Nuts & Bolts #17: Finding Your Mechanics, Part 2
Last week, I started part one of my annual "Nuts & Bolts" article, a series where I explain the technical side of making Magic cards. The topic, finding your mechanics, was complex enough that I turned it into a two-part article. If you haven't read last week's article, I suggest doing that first. Here's part two:
How do you flesh out your mechanic?
The simple answer is that you just design a lot of cards and see what works. With experience, you start to internalize what different mechanics can do, which makes estimating the design space of a mechanic easier.
For newer designers, here's what I recommend. Find your favorite Magic database. Then look up a recent set. Magic: The Gathering Foundations is a good start, being positioned as an entry-level product. Start by looking at the common cards. Then, for each kind of card at common, design a card with your mechanic using an effect of that kind. If your mechanic only works in certain colors, you can just look at those colors. Some kinds of cards might not work with your mechanic. That's okay, you've just learned something important about your mechanic.
Sometimes a certain card will inspire you to design more than one card. That's fine. The goal here is to just design as many cards with your mechanic as you can. You're exploring the design space, so anything and everything is fair game if you follow the rules for color pie, card types, and rarity. Generally, if we've done it in a recent Magic set, it's acceptable in that color, card type, and rarity. Yes, we do bend things from time to time, so there are things you might copy that we wouldn't do in just any set, but I wouldn't worry about that early in the process.
Now it's time to think about mana costs. The goal in early playtesting is to make cards that are in the ballpark of real Magic cards. You're not worrying about overall balance yet, just the balance of your playtest. These disruptive cards can warp the data you're trying to get. In actual Magic design, I have a play designer on my team that assists with mana costs. If you have a friend that you think would be good at helping you cost cards, they can be a valuable resource.
I'll walk you through what I do when I don't have a play designer to ask about mana costs. First, think of an existing Magic mechanic that's like yours. For example, if you have a cost-reduction mechanic, you could use convoke cards. If you have a kicker-like mechanic, you could use kicker cards. If you have an evasion mechanic, you could use skulk. Then, look at cards with that existing mechanic and use their costs as a baseline.
Another technique is to look up a version of what you're making. For example, let's say I'm trying to make creatures with the rumble mechanic (my made-up mechanic from last week's column). I have a red 3/3 creature with trample and rumble. I can look up other red 3/3 creatures with trample and use those as a guideline to figure out the card's mana cost. Again, these costs don't have to be perfect. You're just trying to get close enough. If you're wrong, you can change it for your next playtest. Or you can just change it in the middle of the game. That's what we do.
Once you've made some cards, it's time to playtest them. In early design, we'll make 40-card decks. R&D sometimes makes 30-card decks for playtests, but 40-card decks are easier to build, for enfranchised players with experience in Limited. You should have four to eight cards with your new mechanic. You can have a bit less if those cards are permanents because permanents stay on the battlefield, or a bit more if the mechanic is only on instants and sorceries. You can choose to start with one card with the new mechanic in your opening hand if you wish, but draw one less card, meaning your starting hand size should be the same as a normal game.
The rest of the cards in your playtest deck should be existing cards because you need a control for your variables. If too much is different, it becomes hard to judge what is happening. When building playtest decks, we allow for two of any common and one of any uncommon, with no more than eight uncommons in total. And yes, this means in early playtests, you should only be playtesting commons and uncommons with your new mechanic.
If the cards are too strong or too weak as you play them, you're allowed to change their costs as you play. Remember the goals of playtesting are different than normal play. The goal isn't having a good time but rather learning about your cards. If you need to adjust things as you play, that's fine. Also, if you or another playtester draws a bad hand, they can just draw a fresh hand. Early playtesting isn't about environmental balance, so there's no need for things like mulligans.
As you play, take notes about what is and isn't working. The goal of early playtesting is to "find the fun." Is your mechanic enjoyable to play? On which cards specifically? Where does it shine? Where does it have problems? Look for what does and doesn't cause players to have fun. If you have more cards than fit in your playtest deck, feel free to swap playtest cards in and out between games. If the non-playtest cards are warping things, you can adjust them as well.
Here are some questions to ask yourself after each playtest.
Was the mechanic fun?
