make. What I'm looking for is one effect in white and one effect in blue that have some synergy with one another. It's important on a two-color card that both colors feel represented and that the whole package feels like a singular thing rather than two abilities stapled together.
If we were designing a card for an actual set, this is the point where we would also want to think about what that color combination is doing in the set. We'd steer our design to reinforce the themes already present. As this design test requires us to design cards in a vacuum, that isn't something we have to worry about right now.
Here's what catches my eye. I like the idea of the spell granting +1/+1 to all my creatures. None of the previous white-blue cards have pushed toward wanting more creatures, and I like the idea of stretching white-blue in a new direction. That ability is white, so I need something in blue to attach to it. Since my goal is to encourage having more creatures, I like having a sliding effect that counts how many creatures I have. Common white-blue hasn't done any card drawing yet, so I'm looking there. As this is common, I can't draw a lot of cards, so instead I'm going to do an "Impulse" effect (look at top X of library and pick one to put into hand).
All creatures you control get +1/+1 until end of turn.
Look at the top X cards of your library, where X is the number of creatures you control. Put one card into your hand and the rest on the bottom of your library in a random order.
Once I make my card, I need to check back on my requirements. Is it white? Check. Is it blue? Check. Is it an instant? Check. Is it common? Eh, not really. I've designed an uncommon, not a common. It has a bit too much going on and it can turn into a pretty big effect. What this means is either I can reconfigure my grid to allow me to have an uncommon white-blue card, or I can design another card.
As the design test goes on, I'm going to start locking cards in, which will allow me less and less flexibility to tweak my grid. I'm going to try and design something that's common.
I start again with my white effect, as it's a more limited choice of effects. How about "Destroy target creature with power 4 or greater"? What blue effect would fit with this? I'm going to stay away from card draw because if it's small enough, it feels like a cantrip, which mono-white can do, and if it's big enough, it starts pulling away from common. How about a milling effect? White-blue leans toward control, so it might be neat to combine creature removal with milling.
We need to find a way to make the two effects feel connected. The trick I'm going to use is having both effects care about the same thing. The white effect cares about power, so let's have blue also care about power.
Forget the Mighty
Destroy target creature with power 4 or greater. That creature's controller puts cards from the top of their library into their graveyard equal to that creature's power.
Kill a big creature, then mill the opponent for at least four. That seems pretty cool. Let's run through the requirements. White? Check. Blue? Check? Instant? Check. Common? Check. Okay, we're good.
Now we have to do the same thing with each of our nine other cards. The additional thing I would add is I would try to avoid repeating effects if I could. That's not essential for very basic effects like card drawing or life gain, but I would actively try to make each of my cards feel like it has its own space. What this means is that the restrictions are going to get tighter and tighter as I make my cards.
To help with this, I would do my research and compile my options before committing to any of my designs (beyond the few that I chose to build my test around). Knowing that white-blue is more pinched than red-white might help me avoid later painting myself into a corner.
Once I have all my designs, the next step would be to playtest the cards. We allowed each designer to have one playtester they could work with (who was also vowed to silence). Playtesting is crucial because there's no better way to stress-test a card's design than actually playing it. It will also help you greatly with costing your cards.
The final piece is naming. The contestants aren't being graded on the "creative elements" of the card (the stuff the creative team would do), but the name does help sell the overall concept of the card and can often take things that feel disconnected and make them feel more like a cohesive whole, which is important for your design.
And that, my friends, is how I would tackle the design test. Obviously, there's a lot more that goes into it. Colors, card types, rarities—each has their own design requirements and each requires attacking the design slightly differently, but I only get so many words for my column.
As always, I'm eager for feedback both on this column and on the design test in general. You can email me or contact me through any of my social media accounts (Twitter, Tumblr, Google+, and Instagram).
Next week I'm going to be covering a topic I assumed I'd covered before but failed to find for this column when I looked it up: I'm going to be talking about how to design for each rarity.
Until then, I hope you've been enjoying the Great Designer Search 3 so far and hope you tune in Friday, March 9, for the first "show," when we'll finally meet our Top 8 contestants along with the other judges. We'll judge their design test and give them their first challenge. Check it out!
This is a companion podcast to one I did about designing sets that revisit worlds we've been to before. In this podcast, I talk about designing for a new world.
This podcast was made for you to share with someone new to Magic. I walk through the basics and demonstrate the important things to do when introducing the game.