Magic design, like all of game design, is an iterative process. That means we come up with an idea, test it, get feedback from that test, and make changes based on the feedback. The main way we test our new designs is through playtesting. We make a game, so the best way to understand if it's working is to play with it. We do plenty of what we call "theory crafting," where we talk through our expectations for new cards and mechanics based on years of experience, but nothing can replace actually playing with the cards. Today's article will go through the different ways we playtest as a set makes its way through various stages of design.

Before I get to each stage, let me give you a quick overview of the four stages of Magic design. To explain how they differ, I am going to use the metaphor of building a house. (I do enjoy my metaphors.)

Exploratory design is the preliminary work you do before you begin building the house. This is where you figure out what exactly it is you want. For example, what type of house do you want? What style of design? How big should it be? What type of materials should you use? Where do you want to build it? Are there any new building techniques you want to learn about? Is there any special paperwork you will need? Are there examples of successes and failures from similar houses that you can study? Building a house requires a lot of decisions. Exploratory design is where you explore what all the possible decisions are. My main rule for the exploratory design process is that I want it to go broad, not deep. The team should learn as much as they can about the new design space we are playing in to help the Vision Design team start to narrow down the vision for the set.

Vision design is about the architecture. This is the part where you draw up the blueprints which will be the source of information the builders will work from. In Magic terms, this is where you create the basic structure of the set, including the major mechanics and themes, and establish what mood and tone the set wants to create with creative design and gameplay. True to its name, vision design creates the vision for the set that everyone downstream will use to create the set. The Vision Design team will hand off a card file, but these cards are more of a proof of concept to show the Set Design team the general feel they're going for. Some of those designs might stay, but nothing is essential.

Set design is about being the builder. This is the process of actually building the house. The Set Design team's job is to create the set following the blueprints given to them. But sometimes when you start building a house, the reality of building it requires you to alter the blueprint. The blueprint calls for a basement, but when you go to dig, you find bedrock. The Set Design team adapts to this, creating new mechanical elements if needed. Most of the cards for a set are made during set design.

The play design process is like interior and exterior design. This stage is about doing all the finishing touches on the house to make sure it's the best that it can be. Mostly, the team makes small changes to the finished house, but if need be, they can make larger changes. Set design and play design are the only two parts of the design process that overlap.

With that metaphor out of the way, I'll explore how each part of the design process uses playtests.

Exploratory Design

The Exploratory Design team's main job involves being the keeper of the three buckets. The first bucket consists of good mechanics. These are things the Vision Design team will want to look at when deciding what should go into the set. The second bucket consists of bad mechanics. These are the things that Exploratory Design has figured out aren't worth the Vision Design team's attention. This is very important, as knowing what not to waste energy on saves a lot of time. The third bucket consists of mechanics that need work. These mechanics aren't in good shape for the first bucket, but they shouldn't be thrown away. There's something salvageable, but it will require more work.

The point of exploratory design playtests is to figure out which bucket each mechanic belongs in. To do this, we usually build Constructed decks. The decks are normally either 30 or 40 cards and have anywhere from about 4 to 12 cards with the mechanic we're testing. Most the cards that aren't the thing we're testing can just be existing cards, usually simple ones. The goal is for the existing cards to not muddy the water so we can judge what we think of the new mechanic.

The default is that these decks usually only have one of each uncommon card and no more than two copies of any common card. We normally don't put rare or mythic rare cards in at this point. Our decks can be made using our internal design database. Each designer has their own little sandbox in the database to make their own cards. The decks are normally two-color decks, and all the lands are just dual lands of the two colors. The reason we do that is to lessen non-games. We have a limited amount of time to playtest, so we just want to make sure every player has the mana they need. Environmental balancing won't happen until set design.

In a typical exploratory design meeting, everyone brings a deck or two (sometimes more) based on the prompt provided by the exploratory design lead. For example, during Secrets of Strixhaven exploratory design, we spent meetings on each school and one meeting on things outside of the school.

