Limited_Information

An Elf Warrior, a Goblin Warrior, and a Goblin Shaman walk into a bar....

The Creature Type Web

  • Boards
  • Print
Author Image

The letter H!ello! You're probably asking yourself who the heck this Guskin guy is, but fear not, reader: I'm your first guest writer. Known as "Guuuuuskiiiiiin!" to many here at Wizards, I'm pretty much a draft-a-holic and avid fan of all sorts of Limited formats. Back when I was out in the world, I was a competitive PTQ player about once monthly and a casual drafter with college and work friends most evenings and weekends. Now that I work for Wizards, I tend to draft about three times a week on average (it used to be once a week, but now we have the technology of Cubes and Winston to get one-on-one drafts going all the time.) You might be interested to know that I also have two degrees in physics. Currently, I work as a web developer here at Wizards, but whenever the word "draft" reaches my ears, I still perk up and start salivating. So when I was asked to write an article on Limited, I happily accepted. This is my bread and butter!

Oh What a Tangled Web We Draft

I'm hijacking Noah's column today to discuss a concept I've been knocking around I call the Creature Type Web. Learn it, love it and use it to thwart your enemies! It can even be extended past the new-hotness creature types of today's Lorwyn-Morningtide format back to old-and-busted mechanics of Magic's murky history. I'll get to those further on down.

Your Creature Type Web is how tribally connected your deck is. The most tribal deck possible has the appropriate creature type on all non-land cards in it. The least tribal deck possible has no matching tribes among any of its number. (I doubt this is even possible outside of drafting specifically to avoid shared creature types. Man, would your draft have had to go wrong for that to happen in Lorwyn / Lorwyn / Morningtidedraft...) Usually, your deck is going to fall in between those two extremes.

Back in Lorwyn, it was easy to figure out how good your deck was at utilizing tribal effects: just count up the number of cards with that type you have. The intersection of two "races", such as Goblins and Faeries, was facilitated by only one kind of card – changelings – and so it was straightforward for each possible draft pick or deck choice how to go about evaluating tribal effects. The big shakeup comes in with Morningtide: now there are effects that care about multiple possible matches and which let you choose the kind of match you care about at the spur of the moment, and class types make it extremely easy for most limited decks to have that intersection of tribes even without changelings.

Good luck finding non-gelatinous Goblin Faeries in Lorwyn!

Visualizing three-set intersection is tricksy – maybe I can patent this crazy triangle method...

So, how do we use this crazy Creature Type Web to further our Limited games?

Matty Lab and the Monte Carlos

I tried to solve this problem using statistics and equations. I ran into what fancy analysts call the "multivariate problem": trying to reconcile a million different factors that impact a game all at once. I simplified to just worrying about kinship, since that's an effect that's interesting and new, but clearly informs use of other tribally-interested cards. But it turns out, I quickly found the math even for this is more complex than I could solve analytically. Heck, people have been trying to solve problems involving drawing with replacement from a random deck for years, so this didn't surprise me overmuch. What it did inspire me to do is what scientists always do when the math gets too hard: run an experiment!

With the help of my science buddies Todd and Sam, I created a little trial program to figure out how many kinship triggers one could expect over the course of a game, given some number of tribally connected cards. It's a grossly simplified model from cardslinging as we know it, with a number of finicky assumptions, but it's illustrative. I ran this simulation one MILLION times and recorded the results. An interesting answer emerged: sure, as you increase the total number of tribal cards in your deck, you get better kinship results as one would expect... but!

  • Zero triggers is by far the most common result (the mode), even for decks that are heavily tribal. (This makes sense if you think about the number of lands you have to run.)
  • You can see clear "targets" for what number of tribal cards to strive toward if you want to make efficient use of your kinship cards.

 Assumptions I Made  
  • You make every land drop.
  • The game is 20 cards drawn total.
  • You play the kinship card a.s.a.p.
  • Your kinship guy never gets killified.
  • Two kinship guys are twice as good.
  • Tastes great! Less filling!
Interestingly, the "swingier" kinship effects such as Leaf-Crowned Elder's free spell and Pyroclast Council's Pyroclasm are effects you're fine with triggering only once or twice and not necessarily every game (since they're respectable cards even without triggering in terms of impact on the board), so you can shoot for the lower targets, but the "solid" kinship effects such as those on Kithkin Zephyrnaut and Winnower Patrol are effects you want to max out on, so the higher targets are the more relevant ones.