This is the most important question. Did you enjoy playing with it? And it's okay if you didn't. The point of playtesting is learning about your new cards and mechanics. If it's not working, you need to respond to that. There are two basic options. One, you can tweak it. Take the feedback from the playtest and figure out how you can change your mechanic to adapt to the concerns. Two, you can get rid of the mechanic. Not every mechanic works out, and that's okay. For every mechanic that sees print, R&D probably designs 40 to 50 ones that don't. Magic design is hard, and a lot of ideas that sound good in concept fall apart in execution. When I walk away from a playtest and decide to get rid of a mechanic, I consider it a good playtest. I learned something important.
Which designs shined the most and the least?
Even if a mechanic shows promise, that doesn't mean every execution of it will. Playtesting is valuable because you can learn when a mechanic shines and when it doesn't. Keep the cards that are working and ditch, or tweak, the ones that aren't. As you iterate on the mechanic, it will get better as you find where it works best. Use the successful cards as inspiration. It's this kind of iteration that will inform you how much space your design has at each rarity.
Which cards worked best with your mechanic?
Another important part of playtesting involves learning which themes your mechanics work best with. Part of designing a set requires leaning into synergies. If certain cards play well with your mechanic, that probably means your set will want to lean into those themes.
Were the new cards dominating the gameplay and which were overshadowed? Was there a good balance?
You'll also learn about the general strength of your mechanic during playtests. The less like other things it is, the harder it is to judge in a vacuum. Playtesting does the best job of showing a mechanic's strength. Make sure to keep tweaking mana costs, and other numbers, such that the mechanic is having the right level of impact. As your file progresses, your mechanic will usually shrink in size. You want it in higher volume early to experience it, but as your set evolves you'll usually want to lessen its impact.
Did other players understand the new mechanic?
You will obviously playtest with at least one other person. Ideally, it's great to have more playtesters. You want people who aren't as familiar with the mechanic as you to judge things like comprehension, ease of play, and enjoyability. This is your baby, so it can often be hard to be objective about it. After playtesting, it's nice to sit and talk with your playtesters. If that isn't possible, at least ask them to write up their thoughts. A good tip: when commenting on a playtest, the designer of the thing being tested should talk last. You don't want your opinion to influence their feedback.
You'll want to repeat this process until you have your one or two main mechanics. Playtesting will help you understand how many cards you could design. I should note how many you could make and how many are best for your set are not the same thing. Once you have your main mechanics, it's time to add them to your set skeleton.
Figure out what existing mechanic is most like yours. This is less about mechanical effects and more about the role the mechanic plays in a set. Then, look at the first set that comparable mechanic appeared in. Note how many cards were included with it at each rarity and use that in your first common playtest, which is a playtest of only commons (and maybe a few uncommons). That number might ultimately be wrong, but you'll get a sense of things as you start playtesting your set. You have to start somewhere.
Previous "Nuts & Bolts" articles explain how to fill in your set skeleton. Today, I'm focused more specifically on how to mold your set's main mechanic. Each playtest will give you more information about the overall volume of your mechanic. Usually, you'll want some cycles with it (assuming it goes in all five colors). As you add rarities to your set, you'll have a chance to start exploring with mechanic volume. I recommend erring on the side of having too much rather than too little. Too much will help you adjust it to the right level faster.
After each playtest, ask yourself if you liked the density of the mechanic, or the amount of it in the set. You don't want your set focused too much on your main mechanic, but you want it showing up enough to be impactful. Different mechanics need different as-fans, so this is why playtesting is so important.
What is as-fan and how do I figure it out?
I just referenced an important design tool—as-fan. Let's dig into that. As-fan stands for "as fanned" and represents how often a particular subset of cards will show up in Limited. If, say, kicker has an as-fan of one, you can expect an average of one card with kicker in every pack.
Getting the as-fan of your mechanic correct is important. To figure out your as-fan, count how many cards of each rarity your set will have. Let's look at last week's hypothetical mechanic, rumble, as an example.
For rumble, early designs and playtests tell me that I can make one common cycle and then one extra card with rumble in red and green, the two colors I want to focus the mechanic in. At uncommon, I'm going to try the same thing, a five-color cycle with an extra red and green card. Then at rare, I'll assume one red card, one green card, and one non-red, non-green card. Rumble won't be on any mythic rares.
This means I have seven commons, seven uncommons, three rares, and no mythic rares. A Play Booster has fourteen cards. Let's assume that one of those cards is a land, nine are commons, three are uncommons, and one is a rare or mythic rare. Actual Play Boosters are more complex than this, but I'll keep it simple.
