Tuesday, November 4, 2014

CCDD 110414—Sliver Breeders

Cool Card Design of the Day
11/4/2014 - Here's a novel way to do slivers. Given how mechanically different it is from existing slivers, it likely makes more sense to use breed on another type of creature, but thematically it's perfect for slivers, so that's how I'll present it for now:

Breed lets you generate new creatures, which is powerful in and of itself. Selesnya Evangel isn't exactly a chump. Breed also lets you mix and match abilities, to generate progressively stronger creatures through multiple generations of breeding. It's intentional that when two Oakshell Breeders breed, the resulting creature is 4/4. It's also intentional that if that creature then breeds, its off-spring will be at least 4/4. Since breeding even once significantly increases your capacity to breed again, and the whole thing can grow exponentially, breed likely requires a mana cost. It does leave you tapped out like outlast, but just taking one turn off to breed puts you up a creature, and quickly mitigates that tempo loss.

The primary complexity with this mechanic is tracking which tokens are children of which slivers, or more to the point, exactly what abilities every token has. I'd love for this set to have one sliver token in every booster pack, and leave room for players to write the combinations on, but that still might not be enough. This tracking difficulty is the biggest problem with breed.

But it's not the only one: There's a design challenge because tokens can acquire multiple copies of abilities. Having trample twice or flying thrice is fine because they're not cumulative, and having Oakshell's ability multiple times is positive because it is cumulative, but Glassskein's ability being cumulative is negative because it's mathy at best and unintuitive at worst. (A child of two Glasskein's will cost {4} more to target despite its power being 2.) That means we'll have to be careful how we design breeders.

While it's neat that the child of an Oakshell Breeder and Glassskein Breeder is harder to target than either of its parents, that's not as intuitive as it would be if we make a simpler hexproofy parent:


  1. To clarify, it was my intention that the tokens would gain the breed ability too. That said, it occurs to me that things get much simpler if they don't.

  2. This strikes me as the kind of mechanic that would work better in an online-only game. The "infinite card advantage" part is still a problem, though.

  3. Thinking about why it might matter which Sliver parent you activate the Breed ability with, and how we could introduce variable Breed costs and base p/t, I came up with this version of Breed:

    Glasskin Sliver 1U
    Creature - Sliver 1/1
    Spells that target this creature cost 1 more to cast.
    Breed 1U (1U, T, Tap target untapped Sliver creature you control: Put a Sliver creature token with Glasskin Sliver's base power and base toughness onto the battlefield tapped. It has all its parents' abilities.)

    Oakshell Sliver 3G
    Creature - Sliver 2/2
    This creature enters the battlefield with an additional +1/+1 counter on it.
    Breed 3G (3G, T, Tap target untapped Sliver creature you control: Put a Sliver creature token with Oakshell Sliver's base power and base toughness onto the battlefield tapped. It has all its parents' abilities.)

    Haven't tackled the complexity issue at all yet. I guess one way to simplify it would be to put it on a single Rare.

    Riptide Project Master (RAR) 3UU
    Creature - Human Wizard 3/3
    2, T, Tap two untapped Sliver creatures you control: Put a 1/1 colorless Sliver creature token onto the battlefield tapped. It has all its parents' abilities.

  4. I love this mechanic but agree that it would work a lot better in a video game based on Magic, than the actual game of Magic.

    1. Chip Beauvais mused on Twitter: "Quite possibly, this mechanic deserves its own game." He's got something there. This is dangerously complex (and powerful) to add to Magic, but another game *designed* around it (perhaps with dry-erase cards, or bits to place on spawn to indicate their abilities) could potentially use it better.

    2. I'm imagining a board game-style setup where each player begins the game with his or her own Breeding Pit and a selection of Slivers. You get a number of actions each turn, which you can use to research genetic engineering (spawn a parent Sliver in your Pit), breed a new Sliver, deploy your Slivers to contest locations within the hive, or use your Slivers' abilities to fight nearby Slivers. Different colors of Slivers could have different families of effects and give you bonuses depending on how many you control or what territory you've taken. When all other Sliver lineages are wiped out or one player controls the majority of the Hive... he or she is crowned the Sliver Queen!

