#442: New vs. Returning Worlds
In this podcast, I talk about the difference of designing a set for a new world and designing for a returning one.
Posted in Making Magic on June 12, 2017
Making Magic is an iterative process. We create something, we playtest it, we get feedback on it, and then we use that feedback to either make more new things or to adapt the things we already made. This iterative loop is not only applied to individual sets but also to the game as a whole. Three years ago, I wrote an article titled "Metamorphosis" where I explained a big change the game was going through. Seeing that this column is entitled "Metamorphosis 2.0," perhaps you have some idea where I'm going with this.
The change over to the Two-Block Model (two blocks per year, each with a large set and a small set) had some successes and created some challenges. Today's column is going to examine what we learned and then discuss what we've done to create the next iterative loop.
As I said in "Metamorphosis," I'm going to discuss times of the year when sets get released and the dates shifting from year to year make months inaccurate, so I'm going to use seasons as they apply in the Northern Hemisphere. For any readers from the Southern Hemisphere, I kindly ask that you mentally swap in the proper season accordingly.
Let's begin by examining the successes of the Two-Block Model:
Success #1—More Worlds Per Year
For most of Magic's life, we visited one world per year. We'd make a big splash in the fall and then stretch out the world for nine months' worth of content. For some worlds, like Ravnica, that was fine. For many others—Theros being a classic example—it was not. Most worlds simply couldn't support nine months' worth of content. Players tended to get bored of the world after playing in it for over six months, so we changed that.
Now we're visiting two worlds per year, and response to that has been overwhelmingly positive. First, it shakes things up on a faster timetable. Seeing a new expansion, be it a trip to a brand-new world or a revisit to a beloved old one, is like opening a present. It's exciting and it's fun. We just get to do that more often now.
Second, not every world is for every player. We constantly push the pendulum in new directions, and sometimes we hit something either mechanically or flavorfully that might not be your favorite. With two worlds a year, though, you have less time to wait until something new comes along.
Third, it allows us to try more things. We now go to a new world, on average, once a year. We also now revisit an old world, on average, once a year. There's greater opportunity to experiment with new things or bring back popular old ones. There's more mechanics and more flavor—everyone's been pretty happy.
Success #2—More Large-Set-Alone Draft Environments
We do a lot of market research. One of the things we ask about is Draft environments. What do the players who draft enjoy most? The answer, by a pretty substantial margin, is drafts where all the boosters come from a single large set. Well, we're doing more of that now.
Go back ten years or so and Magic had just one large set per year. Then we tried the experiment with Lorwyn and Shadowmoor back in 2008 and had the first large set outside of the fall. That led to a new pattern where we had a second large set every other year, raising the average to one and half large sets per year. The Two-Block Model got that average up to two large sets per year, and from our data, the drafters seem pretty happy about it.
Success #3—Quicker Storytelling
Another big problem with having one world per year was the rate of our storytelling was pretty slow. Things that got set up in one block would often take years to play out. Changing over to the Two-Block Model literally doubled the pace of our stories.
That, combined with our more focused storyline, our new model of short stories on the website (as opposed to novels or novellas), and more story integration on the cards, has caused an explosion of player involvement with the story. The short stories have gone from being one of the least read things on our website to being the most read. We've also seen the story embraced on social media in ways we'd only previously dreamed of.
Success #4—Planeswalker Decks
Part of changing to the Two-Block Model was creating a different introductory path for new players. A key component of that was a new product called Planeswalker Decks. The Planeswalker Decks are preconstructed decks built around a Planeswalker. They use Standard-legal cards, with a few new designs including the planeswalker face card, and are created to be a good beginner-level play experience. (For more info on the Planeswalker Decks, you can read my column "Ramp of Approval.") The Planeswalker Decks have proven to be very popular. Preconstructed decks plus Planeswalkers is a potent combo.
Now let's talk about some of the challenges the Two-Block Model has created:
Challenge #1—Two-Year Rotation
The Two-Block Model allowed us to change how we were handling Standard. Instead of using two years' worth of sets, we could now use eighteen months' worth, rotating twice a year, once every six months. This, we felt, would allow us to have a faster-changing environment.
That didn't go over well. Players disliked having the lifespan of their cards shortened by six months, and they didn't enjoy more opportunities for their decks to become illegal. The response to this was so strong that we changed back to the old rotation system as soon as we had enough data to show how unpopular it was.
