Thursday, December 4, 2014

Abilities and layers, redux

I have touched this subject before, but here's a much simpler case:
  1. You control a Bonescythe Sliver, and someone casts Snakeform on it. Will your other slivers have double strike?
  2. You control a Stormtide Leviathan, and someone casts Snakeform on it. Will all lands be Islands in addition to their other types?

Instinctively the answer to both questions would be "no", but of course it's not that simple.

In the first case, even though the Sliver's ability has an earlier timestamp than the ability-removing effect created by Snakeform (which normally would mean that all other slivers get first strike before the Snakeform effect removes Bonescythe's abilities), these are dependent effects. In other words, both continuous effects exist on the same layer (the "ability adding/removing" layer, ie. layer 6), and the existence of one depends on the other. This makes them dependent, and for this reason Snakeform's effect is always applied before Bonescythe's, regardless of their relative timestamps. Which means that no slivers get first strike. This makes it intuitive even without knowing about layers, timestamps and dependent effects.

What is much less intuitive is what happens in the second case. One could easily think that it works the same way: Snakeform removes the ability, and thus lands are not Islands in addition to their other types. However, it doesn't work like that.

The Leviathan's effect is not an ability-adding effect, but a type-changing effect. Types are set in a lower layer (more precisely layer 4). This means that the effect is evaluated before any ability-removing effect.

These two effects are not dependent because they are on different layers. Only effects on the same layer can be dependent.

This means that even though Leviathan ends up not having any abilities, all lands will still be Islands in addition to their other types, as unintuitive as that may be.

Friday, October 3, 2014

Doubling season... or is it?

This is Doubling Season (click on any card image to see a larger version):


It has the text: "If an effect would put one or more tokens onto the battlefield under your control, it puts twice that many of those tokens onto the battlefield instead. If an effect would place one or more counters on a permanent you control, it places twice that many of those counters on that permanent instead."

Seems simple enough. Here are some statements. Can you figure out for each one if the statement is true or false? All of the statements assume that you control a Doubling Season.


1) You control a face-down Hooded Hydra, and you turn it face up by paying its morph cost. The hydra will only get five +1/+1 counters.


Hooded Hydra says: "As Hooded Hydra is turned face up, put five +1/+1 counters on it."


2) You control a face-down Hooded Hydra, and your opponent casts Break Open targeting it. It will get ten +1/+1 counters.


Break Open says: "Turn target face-down creature an opponent controls face up."


3) You control a Words of Wilding, and you activate its ability during your upkeep, and proceed to the draw step. When you would draw your card, you instead put a 2/2 green Bear creature token onto the battlefield. Only one.

  

Words of Wilding says: "1: The next time you would draw a card this turn, put a 2/2 green Bear creature token onto the battlefield instead."


4) You control a Words of Wilding and an Alchemist's Apprentice. You activate the ability of the former, and then the ability of the latter. Instead of drawing a card, you instead put two 2/2 green Bear creature tokens onto the battlefield.


 Alchemist's Apprentice says: "Sacrifice Alchemist's Apprentice: Draw a card."


5) You play a Vivid Grove as your land drop. It will enter the battlefield with just two charge counters.


Vivid Grove says: "Vivid Grove enters the battlefield tapped with two charge counters on it."


6) You control a Sakura-Tribe Scout, and you use its ability to put a Vivid Grove from your hand onto the battlefield. It will enter the battlefield with four charge counters.


Sakura-Tribe Scout says: "(Tap): You may put a land card from your hand onto the battlefield."


It may be a surprise to many players, but the answer to each one of the statements above is "true".

Why does Doubling Season sometimes double the number of counters/tokens, and sometimes it doesn't? The reason for this is quite obscure and technical.

The key word here is Doubling Season's wording "if an effect would put..." It's also related to a technical aspect of replacement effects.

The word "effect" in the card's text is not just random or inconsequential. The rules of the game explicitly define "effect" as meaning "spell or ability". In other words, what the text in Doubling Season is really saying is, effectively: "If one or more tokens would be put onto the battlefield under your control as a consequence of a spell or ability..." (and the same for counters).

Turning a face-down permanent face-up by paying the permanent's morph cost is not activating an ability. Instead, it's a special action (this is explicitly specified in the rules), and special actions do not count as activated abilities. Hooded Hydra's "As Hooded Hydra is turned face up..." is a replacement effect (as specified in the rules), and replacement effects do not change the nature or type of the event they are affecting; they simply modify the event in some way. Thus the reason why the hydra is getting counters is a special action, not a spell or ability, and thus Doubling Season has no effect on it.

If the hydra turns face-up due to a spell like Break Open, this changes things. Now it's a spell that's turning it face-up, and thus Doubling Season now has an effect on it.

Drawing a card at the beginning of the draw step is not a spell or an ability either, but a so-called turn-based action. A replacement effect like that of Words of Wilding doesn't make it anything else. Even though the event was modified, it's still a turn-based action, and thus Doubling Season has no effect on it. Thus you get only one bear token.

However, if you are drawing a card with an ability like the one on Alchemist's Apprentice, this once again changes things. Now it's not a turn-based action, but a resolving ability that's making you draw a card. Thus Doubling Season affects it (or rather, the modified version of it by Words of Wilding).