    3. I'm on board with the "own game" camp. This mechanic has the potential to be awesome, but it also has a lot of complexity and requires enormous density, both of which are hard to support on top of everything Magic is already doing.

      If we want to build it into Magic, we definitely need to track the parents more explicitly, which probably involves exiling them. I'd start somewhere like:

      (As a Sorcery--Exile ~, another Breeder you control, and all cards exiled by each of them: Put a Breeder creature token onto the battlefield with power and toughness equal to the combined power and toughness of the exiled cards. It has all its ancestors' abilities.)

    4. Jenesis, you should make that game. I'll consult if you like.

  5. This is certainly very novel and interesting!

    Is it your intention that breeding cost 3 and tapping two slivers, or tapping and tapping another sliver? Also, just a templating thing, it needs to say "tap another untapped sliver creature you control..."

    I don't think I would want this mechanic to go on Slivers, in the way that Allies weren't slivers, I would rather make a new creature type.

    The Volume to make this work would have to be very high to assure players could activate these cards. As an example, the volume on slivers in M14 was much too low, and as a result they saw very little play.

    The reward of making infinite Gray Ogres (with upside!) is very, very high, and I think the correct analogy here is actually Sprout Swarm, since the more creatures you make, the higher the rate at which you can make even more. If I get two Oakshell Breeders in play, I probably never need to play a card from my hand again. (in other words, I agree with Vetinari)

    The complexity board tracking issue makes me wonder if we could somehow have the parents leave so we may exile them as a reminder. Perhaps exile them and give them the ability to make offspring.

    1. I think breeding needs a mana cost. Maybe {3}, maybe more.

      Agreed. I don't even want to think about how breeder slivers would interact with static-effect-granting slivers.

      Yeah, M14 had a frustrating number of Slivers. They could help the slivers Standard deck that existed for 3 months, or Modern Slivers, but they were nearly useless in Limited. We could include creatures of the chosen type without breed to increase the number of combinations while limiting token making.

      It's true. Maybe the cost could scale with the number you have. Pay {X}, where X is the number of 'slivers' you control.

      That no longer reads like parentage, but I could see a mechanic where two cards exile themselves under a token and add their stats to it, provided that those cards come back when it LTB, or otherwise compensate you for 2-for-1'ing yourself.

    2. I picutred the two were off in some breeding pool so they couldn't fight but they were making tokens a la Soul Foundry or something.

    3. Could we perhaps make it so the babies come from putting a card (perhaps a creature card?) from your hand face down into play, so it is a base 2/2 but then gets these abilities?

    4. That does drastically reduce the power of breeding, which is good. Practically, every (eligible) card in your hand can be played as an A+B Sliver. That's got potential.

    5. I love this idea. I would love it to see print, but with minor changes people have addressed:
      1. Change type. We can create a new creature type with same feel but different name. That way your 2/2 is not buffed by any of the old slivers
      2. Board complexity. I am going to overcompensate this and say how about the discard a card idea and have the tokens be of a different type, like spawn. Either that or have them not be able to breed by themselves.
      3. You are relying on flavor to define the term "parents". While the word choice is amazing flavorfully, it means nothing in goame.

    6. Everyone's saying that this should be a different creature type so it doesn't inherit abilities from old Slivers, but also that it'll be difficult to create enough different cards with different abilities. I thought it was a deliberate part of the old design that if you breed, say, an Oakshell Sliver with a Winged Sliver, the child will grant flying to all your Slivers even if the parent Winged Sliver dies.

      As to Theo's point 3, new mechanics add new game terms to the rules all the time. It'd be as simple as a one-sentence addition to the Comp Rules saying "During the resolution of an activated ability, the term 'parents' means 'creatures which were tapped to pay nonmana costs to activate this ability'."
      Or it could be a subrule within the Comp Rules definition of the Breed keyword ability. Either way, it's fine. The word "parents" is perfect flavour-wise for communicating precisely what we want, and it can have the rules to back it up where necessary.