During these meetings, everyone tries to play with or against as many of the decks as they can. While you normally play a couple games with your own deck, you should play a wide variety of decks in the meeting. Often for the last fifteen minutes, we'll discuss people's thoughts and decide what bucket to put the new mechanics in. If the mechanic goes in bucket one of the first two buckets, we're done with it. If it's in the third bucket, we talk about what would need to be done to fix the mechanic. Sometimes the person who made it will try a new version, sometimes someone else will. Exploratory design lasts two or three months, during which we will have as many playtests as we can.

Vision Design

Vision design normally starts by examining the first and third buckets handed off after exploratory design. (Okay, we might peek at the second bucket out of curiosity.) We then choose the mechanics we like best and put them into a card file. This is the first time a card file exists for a set. Once we've filled in the common card slots (and usually a healthy part of the uncommon slots), we have our first playtest.

This is what I refer to as a sampling playtest. The goal is to have the Vision Design team actually play with the different mechanics and get a sense of what they like and don't like about them. Sampling playtests are usually Limited playtests, where each member of the team is given a certain number of colors. The default is two, but that can change based on what the theme of the set is. For example, if it is a three-color faction set, we'd play with three colors.

The reason we assign colors rather than letting designers choose is that the goal of the playtest is to have everything played by someone. Before we did assigned colors, we found that certain colors (namely black and red) were over indexed. Early playtests aren't balanced yet and often utilize mechanics that are unknown, so it's easy for deck builders to lean toward the colors that have more removal spells. By assigning colors, we guarantee every color will have equal exposure in playtesting.

The outcome of a sampling playtest is twofold. First, we figure out what we did and didn't like. This is less about correct balance and more about our excitement for what you played. We ask a few key questions after the playtest. Was it fun? Did it create new interactions or play patterns that you enjoyed? What emotional response did you have when playing with the cards?

Second, we tend to go very broad in execution for the early sealed playtests because we want to sample a variety of what each mechanic can do. This allows us to figure out which versions we like best. If there are executions we like, we can do more of them. If there are executions we didn't like, we can cut them from the set.

Every week, the Vision Design team has two one-hour meetings and one two-hour meeting. The two-hour meeting is when we most often playtest. We will build and play for an hour and a half and spend the rest of the meeting talking through what we learned. The biggest takeaways from a sampling playtest is what stays as is, what stays but is changed, and what leaves.

The next batch of vision design playtests are Sealed. The colors are no longer assigned. This happens after the file is a little more settled (although things will still go in and out) and is roughly balanced by the play designer on the team. For vision design, we tend to have a flat power level where every card is set to be playable in Limited and is roughly close to one another. This allows all the cards to get played in playtesting. If we make cards particularly weak, they just see no play, and we don't learn anything from them.

As we playtest more, our feedback gets more granular. Let's say we're making a faction set. At some point, we'll say we're happy with a faction mechanic. The feedback then turns to finding the best execution of that mechanic. What types of cards make it play the best? Are there synergies that help enable neighboring archetypes? Usually, the uncommon slots are filled out and we start making rare cards at this point in design.

Then in the third month (of four), we tend to start drafting. Drafting isn't useful until the file is a bit more settled (and we have enough rare cards), and we've started weaving in synergistic cards that connect the different components of the set. Our Vision Design teams have four to six people on them, so starting to draft also means we begin inviting outside people who are not on the Vision Design team to playtests. This is our first chance to get fresh eyes. After working on the set for many months, it becomes harder to sense first impressions. When the Vision Design team started, the file was radically different, and we've all watched as it slowly evolved over time. A new drafter will experience it without the baggage of what we tried before.

Sometimes in vision design, we want to get a sense of how certain elements of the set will play in Constructed. When that happens, we build decks, similar to how they're built in exploratory design, except with more access to higher-rarity cards. This is especially helpful for mechanics that aren't aimed at Limited (for example, mechanics that have no common cards) or have a lot of synergy needs.

The last playtesting needs from the Vision Design team are two big elements of the handoff. First, there is the leadership draft where four members of R&D leadership do a four-person pick-two draft with the set and give larger feedback about the set. Some of this might require changes before handoff to set design, but mostly it tends to be big-picture issues for the Set Design team to think about.

Second is the Vision Summit, where the team shows off the set to a variety of R&D members that didn't work on it. For the Vision Summit, we make ten preconstructed decks, normally focusing on the ten major archetypes of the set. We make several copies of each deck to allow the people attending the Vision Summit the ability to play with a good chunk of the set. Each invitee gives feedback relevant to their area of expertise.

