Hello, everyone, and welcome to a rare (mythic rare?) article written by yours truly. As some of you know, at one point I wrote a weekly Play Design column here on DailyMTG, but that column has largely been replaced by a weekly stream where I and/or Paul Cheon play MTG Arena and answer your design questions.
As we do more and more streams, viewers seem to ask the same questions on different shows. In today's article, I'm going to address these questions. Rather than having a VOD people have to watch through, getting these answers out in written form will offer something that's easy to link to and search. I hope you enjoy, and be sure to check out the stream every Monday at 2 p.m. PT at twitch.tv/magic.
How far ahead do you work?
We get so many questions about this, usually in the form of "Did you make [Card X] with [Card Y] in mind?" where it's clear that many players don't know our timelines and how far ahead we work. Play Design does the bulk of its work roughly one year before a set is released. Right now, in May 2019, the newest set we are playtesting Standard for is the spring 2020 set. We are able to change cards that are in the newest set and the one right before it. This means that if we wanted to make a card now after seeing real-world War of the Spark Standard results, that card won't see print until winter 2020.
That said, Play Design works on a very different timeline than the old development team did. We work earlier in a set's lifecycle, meaning that more impactful changes can occur as a result of real-world problems. We have impact on sets in Vision Design and Set Design, and since we know what's coming down the line, we can appropriately design cards that can make future sets successful.
How do you define fun?
There are lots of important pillars of designing Magic sets, such as balance and diversity. Our most important one, however, is fun. Fun is our primary goal for designing our formats.
Fun is a hard thing to define because it means different things to different people, depending on how they enjoy playing Magic. Some players like attacking, others like denying resources to their opponents, while others like building creative decks or tuning decks to beat a certain metagame. It's difficult to cast a wide enough net to encompass so many playstyles, but Play Design has developed its own philosophy on what makes a fun game of Magic.
We want games of Magic to have interaction points at most points in a game. It's important for players to feel like they have the tools to handle the many different situations thrown at them. We try to put answers to a variety of threats in all colors. We try to spread these answers everywhere on the curve to make them accessible to more decks.
We also strive to make games dynamic so that games play out differently based on the decisions the player made. We usually do this by making cards and decks that have many different play patterns available. For example, let's say we have a planeswalker that we are testing and we find that it's always correct to use only one of the abilities. We would change the numbers scheme or the abilities to ensure that different abilities get chosen in an array of situations. We also make creatures that are relevant at all points in the game. A good example of this is Thorn Lieutenant. In the early game it's a reasonably sized creature, but as the game goes on, it can fight the larger creatures with its activated ability.
While fun is very subjective, players tend to have more fun in a game when they feel like they are making meaningful plays and their actions matter. Providing many options of threats and answers at all different points in the game is one of the ways we achieve that.
How do you prevent power creep?
We want each set to have an impact on a variety of formats. Standard is the most important, as it's Magic's most popular format and is the easiest to get into as a new player. We also want our sets to impact formats that use a much deeper card pool, like Modern, Commander, and Legacy. However, it's difficult to have an impact on formats that pull from so many years' worth of cards. If we were to make a large impact on these formats, power creep is likely to happen.
When War of the Spark was first revealed, this question started popping up. How can we possibly balance a set and prevent power creep with so many planeswalkers? In most Standard formats, there are about fifteen or so planeswalkers. In WAR Standard, we were adding an additional 37! Given that planeswalkers are often the most powerful cards in Standard, we had to be very careful. We couldn't follow the normal paradigm for planeswalkers. Doing that would be too difficult to balance, as we'd have to make some very weak-looking planeswalkers to go with the many strong ones.
Our approach for War of the Spark planeswalkers, and our approach to power creep in non-Standard formats as well, was to make sure each card has some kind of role, even if it's a very niche one. If each card has a job, then no matter how the format shapes up, counters and answers are available to whatever is thrown its way.
For example, we knew there was a real possibility that Hydroid Krasis would be very strong in Standard. Since its triggered ability can't be countered, we wanted an effect that could naturally stop it. We put that effect on Narset, Parter of Veils. Narset is a narrow card in general, but its static ability is very powerful in the right metagame. That metagame is happening in current Standard, and Narset is seeing main deck play. If Gruul and White Weenie begin taking over, Narset may move to sideboards. In older formats where cantrips and draw effects are running rampant, Narset plays a much bigger role. We try to make many niche effects like Narset's with the expectation that they will be impactful when the metagame calls for it.