      ("Nonmana" in order to avoid extra abilities proliferating if you use Llanowar Elves to pay the ability's mana cost, etc. I /think/ that closes the most obvious loopholes.)

      That said, I think I agree that this mechanic might be better suited to a separate game.

    7. FWIW, I assumed the comp rules would define breed as something like "it has the abilities of each of those creatures", but the reminder text would say "parents".

      Wizards has done this before, choosing shorter clear flavour text, and I think it has to be done very carefully to ensure it's completely unambiguous. But when it is, I think it's really great -- if the reminder text can be shorter AND clearer, that's better all round :)

  6. Sacrificing the token at EOT helps power level and complexity significantly.

  7. honestly, just keeping the abilities tied to the slivers seems like the best fit. It's a lot of the same power level, but WAY easier to track.

    1. When you say keep the abilities tied to the Slivers, what do you mean. Do you mean Breed? Like only the parents have Breed and the tokens don't? Or the tokens get nothing beyond being Slivers?

  8. Based on multiple excellent comments:

    Breeder ({3}, {T}, Tap an untapped creature you control with breed: Put a card from your hand OTB face-down. It's a 2/2 colorless creature. It gains all abilities of its parents, except Breed.)

    Using cards eliminates the problematic card advantage, and not passing breed to children reduces the number of possible variations (which is both good and bad).

    Breeding ({3}, {T}, Tap an untapped Sliver you control: Put a 2/2 colorless Sliver creature token OTB with all its parents' abilities. Sacrifice each Sliver token you control that isn't identical to it.)

    If there can only ever be one type of spawn running around at a time (per player), remembering what they are becomes much easier.

    Not really sure we can combine both ideas. Prove me wrong?

    1. Would it be cleaner to say "~ and its offspring enter the battlefield with a +1/+1 counter." and let the offspring itself be vanilla?

    2. Might take a bit less text, and could reduce too-much-power, but it's not as thematically intuitive. I won't lose my hazel eyes when my mom dies.

    3. Agree that is the tradeoff. However, I think it is much less text, and provides a clear way to show the board state (using beads like some people did for Soulbond) so it totally removes memory issues.

      I could believe that, in a short term setting, you have flying because your mom is there to show you how to fly.

    4. That's what I was trying to suggest above.

      WInged Breeder 1U
      Creature Sliver
      Slivers you control have flying.
      2, t, tap an untapped sliver you control: Put a 1/1 colorless sliver token onto the battlefield.

      That's just the way slivers work. You won't forget how to ride a bike when the person who taught you dies, but for slivers it's just like "Oh, Bob's dead, guess I forgot how to Trample."

    5. I think that loses too much of the heart of the design. As has been said, these aren't really Slivers.

    6. Think of the surprise potential when your sliver turns out to have Morph!

    7. I thought about that, and I think that part is awesome, but sadly I don't think both abilities can go in the same block, because in competitive play it is important to be able to verify if everything face down was put there legitimately.

      But Breed and Morph have different "check times" for that... Breed you check when you put the card face down, and Morph you check when the game is over. If you didn't remember whether a card was put face down with Morph or Breed you wouldn't know what to do at the end of the game.

      I think there is potential in the block having a mechanic that is like Morph without the ability to play it face down. If there was a second mechanic (other than Breed) that could enable putting cards face down in an interesting way, that would help.

  9. Doesn't this have rules issues if you Break Open an offspring that happens to be a face down instant or sorcery?

    1. Not really. Wizards have so far avoided anything that allows non-permanent cards to be on the battlefield face down, but I wouldn't necessarily expect that to still be true by the end of Tarkir block. Any time they want to do that they can just introduce a rule "an instant or sorcery that's on the battlefield is put into its owner's graveyard as a state-based action", alongside 400.4a.