Last year's "Nuts & Bolts" article went over the cards per rarity in Play Boosters, which I'll use as a baseline. Take the number of commons you have for your mechanic, in this case seven, and divide it by the number of commons, 81. That gives you 0.111. Now multiply that by the number of slots you have for that rarity. There are nine common slots. That gets you 0.999, which rounds to one. So, with just your commons, you have an as-fan of one.
For uncommons, divide the number of uncommons you have with your mechanic, seven, by the number of uncommons in the set, 100. That gets you 0.07. You now multiply that by the number of uncommon slots. That gets you 0.21. Finally, take the number of rares with your mechanic, three, and divide it by the number of rares in the set, 60. That's 0.05. You multiply that by the number of rare slots, which is six over seven, as every seventh booster could have a mythic rare. You get 0.042. Add up all your rarities, and you get an as-fan of 1.252.
The as-fan is a little more relevant for me than all of you because I can look up the as-fan of previous mechanics. If you're curious how your mechanic measures up against famous mechanics from the past, you can do the same calculations I just laid out. You will need to look up things like how many cards of each rarity the set had and how many cards of each rarity were in the booster slots. Draft Boosters were at (mostly) one land, ten commons, three uncommons, and one rare or mythic rare, although in the past, mythic rares used to be in one out of eight Draft Boosters.
The other useful thing about as-fan is that it's a locked number that you can reference. As you go up and down, you can track it, and it will help you figure out the right feel. In actual Magic design, we have some tricks to help us with as-fan. The most useful is the "locked slot." If we have a theme that we think is important, we can dedicate a slot to it. Dominaria, as an example, had a legendary creature slot. March of the Machine had a battle slot. Original Innistrad had a double-faced card slot. For casual designers, I don't think you have to worry about stuff like that, but you can if you wish to.
How do you make the mechanic better?
Here are some tips:
Iterate, iterate, iterate
There is no resource more powerful in game design than an iterative loop. Playtest a lot, ideally with a variety of different people. Always get feedback from those playtesters. Be smart in how you change your file to address the feedback. Then playtest again. If you do it correctly, your set will improve over time. I like to think of this as iterating toward fun.
Don't be afraid to experiment
Playtesting is a tool to help you learn. One of the ways to learn is to try bold things. We spent six weeks in Khans of Tarkir design, putting our current file on hold, so that we could see what morph for
Be honest with yourself
There can often be a big gap between what you want the mechanic to be and what it is. There's an expression in creative circles: "you have to be able to call your baby ugly." Part of truly advancing your mechanic is recognizing what is and isn't working. It's not enough that the card reads well. It has to play well. If you find yourself not playing with your mechanic, or playing it and always regretting it, something has to change.
Harness synergy
Magic is a trading card game, which means it's very modular. Cards can and will work well with other cards. As you playtest and iterate, it's not just your mechanic that will change, your set should change around it. As you find yourself having fun moments, keep track of those. Then adapt other parts of your set to play into the successes the playtesting has illustrated. It's important to remember that a mechanic's success is often about the other mechanics in the set, especially in Limited.
Keep your cards focused on your mechanic
Another mistake is loading your new cards with too much text. If you want people to focus on your new mechanic, most of what the card should be doing is that new mechanic with just enough text to allow it to have its own identity. In movie writing, there's an expression: "don't pull focus." Make sure you're letting the audience focus on the part of the story, or in this case the card, that matters. Any text on your card should serve the larger goal of the card. If it's a new mechanic, it should be about helping that mechanic shine.
How Do I Know When My Mechanic Is Done?
There's always more to do with a mechanic. You can keep tweaking it. We have a great tool for dealing with this: deadline. At some point, I have to hand my set off to other designers, so I have a deadline on the calendar. I recommend giving yourself artificial deadlines. They're a great motivator.
But let's address the larger issue. How do you know when to let go of your mechanic? Where can you stop changing it? It's mostly based on feeling. I know I'm done when there are no obvious big things to change. At that point, I feel I've had enough time to explore the design space and my team and I have created what we feel is the best execution of it. A good rule of thumb: if you have three playtests in a row and haven't made any substantial changes to your mechanic, you're probably good.
"My Work Here Is Done"
I hope you enjoyed both parts of this year's "Nuts & Bolts" column. I enjoy taking time to dig in deep on the technical side of design. As always, if you have any feedback, you can email me or contact me through social media (Bluesky, X, Tumblr, Instagram, and TikTok).
Join me next week, for another installment of "Design Files."
Until then, may you find a mechanic that brings a smile to your face.