Did you know that [Card X] was good in Modern/Legacy?
Similar to the previous question, we try to make cards for all different formats, including Modern. Modern is a format that spans over sixteen years. As more and more sets come out, their likelihood of having an impact in Modern (and Legacy and Vintage) is less and less. One thing that we do to impact formats is add unique effects to permanents that are hard to deal with. For example, Narset, Parter of Veils has an effect that we don't see very often, but it's quite powerful in formats that contain one-mana cantrips. That line of text does exist already, but usually on creatures that are hard to cast (such as Leovold, Emissary of Trest) or have low toughness (Notion Thief). It's much harder to deal with a 5-loyalty planeswalker than a small creature. As you can see, we have added these unique effects to static abilities of planeswalkers in War of the Spark to impact older formats that play fewer answers to planeswalkers. Doing this creates churn in an environment. Maybe players add more creatures to their decks to attack planeswalkers, play a different removal suite, or avoid playing cards that get hosed by these effects.
Another way we can impact non-rotating formats is by creating more low mana cost, interactive cards that are less powerful in Standard than Modern or Legacy. One example of this is Fatal Push. In Standard, it's quite hard to turn on the revolt ability, but in Modern and Legacy, where your mana base contains fetch lands, it's achievable nearly every turn. There are lots of mechanics that are naturally stronger in older formats than in Standard, graveyard mechanics and cost reduction mechanics being some examples. That's where we try to aim designs for Modern and Legacy.
Did you know about [Powerful Interaction] when you designed it?
You may have noticed that there are more combos in Standard than there have been in recent sets. This was certainly intentional. We believe that Standard is healthy when all the major macro archetypes (Aggro, Combo, Midrange, and Control) are available. Standard has been deprived of combo for a while, especially after Kaladesh, where combo was stronger than we wanted it to be. Our team has learned from past mistakes and we are more calculated in our approach to making combo decks. There are a few things we look for when making combos in Standard:
- How many cards do I need to combo off? Two-card combos mean that you need to put together fewer pieces to combo off, and that can be problematic.
- Does the combo have redundancy? Splinter Twin has two different cards for each combo piece (Card #1 is Pestermite or Deceiver Exarch, Card #2 is Splinter Twin or Kiki-Jiki, Mirror Breaker), giving the deck lots of redundancy. We try to make combos less redundant so that games play out more dynamically.
- What are the interaction points? How many turns of setup do I need? Can my combo be interacted with by cards that people usually play (such as creature removal)? As long as there are plenty of interaction points, the combo is usually healthy.
- What is the deck-building cost? Do I have to fill my deck with lots of do-nothing cards? For example, Splinter Twin's deck-building cost is very low. You get to play a control deck with lots of cantrips and don't have to devote many slots to the combo. Conversely, the deck-building cost of Modern Cheerios is very high. The majority of the deck is part of the combo, and you can't play much in terms of interaction if you want to combo off consistently.
We are actively trying to put healthy combo decks in Standard, such as the Ral, Storm Conduit plus Expansion // Explosion combo or a combo deck built around Bolas's Citadel or Dreadhorde Invasion. When we come across interactions like this, we put the cards through their paces to be sure we get them in a fun spot for Standard.
Why doesn't control have a faster win condition?
A control deck looks to answer every threat thrown its way, play draw spells to gain more answers, and then once it has full control, win the game. However, sometimes there aren't strong enough win conditions in Standard. When that happens, decks like pre-WAR Esper Control emerge, a deck that's only win condition is using Teferi, Hero of Dominaria to "tuck" itself back into the library over and over, so that you'll never run out of cards. Once the deck has full control, it takes many, many turns to actually win, and the opponent has no options and has to sit there for many turns until they die. We don't view that play pattern as fun. This is the kind of win condition that we don't want for control decks. The reason this card is the primary win condition is because the other options available are not strong enough. Moving forward we are striving to make sure control decks have strong, fun finishers to win the game once full control has been established.
