Saturday, April 21, 2012

Knowledge from the Helvault (Part One)

Though you may have come to Magic design as an extension of previous and/or concurrent game design experience, one of the more challenging aspects of designing for this particular game is that much of what we consider to be "good design" is defined in less-than-equal parts by:
  • Personal play-style / psychographic profile
  • General game design philosophy
  • Awareness of what Wizards of the Coast tells us is good design
It's this last quality that might be the most uniquely troublesome: having access to the actual designers and developers of the game is often a boon to our productivity, but it can also lead us to false conclusions and rigid behaviors, because as educational as Mark Rosewater and company can be, they can never give us the combination to the safe. So while we're operating from the perspective that the most recently released set is the One True Document and the Word of MaRo is absolute, each spoiler season has a habit of betraying our dogma.

It is with that in mind that I'd like to take a look at what we know of Avacyn Restored and pay attention to the designs and themes that are reshaping some of my design philosophy and get us all thinking about what the set might suggest about our future in Magic design.


A Gently Flickering Light


Sometime last week, ye olde luminumcan was pondering the power level of a Bushido Bear, when he recognized a problem that I see affecting so much of my own work and that of many other amateur designers:
I tend to get a bit paranoid/conservative about pushing power in my own designs. I need to be bolder. 
No matter how quick of a learner you may be, we all have to start somewhere, and the starting point for many designers is the intersection of Convolution and Obscenely Powerful. As we progress under the tutelage of our peers, we ratchet down the power level, massage the text to something approximating intelligibility, and eventually come to a point where we're designing what bears a passing resemblance to actual Magic cards. Yet it's in that developmental process that many of us contract a terrible fear of breaking those rules that helped us reach that degree of realism. We become paralyzed by our guiding principles because we fail to let them simply guide us, instead elevating those principles to positions of absolute authority. No longer are we thinking "it's best to avoid using variables at common", but rather "variables aren't allowed at common", even as we gaze upon a Heat Ray.

So let us first look at a pair of commons that show just how far we can go:

In conjunction with Cloudshift, Ghostly Flicker illustrates a willingness to do what's necessary to flesh out a set's themes, even if it means pushing your complexity level. The abundance of Enters the Battlefield effects and Soulbond clearly show why a mechanic like flicker is important to the set's overall design, but there are two aspects about these flicker cards, including something particular to the flicker-effects in Avacyn Restored, that surprised me most about their appearance at common.

1. "Flickering" a permanent counters all spells and abilities targeting just the flickered permanent.

While many players grasp the obvious aspect of the effect, that the targeted card temporarily disappears and then comes right back, what they don't quite understand is that even though it's essentially the same card, it's no longer the same card that was being targeted by that Doom Blade — that card got lost in a quantum mechanical swap. Compare this with the Astral Slide lineage of long-term flicker effects, which make it abundantly clear that the targeted permanent is no longer available to be targeted.

2. "Return those cards to the battlefield under your control"

Whereas previous flicker-effects (excepting Ruin Ghost) return the flickered permanent to the battlefield under its owner's control, Avacyn Restored has twisted things ever so slightly. This subtle difference means that you can now use your Zealous Conscripts to permanently steal creatures by Cloudshifting that stolen creature before your turn ends, as the creature (or permanent) no longer "remembers" who its owner is, only that you controlled it before it was exiled.

These two functions are part of why flicker-effects may be some of the most lenticular frameworks for a designer to work with, a framework in which a design possesses greater impact on gameplay than is readily apparent just by reading its text. This is an ideal quality for any design, commons especially. Nonetheless, I was surprised to see these cards at common not only because I perceive the above aspects of the mechanic to be unintuitive, but because I have been conditioned to expect these effects to appear at uncommon and above. Were I to have been working on a very similar set in parallel, I doubt that it would've occurred to me that I could use a high density of flicker-effects to support my set, as the precedent of ~15 Uncommons/Rares vs. Momentary Blink & Flickering Spirit would likely have dissuaded me from pursuing such a path. In other words, instead of taking a balanced and educational perspective to evaluate the formats that surrounded prior flicker-effects, it would have been very easy for me to take a short-cut and make assumptions that would have engendered a more conservative approach.

