Friday, July 13, 2012

M13 Designer Diary VI

Today I look at the lands and artifacts from M13 and the lessons we learned choosing/designing them. Here's part five. Here's a directory with all the card links from the final set. And here's the text spoiler Chah posted.

Colorless Commons

We cheated a little bit. Our set has exactly the right number of cards in every color and rarity except we made a 102nd common to fit in both Evolving Wilds and Voyage Chart. We did that because the multicolor nature of the set really wanted more than one common mana fixer that every color had access to. I'm quite happy with Voyage Chart and not just because it's resonant and the flavor matches the nautical subtheme of the set, but it can do so much for a player. Half Wanderer's Twig, half Armillary Sphere, it'll take you from two land to three or four as you prefer. You can put a single land from your splash color into your hand, or you can play an all-bond deck and go four- or five-colors relying on your Chart to get you two every time. While we wanted to push players toward two-color allied decks, since those are the most consistent, enabling three-color decks and enemy decks that splash the allied color for bond should be very doable, and making the occasional rainbow deck possible is important for archetype diversity.

Uncommon & Rare Lands

While the possibility of reprinting the original Ravnica shock lands (Hallowed Fountain, Watery Grave, Blood Crypt, Stomping Ground and Temple Garden) never left our minds, we originally wanted to try making an uncommon cycle of dual lands that ETB tapped, partially to support multicolor play in M13 Limited and partially to offer cheaper options to Modern and Legacy players. Ultimately, the argument that they wouldn't be good enough to affect these larger formats pushed us to go with the real deal.

Having seen R&D's choice to continue with the existing core lands all but guarantees that the shocklands will be reprinted in Return to Ravnica itself, which honestly makes a great deal more sense. Why should the core set steal the fall set's thunder? In that light, we probably should have done the uncommon lands or stuck with the core lands like Wizards did.

We did make an uncommon land cycle with basic land types, but these "Invoker lands" don't serve to fix your mana, so much as to make lands good enough in the set that land destruction is reasonably main-deckable. We definitely succeeded there, but developing this cycle was quite tricky. We settled on 6C,T for the activation since that takes a total of eight mana and that proved high enough in testing to be strictly end-game but low enough to see play in significantly more games overall than the original nine mana total. Including a colored requirement in the cost was quite the debate, as you can see in those development links, but ultimately we wanted to preserve the color pie by requiring you to run another land to get the colored effect of the land. And we kept all the lands at the same cost rather than staggering them because the difference between, say, a five-mana land ability and a ten-mana land ability is massive and impossible to balance.

Settling on the abilities themselves took a lot of discussion as well, since we wanted effects that were powerful enough to get excited about, but not so strong that they break the game (which is easier to do than it sounds since these are both free and repeatable effects).

Sunbleached Plains taps a creature. That's not the kind of thing you want to hold onto your spells in order to be able to do, but when you've got nothing else to use your mana on, removing your opponents best threat or biggest blocker can certainly turn a game around.

Misty Island makes a creature unblockable. That's not as versatile as flying since it can't help you on defense at all, but it's also better evasion so as long as you can hold the fort, this Island will get you there.

Darkrite Swamp is less impressive on offense or defense, but the fact that it helps both at the same time is a nice bit of streamlining for your damage race.

Lavacracked Mountain helps make up for the lack of Trumpet Blast in the set, offering you the same mini-Overrun capacity every turn once you can afford it.

Finally, Primordial Forest is a beast, offering a repeatable Titanic Growth on demand, letting you bluff through attacks or just deal 4 extra damage every turn. If I could tweak any of these cards after the fact, I would retcon this card down to +3/+3 because it has proved to be both strong and versatile as well as granting you value even when you don't tap it.

We didn't get enough time to test these nearly as thoroughly as we would have liked. I suspect they're strong but manageable in Limited and in Standard they could be too slow to bother with or auto-includes so prevalent that players regularly sideboard land destruction, depending on the metagame.