I bring this up now because this was a lesson of the Two-Block Model, just one we've already adjusted to.
Challenge #2—Small sets
For years we've had three-set blocks with one large set and two small ones (with some later years having two large sets and one small one). Throughout those years, we struggled with the third set. How do we add enough variety to keep the players from getting bored with the world while still making something that played well with the first two sets? The Two-Block Model solved this problem by getting rid of the third small set.
One of the most eye-opening things about the Two-Block Model was realizing that some of the problems we attributed to the third set were in fact about small sets. Giving a small set its own identity that also plays well with the large set is problematic. Change too much and the sets feel disconnected; don't change enough and the new set isn't exciting. The third set hid this problem by making the second set seem better in comparison. By removing it, the second set got more focus.
We experimented with a bunch of different approaches to help the second set. Oath of the Gatewatch had a huge mechanical differential (the two sets were mechanically more distinct than normal). Eldritch Moon had a giant tonal shift. The block changed from mystery to cosmic horror. Aether Revolt tried keeping things more the same, being additive rather than subtractive. Players were unhappy when mechanics they liked dropped out between sets, yet also complained that we didn't explore new mechanics enough. For example, Eldritch Moon both didn't have investigate and also didn't have enough meld cards.
In addition, there was the Draft problem. There's a consistency with drafting with only large set packs that we can't replicate with the small set. They're not big enough to draft alone, but lining them up to draft smoothly with the large set is tricky. Once again, we want to continue themes so that the two sets play nicely together, but we also want to do something different to give the small set its own identity.
We've made numerous changes to try to fix this problem. We started drafting the new set first. We put in more packs of the newer set. Starting with Oath of the Gatewatch, we even began making the small sets a bit bigger to try to fit in more things to make the draft work. While we've improved things, as the data I talked about above showed, we're still not making drafts with two sets as popular as drafts with one.
Finally, we discovered that some of the third set complaints turned out to be "last set of the block" complaints. There's a fatigue that sets in on any block. We discovered that nine months was too long. For some worlds, it turns out six months is too long.
Challenge #3—No Core Set
You never know how much you appreciate something until it's taken away. The core set's absence taught us that it did a few important things. First, it allowed us to reprint whatever cards we wanted for Standard. We had assumed that we could just put reprints in the normal sets, but that proved more difficult than we imagined. Reprints, for instance, have to keep their name and, if a creature, their creature type. Often the flavor of that name doesn't fit in a world or the creature type is absent from it.
Other times, elements of the world don't line up with the mechanical definition. Inquisition of Kozilek, for example, was black in a world where all the Eldrazi's spells were colorless. Ghostfire, ironically, had the opposite problem of being colorless in a world where all the colorless spells were Eldrazi spells.
And then there's the focus issue. We want new sets to be something players have to explore and discover. Putting in a powerful reprint both lessens that discovery and tends to make the new set about something old rather than something new.
Second, the core sets allowed us to make relevant cards in a set outside the world where they made sense. Need a Goblin lord a year later to help out a Goblin-themed deck based on the trip to Goblin World from the previous year? The core set could do that. Its looseness in theme allowed it more flexibility to create the exact card Standard needs.
Third, as I explained in my "Ramp of Approval" column, our goal was to make a sequence of products to help provide a smooth transition into Magic for new players. It turns out our ramp ended up having too much of a gap because we were missing a booster product aimed at the less-enfranchised players.
Fourth, it helped with main set fatigue. Under the old system, there were just three major releases and a core set. Many players played with and enjoyed the core set, but it was an opt-in experience. If a player had enough with the main three sets, they didn't necessarily have to care about the core set.
Challenge #4—Too Much Focus on the Gatewatch
Part of the change-over to the Two-Block Model was a greater focus on storytelling, including an ongoing narrative with a continuous cast. While we've had a lot of success in this area, we did discover one common complaint. We turned up the volume a little too high, especially with mechanics.
Players, for instance, liked seeing the Gatewatch on planeswalker cards, but not so many versions. You see, when we first shifted to the Two-Block Model, our goal was to try to keep the main five Gatewatch members always in Standard. This proved to be a big mistake, one that got magnified when we reverted to a Two-Year Standard.
We also made the conscious decision to try to push more story-related cards to be tournament-viable. This resulted in things that were flavorful but didn't necessarily lead to fun tournament environments. (Emrakul, the Promised End from Eldritch Moon is a good example.)
