Friday, December 30, 2011

How Expensive Should Removal Be?

No great truth was ever revealed to someone who blindly accepted what his parents told him. No revolution has come on the heels of accepting the status quo. Civilization is where it is because billions of people have figured out what's good and what's not for thousands of years, yet we haven't run out of innovation, reform or progress ...and we never will.

Question everything. You'll either learn why things are the way they are or how things could be better.

Suppose you're drafting Magic and open a pack with Djinn of Wishes, Condemn, Wurm's Tooth, Levitation, Doom Blade, Shock, Craw Wurm, Peek, Mighty Leap, Warpath Ghoul, Goblin Piker, Giant Growth, Maritime Guard and Siege Mastodon. What do you suppose the first four picks will be in the vast majority of groups? If you said the three removal spells and the big flier, we're on the same page. Why is that a problem?

It's important—nay, vital—to have answers. For every threat an opponent may present, it has to be possible for you to have a card that neutralizes it or else the game breaks down to who draws their unanswerable question first. It is similarly important that no removal spell is the best answer to all threats. Condemn can only deal with a creature that attacks, Doom Blade can only deal with nonblack creatures and Shock can only deal with small creatures. Iona's Judgment can handle any creature (without shroud, hexproof or protection from white) but it costs 5. If an efficient removal spell could handle any threat, it would be played ubiquitously and that's the surest sign of a degenerate metagame.

My question is this. Is it necessary that removal spells be so efficient? If you consider the vast majority of destruction effects that are played competitively, almost all of them cost 1 or 2 mana, unless they target a large number or type of permanents. Most threats, outside of Dead Guy Red, Suicide Black and White Weenie cost 3, 4, 5 mana and up. There's a definite discrepancy there. I don't know that it's wrong. But I do wonder.

One argument in favor of cheap removal is that if it cost as much to remove a threat as it did to play one, then the player on the draw might forever be on the defensive, always spending all of his resources to undo the advances the first player has made; If she misses once, she may never catch up. Another argument is that the applicability of a given removal spell is dependent on the threat, and so a discount to the cost of that spell offsets the opportunity cost that it may be irrelevant. These are not marginal factors.

But let's consider for a moment what would happen if removal were a bit more expensive. Would the game break or would it just play a little differently? Suppose Condemn cost 1W, Shock 1R and Doom Blade 2B. That cuts into your ability to play a creature and kill an opponent's creature in the same turn (by one turn's worth of mana) but does it unbalance the game in favor of one player over another? I can't be sure, but I don't think it does. If they've played a threat worth removing, your creature can wait another turn. If not, play your creature now and remove their threat later.

How do these less efficient cards fare in our pick order? I think Condemn and Doom Blade are still high picks because they answer so many things, but maybe Craw Wurm edges out Shock now, since it's not quite as vulnerable as it used to be. I would argue that's a good thing. Not because one type of card needs more help/attention/respect than another, but because the game is more skillful/interesting if the choice isn't obvious. How many people do you know that follow a route formula for draft picks? Bombs > removal > threats > utility > sideboard fodder.

In some ways, predictability is good for the game. That a brand new player can pick cards by the color of their expansion symbol and not make a complete fool out of themselves is a net positive. But ultimately, Magic is about exploration and discovery. The player who ventures beyond the status quo to discover that a deck full of Ironclaw Orcs can beat a deck full of Serra Angels deserves to be rewarded for the risk of innovation. Is it better to help new/casual players learn the game by making removal so good or is better to add depth to the game by muddying the lines of power?

In a world where an Ulamog, the Infinite Gyre can't be Path to Exile'd, maybe the game is more proactive. Maybe it's more about what you do and less about what you undo. I'm not suggesting removal do less, or even that all removal cost more (cheap threats still need cheap removal), but perhaps the average cost of "destroy most any creature" should be 3 instead of 2.

Modern Magic has seen a marked increase in the power level of creatures. Serra Angel used to be rare. Walking Dead used to be Scathe Zombies. Vengevine used to be... unprintable. That at least shows an awareness that creatures weren't on par with spells. But does it matter how good your creature is if it still dies to a handful of 1- or 2-mana spells? (Part of the reason the Titans were so good was they did something even if they died immediately.) The simpler solution could well be reducing the power level of removal.