Finally, we've got a single rare land that stands on its own: Clockwork Citadel. We considered a Thawing Glaciers update but never found anything that was simple enough and still evoked the idea behind the original. For a good long time, we we're using a Temple of the False God update (2,T: Add 4) but in the end decided we wanted something that tied into the set's multicolor themes. Clockwork Citadel rewards you for playing as many colors as possible (it's Domain, after all) and I think that shifts it solidly from Spike to Johnny. You need three basic land types just to break even, but having all five will net you three mana from this one land. It's not a coincidence that Citadel plays well with the Invoker lands.

Uncommon & Rare Artifacts

Arachnoid acts as both a defensive supplement and an additional reach creature. It's not splashy but the effect on the set is soothing when it comes up.

We really wanted to make a new Gargoyle for M13, but all our designs were flawed one way or another. I'm actually glad we ended up reprinting Gargoyle Sentinel because I think it really deserves a second go.

Solemn Harvester is a neat little utility card that can fix your mana, block intimidating creatures and enable consistent benefits from any of the five landfall creatures. Again, not amazing, but it will contribute to some clever plays and while being generally cute.

Equipment is often a problem because it's inherently strong. Wizards solution has been to make it bigger (Greatsword) or terrible (Kite Shield). We wanted to find something in between and I'm pleased with the result. Great Axe curves nicely with a two-drop, is cheap to play, but expensive to reuse and has a pleasing aesthetic with its 1-2-3 number progression.

We discussed removing the lucky charms (Dragon's Claw, etc) from the set, but ultimately decided we shouldn't because how could we know better than R&D whose kept them for so very long. Haha. Naturally R&D finally took them out this year. Classic. I'll discuss the Rings later. Let's focus on Heavenly Perch, Statue on the Bay, Unholy Throne, Molten Roost and Hunting Tower. While we preserved the slot, we didn't make identical cards because they didn't support the set's land theme quite as well.

If we had been much bolder, we might have replaced them entirely. If we had been just a bit bolder, we might have upgraded them by adding "When CARDNAME ETB, search your library for a basic card and put it into your hand. Then shuffle" which would have made them quite playable. We didn't go that route because we thought they needed to suck. Lesson learned.

We gave Solemn Simulacrum a second year in Standard for the same reason they kept Baneslayer Angel and Primeval Titan yet it's not as warping as the Titans were. "Jens" also supports the set's land theme and multicolor decks.

Adaptive Automaton is too sweet not to give a second print. No question.

We wanted a colossus; Some massive artifact creature Timmy would die for. The original proposal was "Colossus of the Paruns" as a callback to Sword of the Paruns because it also had one ability while tapped and another untapped. Quixotically, that version gave +2/+0 while untapped and penalized your opponents -2/-0 while tapped, but we deemed that too confusing. Giving it the Sword's exact abilities never caught on, so we iterated until we found Guardian of Night and Day. Sweet card, no?

Memory Maker is the most explicit Johnny-pleaser of our rare artifacts. I love that it looks like it does the opposite of what you'd want—by giving your opponent cards—but doubling every mill/forget spell you cast is nothing to sneeze at. (In fact, I'm now thinking this is way too good with Traumatize. Shrug.) That it actually plays amazingly with Innistrad's self-mill theme helps add life to still-legal cards that players may have forgotten about.

Remember Sun Droplet from the original Mirrodin? I loved that card. That was the set where I was just getting back into Magic, so I didn't understand quite how marginal life gain usually is (though I had an idea), but the ability to heal all the damage ever dealt to you given the time is hugely appealing to a lot of players, even if a conditional 2 life per round may not cut the mustard for everyone at all times.

Contested Lamp is easily in the top ten list of cards the team continually iterated on in an attempt to get right. We wanted it be a mini-game the players need to keep an eye on while they go about their normal business. I'm quite happy with the result, but I can tell you it was not easy to find. So many versions were either too risky to play or so quick to trigger that the entire game became solely focused on the Lamp. This still gets a lot of attention—free Djinns are nothing to scoff at—but it rarely takes over entire games.

Oblivion Stone gives every player a reset button and that's important. Doing it inefficiently is also important.

That's it! I've walked you through the entire set and told you every worthwhile lesson and design story I could remember. This has been an absolutely amazing experience and I've learned so much more collaborating on a "real set" than I ever have on my own. I say "real set" not because it's official like Wizards' but because the set feels so real to me. Somehow, seeing it through from start to finish with other people and constant feedback not only from the designers and developers but from this amazing community lends authenticity to it that I could never foster on my own.