Finally, because we were trying to get the Gatewatch established, we made the decision to involve all of them in all the stories. This led to what many players felt was an overexposure of the Gatewatch.
Challenge #5—The Masterpiece Series
There's a phenomenon in Magic design that I call the "cycle of diminishing returns." Here's how it works: You design a card. It's amazing. So amazing, in fact, that you believe you could make more than one card that has the same design. How about a cycle? That sounds cool. You then make the second card. It's quite good. Not as amazing as your first one, but still very good. Then you make your third one. Not quite as good as your second design, but in a vacuum, not compared with the other two, it's a worthwhile design. Then you make your fourth one. It's okay. It's functional. Then you make your fifth one. It's pushing the boundaries of what you'd print. The biggest justification for printing it is it's part of a cycle. One of the cards is amazing and the rest range from good to passable.
I feel this same phenomenon was happening with the Masterpiece Series. The Zendikar Expeditions were awesome. Dual lands are just so fundamental to the game and the concept of land is essential to the world of Zendikar.; it was great synergy. Expeditions were so well received, they inspired us to make more. The Kaladesh Inventions were also pretty cool, but the Amonkhet Invocations didn't have the advantage of having a theme as clean and simple as "lands" or "artifacts." We started with the idea of doing "instants and sorceries" but soon found the theme didn't connect enough to the world of Amonkhet. We changed to a flavor-based theme built around the Gods, but it required explaining, and the audience never quite warmed up to it.
It's important to start with the success and challenges because the change that's coming was built around trying to cope with these issues.
Players liked having more change. Players liked having more large sets. Players weren't enjoying the small sets as much as the large sets. This seemed to have a straightforward but bold answer. What if we stopped doing small sets?
Four large sets a year though would be too much—both in number of new cards and in the amount of work we would need to do to produce them. Okay, what if we did three large sets a year? That would work except it would leave a hole in the schedule in the summer. Was there something we could fill it with? How about the product we were sad went away? What if we filled the slot with a core set?
Not exactly a core set, as there were still some problems with core sets to solve, but how about a new product that was similar to a core set? That got us to the right number of cards and, if constructed correctly, could solve a few other outstanding problems.
The next problem was what would three large sets mean for the creative team. Would each be its own separate world? We have the team in place to design two worlds a year; was three a possibility? And did it even want to be three different worlds? Weren't there some worlds that we wanted to spend more than one set on?
We hashed through all of this and arrived with a brand-new model, what I've been calling the Three-and-One Model.
Before I begin explaining the new model, I want to stress that this new system begins with "Soup" in the spring of 2018. Both Amonkhet and Ixalan blocks will follow the current model complete with a large and a small set. The reason I'm writing this now is that later this week we're going to be introducing sets that fall under this new system. (Yes, "Soup" will very shortly get a real name.)
Change #1—The Fall, Winter, and Spring Sets Will All Be Large Sets That Are Drafted Alone
There will no longer be small sets (in main releases; supplementary sets, such as Conspiracy, may still be smaller). All main Standard-legal expansions will be large, and all of them will be drafted alone. From a design standpoint, each will have its own mechanical identity, although there's potential for a small amount of overlap.
The reason for that is because from a creative standpoint, we plan to stay on worlds for as many sets as makes sense for the story. Some worlds will be a single set while others might be two or three sets. If we stay on a world for more than one set, something about how the world is structured will let us have a delineation between the sets. Sometimes, but not all the times, we will have mechanical carryover between sets that share a world. For instance, under this new model, if we chose to stay on Kaladesh for two sets, we might have chosen to use energy in both sets (most likely playing around in different design spaces), but the rest of the mechanics would have been unique to their set. Note that sets that share a world won't necessarily overlap in any mechanics.
This also means that we are essentially removing the concept of blocks. Sets will share worlds on occasion, but the preset structure of a locked number of expansions being played together is no more.
Change #2—The Summer Set Will Be a Revamped Core Set
The core sets are back—kind of. As I explained when we got rid of the core set, they've always had an identity problem. Are they for newer players or enfranchised players? We've decided to err on the side of new players. Note this doesn't mean there won't be some goodies for the enfranchised player (there will be new and reprinted cards aimed for Constructed play), just that whenever we got to a fork in the road, we picked the path of making this product the best introductory booster release possible.