Probability Jumps
Kinship Triggers (1 kinship card) Tribal count
to hit ~10% of games to hit ~50% of games
1 4+ 13+
2 7+ 16+
3 10+ 20+
4 14+ lots of luck :)
Kinship Triggers (2 kinship cards) Tribal count
to hit ~10% of games to hit ~50% of games
1 3+ 12+
2 4+ 13+
3 6+ 14+
4 8+ 16+

The dotted lines represent a three-cost kinship card (as opposed to the solid lines, which show a two-cost kinship card)–since three-cost kinship guys can't be played as early as two-cost kinship guys, the odds are slightly reduced that you can hit the mark you're looking for. However, it's a small effect (but gets larger the higher cost the kinship card is), since the cost only matters if you draw it early in the game.

Also note all of these calculations are approximate. I know a MILLION trials sounds like a lot (it better – those capital letters for emphasis don't come cheap!) but in the grand statistical scheme of things, there's still a bit of error involved with limited sample size.

Click here for a more detailed explanation of the math.

Choose—But Choose Wisely

But numbers aren't the whole story! (In fact, numbers aren't very good at telling a story at all; they're more of the silent, sensitive type.) Another perspective to take is to think around the possible options Morningtide brings to your tried-and-true Lorwyn experience. You'll find four categories of cards in Lorwyn BlockLimited that utilize this Creature Type Web concept and to which we can apply the scalpel of science:

  • Lords
  • Kinship
  • Choose a creature type
  • "Web" cards

Lords and "Count-Me" Cards

There's not much new ground here; Lorwyn already inundated us with effects that care specifically about count of a particular creature type. Lead Developer Devin Low outlined these kinds of cards in his article Count-Me vs. Threshold Zero. It's worth noting that there's a new twist to this category with the inclusion of Morningtide: cards that care about the count of a particular class, such as Rage Forger or Obsidian Battle-Axe... and as I mentioned above, many more decks with be able to assemble something akin to a race/class-powered Voltron using subtype superglue.

Note that there's a smattering of analogous mechanics from Magic's history that also fall into this category and that the model above fits for these too! Affinity has the same elements as a "count-me": the card with affinity plays the role of the lord (it gets better the more cards you control of the specified type, for example) and your tribe is the type it has affinity for.

Kinship

Kinship is special because it cares about (at least) two different types. In fact, observing how kinship plays in limited was the whole "Eureka!" moment that motivated me to write about this Creature Type Web construction. With kinship, the connectedness of your deck is the key quality. A mono-tribal deck has no worries, but as soon as you venture into duo-tribal or further (we can't all be as close-knit as Mr. Mono-Thoughtweft and his Weenie Beatdown House of Pain) the key becomes the intersection of your tribes along the kinship condition.

You can always cheat, of course. Amoeboid Changeling, Runed Stalactite, instants like Shields of Velis Vel—these will all allow you to "boost" your kinship to trigger on any creature. Statistics is going to catch you with your amorphous hand in the cookie jar, though—overall, you have better chances maximizing your web during the draft and deck construction than you would relying on a two-card simultaneous combo that still relies on the top card of your deck.

Hybrid mana from Ravnica is the closest analog in Limited formats to the enigma of kinship—when you're building your mana base, you just need to be aware that either color match will allow you to play your hybrids. So you can reverse the numbers I have above to give you a quick and dirty rubric for the construction of your mana base—for a two-cost hybrid spell, you need to "trigger" with the land(s) of the appropriate type twice, meaning you need at least 7+ sources of the appropriate colors in your deck to consider playing it.

Choose a Creature Type

I always feel like an Iron Chef when I have any of the "choose a creature type" cards in my pile trying to construct my deck—the "choose" card is like the secret ingredient and you'll trying to see which dish will have the most bang for the judges' buck. These cards are easier to maximize than the lords, but not by much: almost always, you'll have drafted something with a major tribal theme, so these cards let you bank on your secondary or tertiary tribal counts. These are like the proverbial glove to a changeling's hand—an outlet for your biggest tribal effects without regard to their underlying type (as long as it matches).

The "wild" component of this category is partially mirrored in the converted mana cost matters cards from Scourge. You clearly get some effect from any permanent you control (well, except for your Ornithopter. Nobody likes your style, Ornithopter!) However, your Torrent of Flame is going to be much more powerful if you have lots of high converted mana cost permanent cards in your deck, just as a "choose a creature type" card is going to go to the Limit if you maximize one tribe.

Maybe get the band—I mean web—back together?