As fun as this has been, as much as I've improved as a Magic designer, and as proud as I am of the set, I won't be leading another set design as some have anticipated. At least not for the time being. This project took a great deal of time and my other great love, designing games other than Magic, has suffered for most of a year. I will continue to write and make CCDD (and possibly some other ideas I have cooking) and I would absolutely consider joining another design team if someone else is brave enough to lead it.

Thank you so much for your attention, patience and participation.

If could give but a single piece of advice to other aspiring Magic designers as a result of this experience, it would be this: Check your ego at the door. No card, no cycle, no mechanic is beyond improvement. Don't let any single idea blind you from making the whole better. Everything is mutable, nothing is sacred, and the best way to learn what needs to change, die, or be added is to actually play with actual people, watch and listen.


  1. How is Memory Maker way too good? You're spending 11 mana and two cards to die next turn if you don't win right now. It's exciting and terrible, exactly what combos should be.

    1. I thought the idea was to use it as a 2-card kill combo...

  2. New flavor text for Great Axe:
    "There is always a greater power".

    It might be helpful to link one of the Molten Roost cycle, just in case someone doesn't realize they can look at the full spoiler

  3. I couldn't agree more with what an incredible and informative project this has been. I'm honored to have been on the design team, and if I can make time I'd love to try running something similar. You've left some pretty big shoes to fill, but I guess that's just another reason I'll have to corral you onto the team.

  4. Any comment on whether you think official M13's "domain cycle" creatures had anything to do with your project?

    1. I think that has more to do with the possibility of shockland reprints in RTR than Artisans. There are lots of M13 hint at nonbasics: New Liliana and her Shade, Gem of Becoming, Farseek, Ranger's Path, plus the cycle you're talking about. Also keep in mind that Wizard's M13 has been worked on for 2 years.

    2. While the core set design cycle isn't as lengthy as the expert one, it remains that the official MTGM13 file was finished before we began. There is no chance that anything printed in the actual set was influenced by the designs and discussions we had about our version.

      Ben hits on the crux of the matter. The reason there are so many similarities between our sets, beyond the standard convergence that always occurs, is that both teams were working under surprisingly similar parameters.

      Stuff we couldn't screw up: Replace M12, continue the modern core set legacy, follow Innistrad. The big thing that could have gone any direction but didn't was leading into Return to Ravnica as the next fall set.

      Hinting at and supporting the upcoming block is a major responsibility of the core set and the fact that we were able to guess what that set would be was an enormous leg up in terms of matching parameters for our design.

      With the needs and goals of the set in place, how similar the two sets ended up was purely a question of execution. The biggest point of divergence there was that, after examining the list of options for the returning mechanic, Wizards went with the best option that strictly fulfilled the requirement, Exalted, where we went with the (arguably) better option that bent the rules of "choose an existing keyword and bring it back," Bond.

      Wizards will often isolate such decisions. There are 11 white and black cards with Exalted in the set and, unlike Bloodthirst which required a visible shift in the card set to support and counter aggressive decks, there are very few card choices in M13 that were made specifically because of the presence of Exalted or that couldn't have been made without it.

      In contrast, our entire set became focused on "lands matter" because of the inclusion of bond. Most cards that would have keyed off color were replaced with cards that care about land-type. We looked for other ways to make lands relevant including the landfall cycle, land auras, invoker lands, land destruction and higher rarity cards like Vedalken Plotter, Moonlight Druid and Quirion Exarch.

      It's arguable that our method was too focused or over-themed for a core set. I was nervous whether these choices would prevent M13 from feeling really core. Having shared and played with it, I'm ecstatic to say that it does. Not just because we didn't fail, but because our success offers a new and exciting model for future core sets. They don't have to be 100% athematic and lacking in story as they have. It's entirely possible to make a core set that feels enough like generic Fantasy to remain a perfect entry point for new players, while still exploring mechanical themes and telling a story in a way that makes them more engaging to the established players.

      That's kind of a huge revelation and if R&D learns anything from our work at all, this is the takeaway that I'm sure will benefit everyone the most.