This has a bunch of ramifications. For starters, we're going to look at every card through the lens of "is this something that will cause problems for a newer player?" In the past, we would include things that confused new players at the cost of adding value for a different type of player. We're working hard to meet other needs while not sacrificing the vision of being an entry-level product. This means, for example, that we will have a Draft environment that can be mastered more quickly than normal (meaning it should be fun for enfranchised players but not for as many drafts as usual). The core set will also have a stronger integration with the sample decks, Planeswalker Decks, and Deck Builder's Toolkit, allowing for an easier transition between the products.
The set will be roughly half new cards and half reprints, and all the cards will be Standard-legal. The new cards will push toward resonance (aka things a new player would already recognize by having a familiarity with fantasy from pop culture). The set will have some story relevance, but more in a "filling in information of the past about relevant characters" way than telling a piece of the "present-day" story.
Probably the biggest change about the new core set is a philosophy in how we think about it. In the past, we stuck the core set on equal footing with the other three major sets, but with its return we're going to treat it differently. It has a different purpose, different priorities, and it's going to be designed with a different approach. As such, we're not going to treat it as if it's just like the main three expansions.
Change #3—A Different Approach to the Gatewatch
Starting with Hour of Devastation, we're making three fixes:
Change #4—The Masterpieces Series Will Revert to Being in Fewer Sets
The more we worked on making Masterpieces, the more we realized that doing one every set would eventually lead to the overall quality level dropping below our standards, so we've decided to make a big change with them. We're only going to do them when we can do something amazing. That means they aren't going to be something we do often, but when we do them, they're going to be memorable. Hour of Devastation will still have Amonkhet Invocations, but, for example, Ixalan block will not have any Masterpieces.
Change #5—We're Changing Things Behind the Scenes
Today's article isn't really about this one, but I want to bring it up because it was the process of adapting to the Two-Block Model that made us realize we wanted to iterate on how we make Magic. There have been substantial changes behind the scenes, but I'm going to wait to write about these changes until we get to the products that are affected by the new processes (As "Eggs," Ixalan's small expansion, is technically the first set affected, expect an article from me about it late this year.)
I did though want to talk briefly about one R&D change today. Standard has had a rough year, so we wanted to figure out how to change things to prevent a repeat of it. To do that, we've formed a new group in R&D called Play Design. We've hired a man named Dan Burdick to manage the new team. We've allocated a number of existing R&D members to the Play Design team, including turning a couple of our contractors into full-time employees (welcome, Melissa DeTora and Andrew Brown). We're also looking externally for additional people to join the team.
The idea of Play Design is having a core set of people focused on the play environment (mostly Standard, Draft, and Sealed). They will be dedicated not just to playtesting but to be available from the very start of the process to the very end to make sure that decisions at every step are taking tournament play into account. This team is reinventing how we craft environments and how we monitor them throughout the R&D process to minimize the kind of things that have recently happen in Standard. An article by Dan introducing the team will be appearing later this week.
Magic, at its heart, is a game about change. The card mix is constantly shifting, creating an ebb and flow to the game. That change though is not restricted to just the cards but the game itself. Magic today is not the same as it was 20 years ago. Or ten years ago. Or five years ago. Or, as today's article explains, even last year.
After almost 22 years of making Magic, it's exciting to constantly find ways to keep improving it. I hope the new changes I announced in today's column excite all of you as much as they excite me and the rest of R&D. There's a lot of cool stuff coming your way (a bunch of which you'll hear about later this week). Regardless of how you feel about the changes, I'd like to hear about it. Do you like the changes? Do you dislike them? Do you have some mixed feelings? Let me know, either by email or by contacting me through any of my social media accounts (Twitter, Tumblr, Google+, and Instagram). An important part of the iteration loop is feedback, so please help us continue to make Magic better.
Join me next week when the previews of Devastation begin.
Until then, may your relationship with Magic evolve as the game evolves.
In this podcast, I talk about the difference of designing a set for a new world and designing for a returning one.
One of my popular features in my "Tales from the Pit" comic is an ongoing "sit-com" with planeswalkers I make using the Funko Pop! figures. I call the recurring feature "Sparks." In this podcast, I talk about how it came about and what I did to...
Many years ago on my blog (Blogatog on Tumblr), I came up with a system to gauge how likely a particular item, usually a mechanic, was to return in a Standard-legal set. I called this sca...