I honestly don't know the answer. The fact that R&D has kept removal so efficient for so long may well mean there's a good reason for it. But possibly—like all of us until now—they've never really questioned it. Perhaps there's a better game just over the horizon and all we have to do is trot a bit off the beaten course for a while to find it. Then again, it could be all cactus out there.


  1. now THIS is a discussion I can get behind.

  2. I think you're looking at this the wrong way. Or, in another fashion, you're looking at this as someone who would overplay blue.

    The reason that answers are cheap is so that you can do something else on your turn/opponent's turn outside of play answer. YOu need not really look much further than Into the Mist.

    Into the Mist is a Counter Spell that bounces something back for 3UU, and as such is pretty unplayable. The reason that it's unplayable (as opposed to Rewind) is that it dedicates a huge chunk of your mana resources to do so. Unlike the similar Dissipate (which falls into our Cancel category), which by midgame actually lets you do something else.

    Increasing the costs of answers does more than just increase creature playability, it changes the whole tempo of the game. By forcing the choice between "counter" and "Cast", you are handing tempo control to the player who plays first.

    Consider this for a moment. Conventional Wisdom has changed over the years of magic. It used to be accepted that you could run a deck with less land than it currently is being run with. If you increase the price of counters, how much more land will need to be added to lands to provide enough mana to be able to play threats *and* have answers in your hand?

    R&D already has a different solution in mind, when one looks at cards like Divine Reckoning and Sever the Bloodline. Mass Removal might be going the route of "Get rid of most of something", instead of "Remove Everything". A lot of the removal in Innistrad is conditional, and I feel that we are going to see more of this Conditional Answers in the future.

    The way you reign in the effect is not to make it more expensive, but to make the answers specific to one problem, not all problems.

  3. First, they'd have to stop printing one-drops like Stromkirk Noble, Delver of Secrets, Reckless Waif, and Signal Pest. If you don't have a cheap answer to one of those, the game ends before it begins.

    Removal started out really strong (Lightning Bolt, Swords to Plowshares, Terror) and has largely stayed that way, whereas R&D has taken a long time to bring creatures up to the same power level. They can't nerf removal now without also nerfing creatures.

  4. I agree with Havelock when it comes to constructed. Making decks run separate answer suites for weenies and bombs would make a lot more games come down to drawing the wrong half of your deck.

    In limited however, the aggressive decks don't have infinite removal to spend, so defensive creatures can serve as much better defense against the weenies, allowing removal that deals with bombs to be weak against weenie hordes.

    We're also constrained in that we don't want all constructed quality removal at rare because then new players feel powerless against their friends. Basically, I think that shifting this for limited means we have to spend some extra uncommon slots, but that feels like a reasonable price to pay for more interesting drafting and deckbuilding where all of your removal spells aren't auto-includes.

  5. I agree with Havelock. Not only would cheap creatures would be hard to deal with, but late creatures would also be hard to deal with, because you can't run as many removal spells in your deck if the removal spells are clunky.

    It used to feel like every RB deck should play 4 Terror and 4 Bolts, every WU deck should play 4 Counterspells and 4 Swords to Plows, and then you decide what else can go in the deck. Now, because many high cost creatures provide immediate advantage, it makes it bad to commit very heavily to spot removal.

    So many creatures that look exciting by modern standards say "deal with me now or lose!" and that wouldn't be possible if removal wasn't also good.

    I don't think it's impossible for a card game to be balanced with higher costing removal, but it would be a new world order of balancing.

  6. Jule's plan looks doable to me, if the goal is to only nerf removal in Limited without doing so in Constructed. Competitive constructed would function the same if the good removal spells were uncommon rather than common.

    However, there might be a reason why good cards like Lightning Bolt or Doom Blade are at common.

    If I was a beginner, my creature cards would suck and I'll have very few land fixers and no Planeswalkers. If my removal sucked in addition to that, that might feel bad. It would also reinforce the feeling for beginners that it's all about spending money and upgrading ("There's a better, more rare version to everything!"), even if uncommons aren't a big deal to competitive players.