Make 'em count! (Literally.)


It's no secret that Magic involves a whole lot of math; some might even argue that's a selling point. One of the more amusing subthemes within Avacyn Restored deals with the contrast between Evil's nihilistic "loner" ideology (best demonstrated by the terrifying Descent into Madness) and Good's strength-in-numbers philosophy, a subtheme that is then extrapolated to a number of related counting concepts, ranging from effects that scale based on the number of creatures you control to cards like Triumph of Ferocity, which care whether you control the creature with the greatest power.

There are two important lessons that I've derived from these intertwined "counting" mechanics:

The first is that once you've come up with a single good idea, don't be afraid to milk it. When I wrote about the benefits of diagramming the mechanical relationships within a set to develop a better limited environment, one of the greatest flaws in my approach was that I tried too hard to find synergistic-but-distinct mechanics to flesh out my set, rather than listening to my earlier self and pursue the obvious solution — the obvious solution being to simply do "more of what you're already doing, but a little different." In the case of Avacyn Restored, it's no coincidence that Soulbond deals with pairs of creatures and we get a Barter in Blood reprint alongside it. Cycles and mirrored designs (Harvester of Souls and Soul of the Harvest) are not things to be afraid of, they should be embraced.

The second lesson is very much tied to the first, and that's that we can't be afraid to retread the past. While Avacyn Restored is definitely a part of Innistrad block, the leadership of Brian Tinsman has left a distinctly Eldrazi-sized imprint, infusing a large number of themes within AVR that have clear parallels to those in Rise of the Eldrazi. What was once a lonely Aura Gnarlid has now been spliced into a vertical cycle; the importance of Eldrazi-tokens and high creature counts, which once powered Raid Bombardment and Bramblesnap, has lead to an emphasis on non-Eldrazi-tokens and high creature counts that now power Druids' Repository and Goldnight Redeemer; and let us not forget that where we once feared the might of Ulamog, the Infinite Gyre and his Eldrazi-brethren, we now have a wealth of expensive Demons and Angels too numerous to name.

The value of doing such a "re-imagining" is not because it allows us to diminish the amount of thought and labor necessary, far from it —this is about identifying our strengths, the things that we enjoy designing, and refining such designs until they've become the pinnacle of that form. And that's not going to happen overnight.


In the next installment, I'm going to look at how the continuing process to scale back on Activated Abilities has allowed R&D to spend "complexity points" elsewhere, and hopefully I'll get around to explaining why I think we'll be seeing the return of Delve when we go back to Ravnica.

Until then...

Have I got a treat for you! Wait, maybe that should be a question.

Have I got a treat for you? Perhaps.

You see, while the majority of my fellow Artisans have teamed with amateur design All-Stars like Jules Robins, Nich Grayson, Wobbles the Duck Goose, and the presumably pasteurized Pasteur, I've been steadily skulking away over in the Multiverse, tinkering upon no mere golem's head, but a set of my very own. And while I've talked about bits and pieces of it over the course of our existence, today I've brought not only a rough draft of the entire set, but a sealed pool for everyone to puzzle over.

Codename Melody (280 Cards)

Sealed Pool (Printable Visual Spoiler, Sorted by Color/Rarity)


  1. Wobbles the GOOSE :P

    Otherwise, I really enjoyed this. I have no doubt that that Avacyn is going to be one of the most complex (Melvin-wise) set we've seen in a while. I look forward to diving into your set.

    1. I'll get to editing that, pronto.

      But thanks! Innistrad as a whole has a ton to chew on, and I wanted to take a look at AVR that went beyond just single-card evaluations.

  2. This is an awesome article! You pinned down a lot that hadn't quite come clear to me, and I expect to use this as a reminder to push things in the future. Now to look over that set of yours.