Some examples of control decks with strong win conditions are Esper Dragon Control, which mainly won with Dragonlord Ojutai, and Blue-Black Control with Pearl Lake Ancient. In real-world Standard, we have access to Chromium, the Mutable, but that card ended up being weaker than we had anticipated. We predicted that when WAR dropped, control decks would adapt more planeswalkers as win conditions, specifically planeswalkers that can provide a fast clock and win quickly once you've turned the corner.
Play Design has found that control finishers are the most challenging cards to design. It's tricky to find the right balance between being both strong enough and fun, because sometimes losing the game to one resilient card isn't the most fun—and if the finishers aren't strong enough, then players will just not play a win condition. (See Ivan Floch's White-Blue Control deck from Pro Tour Magic 2015; his only win condition was a single copy of Elixir of Immortality to reshuffle his used spells back into his library over and over and ultimately deck his opponent.) We learned valuable lessons from seeing Chromium in the real world and are incorporating them into designing our future control finishers.
Does Play Design have influence on bans?
Also known as: When are you going to ban [Card X]? and/or Why did you ban [Card Y?]
Play Design does have influence on bans. We have a meeting before every B&R announcement to discuss the state of each format, look at data, and determine if anything is problematic enough to ban.
There are many factors to consider when it comes to banning cards, such as:
- Power level: Does the deck have a high win percentage? Is the card/deck significantly more powerful than anything else in the surrounding format?
- Does it have a high metagame share? Sometimes the power level of a deck is reasonable, but the deck still is played more than anything else. This was true of energy decks during Kaladesh Standard where we found the win percentage at a reasonable level, but it was played much more than anything else.
- Is the deck fun to play against? Are there enough answers to it? This was true of the last Modern banning of Krark-Clan Ironworks. This card met many of the other criteria as well, but it was also very unfun to play against and hard to interact with.
- What is the community feedback? Do players actively dislike the card/deck? Are players no longer playing the format?
One question that gets asked on our streams quite often is will we ban Nexus of Fate in Standard. We found the card to be very hard to answer in Best-of-One Standard (where it was banned on MTG Arena) but the win percentage in Best-of-Three Standard was not very high, and we are seeing reflections of that in premier Bo3 Standard tournaments. At the moment, Nexus does not meet our criteria for banning, but we are always keeping an eye on the metagame and listening to community feedback about problematic cards.
How do you design for Best-of-One?
The Best-of-One format on MTG Arena is pretty new to us. We knew that it was going to exist when we were designing Guilds of Ravnica, and we designed some cards with it in mind, but we did not design an entire format around it. Now that Bo1 on MTG Arena is a popular format, we are more mindful of how we design cards for it. Here are some things we've started doing:
- Design more flexible, main-deckable cards with modal or niche effects: Some examples of this are Kraul Harpooner, a card that is a fine creature but also a great answer to fliers, and Knight of Autumn, a card that has utility when you need it, but is a 4/3 when you don't.
- Be mindful of too many one-drops, or cards that power up one-drops: Hyper-aggro decks shine in Bo1, where curving out is powerful and many opponents don't have the right answers. We are trying to be more careful with hyper-aggressive decks and cards that are much stronger on the play than on the draw. Venerated Loxodon and Runaway Steam-Kin are some examples of cards that synergize well with one-drop creatures, and these cards press the advantage more than we like in Bo1.
- More card selection: It's not fun when your hand is full of creature removal and you're up against a creatureless control deck. However, you need to play these types of answers in Bo1 if you want to beat hyper-aggressive decks. We are trying to add more card selection effects to the environment so that your cards are dead in hand less often. This can include scrying, looting, or ways to use your unwanted cards for other effects (such as discarding to a jump-start card).
Since Best-of-One Standard is a new format, we are still learning and watching for trends and cards that are successful. It's a gradual process, and you'll see more cards built with the format in mind over time.
Wow, we covered a lot of stuff today. I hope this article cleared up a lot of questions you may have had about Play Design's philosophy, and I'm happy to take any follow-up questions on Twitter, Reddit, or our next stream. We are always looking to improve our processes and are listening to community feedback. Thanks for reading!