    Since removal cards have wide application, maybe that's why they're chosen as the good cards that beginners have access to.

  7. When it comes to limited, there are also pricier removal spells at common: Arachnus Web, Brink of Disaster, Chandra's Outrage, etc. So it's not true that all removal spells are so cheap as to be outright better than creatures. I'd play Gorehorn Minotaurs over Fling or Merfolk Looter over Ice Cage any day of the week.

    I don't like moving more removal to uncommon for the reasons Chah states, and also because it would simply make removal even more valuable in limited. (The value a removal spell adds to your deck is much higher if you're light on removal to begin with.)

  8. My example confused my message. Sorry about that. I'm not suggesting making _all_ removal more expensive. As I said, "cheap threats still need cheap removal:" We need Shock to answer Goblin Guide and Stromkirk Noble.

    I'm fully on board with having a range of removal spells where the cheaper ones answer fewer creatures. Shock handles creatures with toughness 2 or less. Deathmark gets white and green creatures. On the other end of the specturm, Into the Maw of Hell gets most everything *and* nets you a land. That's how it should be.

    And as Havelock points out, R&D has already started making more second-tier removal spells that fill a necessary role in Limited. Those are perfect and a step in the right direction.

    What I'm suggesting changing are the cheap cards that can handle anything or almost anything. Doom Blade is the big offender here. It kills about 80% of the creatures, big or small, cheap or expensive for just two mana. Sure, it can't handle Abyssal Persecutor but if Doom Blade cost 3 mana it would still be played in Constructed and it would still be a top pick in Limited, just not as ubiquitously.

    Cards like Pacifism, Journey to Nowhere and Oblivion Ring are also cheap answers to any creature but they are reversible. Cruel Edict kills anything, but only once you've reduced your opponents choices. Vendetta is super efficient, but costs you life. Condemn only hits attackers and gives away life. These are all fine.

    Every threat should have an answer, but not every answer should handle all threats. Not cheaply or without some vulnerability.

  9. Ah, I see your point. I largely agree, but there's something to be said for black being better at killing creatures. Yes, Doom Blade would still be a first pick at 2B, but constructed is a different story entirely.

    If you look at the top Standard decks from Worlds, those that run black are packing 2 or 3 Doom Blades, not 4. Some even have 0 or 1. They supplement them with various conditional removal spells: Go for the Throat, Wring Flesh, Black Sun's Zenith, Liliana of the Veil, or Tribute to Hunger. This doesn't strike me as degenerate or unbalanced in any way.

    So if Doom Blade is just one removal spell among many right now, what would happen if it cost 2B? I suspect it would be a second-tier card.

  10. That should be "_most of_ those that run black". A small minority do have 4 Doom Blades.

  11. Anyone know if Terminate was ever a constructed card? (I wasn't playing during Invasion, and I don't remember it seeing play during Shards-era Standard.)

    But on the topic of Shards block, I think that's a great resource for analyzing the balance between the power-level of creatures and the efficiency of removal. As I've been drafting it quite a bit on MTGO recently, it's incredible to see how much top-tier removal it has: Terminate, Path to Exile, Oblivion Ring, Branching Bolt, Maelstrom Pulse, Volcanic Fallout, Agony Warp, Infest, Resounding Thunder, and Executioner's Cap. Like, Agony Warp was a common and frequently resulted in absolute blowouts, but maybe it saw a little play in Cruel Control?

  12. I'm also amazed at how Shards of Alara had so much good removal. Part of it must be due to how each Shard had powerful threats that worked very differently. It's difficult to provide enough answers to such varied threats in the form of creatures.

  13. Terminate was definitely a constructed card. It saw significant play from Jund and to a lesser extent, 5cc; Agony warp got tested again and again in Fae but rarely made an appearance as more than a one-of. I remember seeing a number of cards get ignored/invalidated because paying BR was so easy for Terminate; which is interesting in hindsight as Terminate itself wasn't necessarily a 4-of in the face of Maelstrom Pulse and Lightning Bolt (which are also hyperefficient nearly-universal removal.)

  14. I know I'm responding to an ancient post, but I was just wondering... how does it feel to be proven right by Wizards, three years after the fact?