In a recent tumblr post, Mark Rosewater announced the creation of a new team within Magic R&D: “Play Design.” MaRo said the group’s function “is to be a group solely dedicated to the health of tournament environments, to make sure that playing Magic in structured settings is as enjoyable as possible.” Also announced: Dan Burdick will head the team, with Melissa DeTora and Andrew Brown taking full-time roles in it as well.
Magic: A Game You Can’t (Easily) Patch
To get a sense of the problem Play Design will be tasked with solving, consider that R&D has to make formats complex enough to stay interesting, but not so complex that it’s hard to see or appreciate what matters. They have to make formats that are interesting on day 1 and day 150. They have to make formats where the new cards and strategies are sometimes effective, but not always effective. They have to make sure that the format’s play patterns are fun.
Crucially, they have to do all of this without the ability to change the cards once they are printed. In Magic, it’s always been this way. But the rise of digital CCGs like Hearthstone and others shows the value of being able to “patch” design flaws by changing the text of cards or removing them from the card pool—these steps are easy in a digital CCG, but difficult in a paper one.
Now, Magic R&D does have the ability to ban cards, and Magic Online could always deviate from paper Magic if it needed to. But these actions have costs on a different scale than what it costs Blizzard to change the casting cost on, say, Unleash the Hounds in Hearthstone. If Smuggler’s Copter would be a reasonable card at an increased mana cost of 3, that’s an R&D lesson for the next Vehicle block. For the 2-mana version, Magic R&D decided to ban it.
After Smuggler’s Copter was banned in Standard, people still purchased packs of Kaladesh and opened it in the rare slot. Next thing you know, someone’s brother or sister finds an article online and accuses their sibling of cheating—all this because balancing everything is hard and “patching” the set is impossible.
Ben Stark recently suggested that perhaps Glorybringer should be banned in Amonkhet Limited. This was a dubious suggestion, even for BenS—“If you open it in a new pack, you can play with it” is not a concept we should treat as disposable. It did remind me, though, that R&D also faces decisions at this structural level. Should we make a big change to how Limited formats work? Should we change the rotation schedule so that Gideon doesn’t rotate until 2025? Data really outperforms vague recollections about what’s working if the goalposts move and your data set allows you to investigate the new problem in a thoughtful way.
Play Design as a Solution
We’ve covered why getting it right before the set is released is so important. The old “Future Future League” approach (making playtesting a small part of everyone’s jobs rather than a large part of certain people’s jobs) worked some of the time but not often enough. Here, the proof was in the pudding: problematic cards like Emrakul, Smuggler’s Copter, Aetherworks Marvel, that damn Cat combo (Felidar Guardian + Saheeli Rai), and Heart of Kiran, right on the heels of Gideon, Ally of Zendikar and Reflector Mage.
There are a lot of numbers on these cards, and many of the costs turned out to be too low for the power level of their effects. It’s not just about the numbers, though. Felidar Guardian presents an interesting case that’s qualitatively different from that of a card like Smuggler’s Copter: in this case, R&D simply missed the combo. The formation of the Play Design team is, at least in part, an admission that with more resources they might have spotted it in time to do something about it.
This all feels very healthy to me. Your business has a critical function which isn’t getting enough energy, ownership, leadership, analysis, reporting, etc., so you form a task force or team to address it.
First Impressions of the Initial Team Members
Having faith in a concept is different from having faith in the execution of that concept. Here, I happen to have played Magic and spent some non-playing time with all three announced members of the Play Design team (Dan Burdick, Melissa DeTora, and Andrew Brown) so I think that’s a pretty good start.
Dan Burdick was in our Pro Tour Austin testing house when Ben Rubin, Brian Kibler, myself, and others produced the deck known as “Rubin Zoo” that Kibler used to win the Pro Tour. So he’s got experience with testing in Magic’s “real world” Pro Tour scene, but I suspect his more important contribution here is recent experience at DireWolf, working on digital CCGs and honing his abilities relating to playtest team management, data collection and analysis, and other important skills required to build a tournament-ready CCG format. I can’t speak too much to his specific skill set since we haven’t been very close in recent years, but he’s a smart guy who was recently working on a similar set of problems—that’s a good sign.
Melissa DeTora and Andrew Brown are both players with recent experience on large Pro Tour testing teams. I doubt there is more relevant experience for the job at hand. You need strong players working on this problem, because while anyone can theoretically realize that Felidar Guardian is worded in a way that could create a problem, only a Pro Tour level player can tell you whether Copter ought to cost 3 or Aetherworks Marvel ought to cost 6, particularly when format health is on the line and other interests want the card or strategy to be “pushed.” (Wanting something to be pushed is fine, but lacking a team to determine when it’s been pushed too far is not.)
I will say about the current lineup that it’s missing someone like Sam Black, Patrick Chapin, Conley Woods, Travis Woo, or Saffron Olive (in order of my guess at how effective each would be). To most people, Brewers are just mediocre baseball players, but in the Magic world brewing means identifying non-obvious strategies and putting together first drafts of decks to exploit those strategies. It’s fun to watch, but it’s also important. The Play Design team needs to spot these non-obvious strategies and make sure what’s been spotted doesn’t amount to a glitch or exploit (to put it in digital terms) because, again, you can’t patch that glitch or exploit once it’s released.
Developers were always tweaking numbers in the past, but “this feels like it should cost 6” is not the same output as what Dan Burdick and his team will aim to produce. Their work product will have more context, more relevant data, and more people with an informed opinion collaborating to produce the output, with better feedback loops to improve over time. That’s the most fundamental change here.
Handful of Gatekeepers vs. Hive Mind of Real-World Players
The Play Design team has a big, challenging mandate. They will not nail it every time. Players in the real world are constantly iterating, sharing information, observing what wins, and trying to “solve” the format. R&D will never be able to compete in terms of time and energy they can spend trying to solve it first.
The team won’t be able to “solve” a format, and they won’t be asked to. You don’t have to solve a format to realize Smuggler’s Copter is too much stronger than other 2-mana options. You don’t have to solve a format to realize that Emrakul is the only late game you can play in a Standard where it’s legal. And you certainly don’t have to solve a format to realize that Saheeli Rai plus Felidar Guardian ends the game.
Competitive players should be encouraged by this shift in R&D resources. Mistakes will still happen, but hopefully not as frequently as we’ve seen in the last few blocks.