  3. This really was a solid article.

    Something that struck me while reading about designers taking precedent for hard-and-fast rules is the direct analog to actually playing the game. As you get better at Magic, you start to observe trends which form guidelines that fossilize into rules and eventually limit you (unless your hyper-vigilant AKA a Pro). For example, playing your land and spells after you attack becomes so rote that it took a while to 100% remember to play them before combat in Zendikar to activate Landfall.

    In the same way, budding designers first progress by observing trends and adopting behaviors that conform to those trends, but it's dangerously easy to forget there are no rules, only guidelines and that everything has an exception at some point in time.

    Too bad observing this trap in theory is easier than avoiding it in practice. I'm like the Michael J Flores of Magic Design. Have I mentioned that I beat Jon Becker once?

    1. Thanks Jay!

      I don't think you tweet enough photos of your nonexistent children eating Korean fried chicken to really qualify as the Michael J Flores of Magic Design, but you're close.

  4. I've been looking over Melody, and the set just seems to have too much going on. There's support for both Shard and Wedge multicolor at common, you have a tribal component(Knights/Soldiers, Birds, and Mercenaries), a multiple spells theme, an equipment theme, a cmc 6 matters theme, and a graveyard theme. There are a lot of cool and interesting things here, and they interact in cool ways, so I see the reason you wanted them all together. Maybe these themes could be split among two consecutive blocks so they can play together in Standard without overloading this set.

    1. Thanks for taking a look, Jules. And yes, you're absolutely right, there's way too much going on.

      I may never have a reason to write an article explaining the set as is because I'll be changing them, so I'll try to explain some things right now. But note that the thoughts that I covered in this article are the things that are going be to useful for how I proceed in revising this initial draft.

      To start, hopefully it's apparent that the "multiple spells" theme is the parent thematic element that drives (or is trying to drive) each keyword. Melody dictates spellcasting, rhythm reacts to spellcasting, and Delve supports spellcasting by both feeding on previous spells cast and allowing you to cast more spells by providing a discount.

      Whether that's sensible food chain is up to debate. We can look at Innistrad and see something similar, but demonstrated in smaller clusters, such as how Morbid and Self-Mill fed the graveyard for Zombies, or how Self-Mill fed Flashback which helped counter Werewolves.

      The tribal stuff was intended to just flesh out the flavor of the world, but as tribal ideas tend to demand a certain density, it got a bit out of control. The original idea was to just convey a class-system and define the colors according to regional race/class relationships. For instance, in White you have Humans and Birds as the dominant races and an ascending class order with Soldiers at Common, Knights at Uncommon, and Clerics/Advisors at Rare.

      Much of the very explicit tribal stuff (like what you find at higher rarities) is definitely something I want to scale back on.

      The equipment theme, and by extension much of what you can find in Red, was an attempt to carve out a more interesting niche for Red. Considering this is the SPELLCASTING set/block, it's doubtful this is the place to do such experimentation.

      The Shard vs. Wedge conflict is a result of me porting designs from older work into the set file without really thinking and then me trying to ultimately support both perspectives by just having fun designing multicolor stuff.

      In theory, the keywords break down like so:

      Melody - R/U dominant G middle W/B least
      Rhythm - U/W/G dominant (it's like a weird vigilance)
      Delve - B/G dominant U less

      Because Melody and Delve are Wedge-based and Rhythm is Shard-based, some of that confusion in intent was exacerbated as I tried to fill in blanks in the colors where a hypothetical cycle was getting cut short.

      Anyway, I hope that explaining all that might help you (or anyone) focus feedback.

  5. Hey, I'm in this article! :D I'm grateful to see that I've inspired such insightful musings here with my dumb Bushido Bear thing.

    1. 'twas a dumb Bushido Bear.

      Here lies Bushido Bear
      A fine bear of valor
      All of 2/2 for two
      One white and a colorless
      with Bushido 1 to boot
      And still
      Worse than Loyal Cathar