Set Design

The first thing the Set Design team does when the file is handed off from the Vision Design team is do a pass on it where they make the file a bit more real, replacing a good section of the cards with new designs that fill the same basic function. This is also the first pass where mana costs are adjusted to match a more finalized Magic set.

Once that pass is complete, the Set Design team will start doing Draft playtests. These playtests are similar to the Vision Design team's drafts, except the iteration loop starts speeding up. While the Vision Design team will playtest once every other week, the Set Design team tends to playtest every week. The Set Design team has two two-hour meetings a week to allow any meeting to turn into a playtest.

While Vision Design teams are focused on big-picture feedback, Set Design teams look at more granular feedback. Playtests during set design are more environmental, meaning they look for feedback about how specific cards interact with the larger system of the set and one another. They'll start looking for notes about specific card executions, including mana costs, as balance starts becoming a focus.

Set design is also where we start having external playtests. External playtests are drafts done outside team meetings, most often with R&D members that aren't on the team. This is the first chance for more of R&D to get a sense of what the set is doing. Often the set lead will be at these playtests (usually observing and not playing), but everyone playing is sent a survey to fill out to allow them to give all their feedback. This feedback is looked at and discussed in the set design meetings.

The Set Design team will also do casual Constructed playtests to mimic how players with less access to cards will experience the set. You get up to four copies of each common card, two or three of each uncommon card, and no more than one of any rare or mythic rare, and only a handful of those. The decks tend to be two-color decks and lean into the draft archetypes of the set.

As I said before, set design and play design are the only two processes that overlap, so I'm going to jump to the next section, but be aware set design is not done yet.

Play Design

Limited playtests are good for understanding draft and sealed play and give a general insight into less-enfranchised casual play (because casual players open fewer boosters, their experience is closer to Sealed than Constructed), but it's important to start getting a sense of how a set is going to play in Constructed with playtests.

Play design consists of two different teams. Competitive Play Design focuses on Draft and Standard and keeps an eye on other competitive formats. Casual Play Design focuses on Commander and other casual formats.

The Competitive Play Design team has the Future Future League (or FFL), which tends to play Standard about a year ahead of the real world. We used to have something called the Future League that played six months ahead of time, but we realized that it wasn't early enough to allow us to adapt to the lessons we learned, so we made the Future Future League to jump another six months into the future. R&D thought FFL sounded cool, and the name has never changed.

Competitive play designers play in the Future Future League, putting together Standard decks from that future time frame. There are FFL tournaments and FFL free play where designers play their decks against one another. There are regular meetings where they talk through all the data they collected and work with the Set Design team to make appropriate changes.

Competitive Play Design will also have scheduled drafts where they draft with an eye toward balancing Limited. There's even an off-site meeting that happens near the end of play design where they take a whole day to play multiple drafts. This event is usually what leads to the final adjustments on costs.

Meanwhile the set also goes to the Casual Play Design team, who builds Commander decks with various cards in the set. This is referred to as the Casual Play League, or CPL. Like the FFL, they play many games so the team can give feedback on how cards can be improved for Commander play. At the same time the main set cards are being tested, the Commander decks from the set, if the set has them, are also being tested. Sometimes, cards will get swapped between the main set and Commander decks.

Usually when play design starts, the Set Design team stops having regularly scheduled meetings because the file is locked down to allow the FFL and CPL to do their work. After that data is collected, the Set Design team will resume listening to the feedback of the Competitive and Casual Play Design teams. The set design lead will work with the leads to adjust cards along the way, allowing changes to get some additional playtesting.

This whole process from the beginning of exploratory design to the final pencils-down moment of set design and play design takes about two and a half years. There are another six months required for all the teams downstream of design to actually make and distribute the set. And that is all the playtesting we do!

I hope today was an interesting look into a different aspect of design. As always, I'm eager for any feedback on today's article or any of the different types of playtesting I talked about. You can email me or contact me through social media (Bluesky, Tumblr, Instagram, TikTok, and Twitter).

Join me next week for another installment of Making Magic.

Until then, may you practice what you preach.