"Web" Cards

There are really only a couple cards in Morningtide that illustrate perfectly this connectedness concept, and I have dubbed them "web" cards. Reins of the Vinesteed is a clear example of a card that scales with a tightly connected web—if you can walk along connected creature types, from Elf Warrior to Treefolk Warrior to Treefolk Shaman to Elemental Shaman, you can keep this Rancor analogue in play quite a while.

Rivals' Duel addresses the same problem, but from the opponent's point of view: how vulnerable is your deck to being Rivals' Bait? The stronger your web, the less likely you are to get blown out by this card. Duel actually cares more about web tightness than Reins does, because any long connections (such as the Elf to Elemental path described above) are like chinks in your armor, opening you up to the dreaded two-for-one.

Assassins Treed

This is a deck I recently drafted here at Wizards, and it's a great illustration of the principles I outlined above. One of my pet goals for each Limited format is to draft some absurd Johnny deck (the turbo-Horde of Notions and the Mirror Entity / Ceaseless Searblades / Brion Stoutarm combo deck are two recent ones from Lorwyn), so when Morningtide appeared with five crazy small-class supporting lords, I knew I had to go for each at least once. Finally, Assassins!


When I opened the Morningtide booster, I saw two definite first-pick possibilities: Winnower Patrol and Scarblade Elite. Going into the pack, I had a whopping five Assassins already as Elite-food—the two Lys Alana Scarblades that I had successfully long-ranged, and three changelings—but I had thirteen Elves to trigger the Patrol.

If we refer back to the tables I gave above, thirteen Elves puts me at about one trigger in half my games for Winnower Patrol. Is a 4/3 for 2 ManaGreen Mana with a relevant creature type as good as the allure of multiple Dark Banishings? Although by the end of Morningtide I might be able to beef up my Elf and/or Warrior count enough that expected kinship trigger number up to two or higher, I'd rather select a card which has a much greater potential impact if I do indeed only get one use out of it.

I decided at the time that the additional power of the Assassin's effect, especially the one-two punch of discarding a changeling to Lys-Alana Scarblade and then reusing it with the Elite, was worth passing on a chance for insane multiple kinship draws—especially with more Warriors likely on the way in this booster. I kept my eye out for future changelings to power up the Elite and even found another "real" Assassin in the Weed-Pruner Poplar.

You can also see two interesting deckbuilding choices I made. One is fairly straightforward: the inclusion of two totally off-tribe Dreamspoiler Witches due to their raw power and great synergy with the instants I had drafted. This brought up a side question of whether I should play Nightshade Schemers as another kinship-like effect, due to the large number of changelings, but I quickly dismissed that option due to the huge number of 5 drops I had in my deck.

The second is a follow-on choice to that, a dilemma which I faced in adding in my final card: should I play something more powerful and subtly synergistic with those Witches and/or good on its own merits, or something more conservative and directly synergistic with the Elves I already had? I had four major tribal effects at work in Elves:

"Dave, you buffoon!" I can hear you thinking. (That's right, I can hear you thinking. Science!) "You didn't include Imperious Perfect in your list of major tribal effects at work!" Checking back against my sideboard, though, the only possible fit for the slot here if I use an Elf is a dorky 1/1 Elf. And you know what you get if you combine a dorky 1/1 Elf with Imperious Perfect? A dorky 2/2 Elf.

Offalsnout
Spiderpig?
The alternative was oft-maligned Offalsnout. That's right: some sort of disturbing pig-monster. It had synergy with my Witches triggers, as a cheap instant (or instant blocker on their turn). However, the main reason I wanted ol' Offal was the fact that I could see his primary effects—being a Grey Ogre and giving me a bit of maindeck graveyard hate—as useful to round out my game plan. Sometimes you have to forgo topping out on synergy, as with the Elves here, and play cards that both stand up on their own merits and which shore up other pieces of your deck's strategy and/or combine to strengthen other paths you might need to take in Limited.

The More You Know

I hope I've left you with a better understanding of how to assemble a crack strike force of Rogues, Elf Assassins, Spider Goats—whatever kins your tribe—in this brave new world of Lorwyn with Morningtide. Remember, these aren't rules set in stone (they're more like guidelines, really). Don't blindly follow the numbers, but instead let them factor into your decision making process when you draft and build your deck. If you guys have any questions, comments or thoughts on my writing, feel free to comment in the forums: web development of the forums is one of my primary responsibilities, so I'll be following the discussion. Thanks for reading!

  • Planeswalker Points
  • Facebook Twitter
  • Gatherer: The Magic Card Database
  • Forums: Connect with the Magic Community
  • Magic Locator