Likewise playing a land is also a special action. It's not casting a spell nor is it an ability. This is why playing a vivid land as your land drop won't double the counters on it. However, putting a land onto the battlefield due to a spell or ability (like the one on Sakura-Tribe Scout) does double the counters.

(Note that this is a consequence of the rules that may change in the future, as the rules coordinators have expressed the sentiment that they may want to change this so that it becomes more intuitive. The rules work like above as of writing this blog post, but may become different in the future.)

One of the reasons why Doubling Season is worded like that (ie. "if an effect would put..."), and why it's so difficult to change so that the abovementioned interactions would become more intuitive, is that they don't want it interacting with things like costs that put counters on permanents (such as activating planeswalker abilities).

Wednesday, September 24, 2014

Goblin Test Pilot shenanigans

There's a handy combo that can be used to get an unlimited amount of copies of almost any creature, which consists of enchanting said creature with Splinter Twin and having an Intruder Alarm on the battlefield. (This way each time you create a copy of the creature, Intruder Alarm will untap the creature, which means you can create another copy, and so on.)

It's fun to think what happens with certain creatures when they are copied without limit. One such case is Goblin Test Pilot. This can create some very interesting rules nightmares.


Goblin Test Pilot has power 0, so it can't be used to win the game by attacking (unless something boosts all your creatures). However, it has an ability that deals 2 damage to a target creature or player. (Because the copy will have haste, you can use it right away.) The thing is, the target is chosen at random. This creates interesting situations.

The game supports the notion of "shortcuts". In other words, if you wanted to create a thousand copies of the goblin, you don't have to go through the motions one at a time, but you can propose a shortcut. You can simply demonstrate how the combo works and say "I create a thousand copies of the goblin; they'll all end up untapped at the end." If your opponent has nothing to interrupt the combo, then they can agree with the shortcut.

However, suppose that now you would want to activate the ability of each copy. Can you shortcut this? Normally you could, but in this case you can't because the target is chosen at random. You pretty obviously just can't propose a shortcut for a repeated random effect. You would need to choose the random target individually for each activated ability. So what exactly are you supposed to do in this case? You can't take forever to activate them.

This isn't just a "troll" combo, however. There's a way to arguably use this combo to win, by following this procedure:
  1. If there are no untapped goblin copies, create a copy and let Intruder Alarm untap all creatures.
  2. Activate the ability of an untapped goblin copy.
  3. If it targets your opponent, one of his or her creatures, or one of your goblin copies, let it resolve and then jump to step 1.
  4. Else (in other words, if it targets you or your original goblin) do not let it resolve and instead jump to step 1 in response.
While this is not a formal proof, this informal argument can be used to show that you will eventually deal enough damage to your opponent to win the game, no matter how many iterations it takes. It doesn't really matter if this takes ten iterations or ten million iterations, eventually you will have dealt enough damage to your opponent to win.

Can you propose this kind of shortcut? The official answer is no: You can't. The reason is that the number of iterations is not fixed, but random, and that just can't be shortcut. The rules of the game don't support this.

This creates a curious situation where, at least in a tournament, you would probably be disallowed from running this loop, or even imposed a game loss for stalling, even though there's nothing illegitimate about the combo and you are not really stalling. You are just doing what the cards say, without any kind of stalling. However, if the loop is taking hours, you won't be able to just continue it in an actual tournament. You can't shortcut this either (at least if the judges are strict with the rules of the game.)

Monday, September 22, 2014

Is pile-shuffling useful?

I posit that it is not, and here's my argument.

"Pile-shuffling" is the method of distributing the cards into a number of piles, and then just collecting the piles together. This is rather different from the traditional "mash shuffling" and "riffle shuffling" methods.

This is an important fact: Pile-shuffling does not evenly randomize your deck. This might sound like something self-evident and trivial, but nevertheless surprisingly many people have all kinds of misconceptions about it. Some people believe that pile-shuffling two times into a different number of piles is "sufficient randomization". Many even mention prime numbers (even though they have nothing to do with how randomized the deck will become.) Pile-shuffling does not evenly randomize your deck even if you put each card on a random pile.

Assume that at the end of a game you simply collect all your lands and put them on top of your library, and then pile-shuffle for example into seven piles. This means that you know that each 8th card in your deck will be a land. This is obviously not even randomization. Even if you couldn't tell which land will come after the next seven cards, it's still not randomized because you know for certain that it will be a land. This ought to be pretty obvious.

Moreover, if you did this, it would be effectively mana-weaving (ie. deliberately distributing your lands evenly in your deck, with no randomization afterwards), which is illegal.

Most players who pile-shuffle will mash-shuffle or riffle-shuffle extensively afterwards, and they argue that this is sufficient randomization which nullifies the "mana-weaving", and therefore it's not cheating. And that's the crucial point: If they are going to sufficiently randomize it afterwards, undoing any ordering that the pile-shuffling achieved, then the pile-shuffling was a complete waste of time because it did nothing useful.

Pile-shuffling does not increase the randomness of the deck (by its very nature), and the deck is going to be randomized afterwards anyways, so why exactly do it? What's the use?

Some may ask "why not? What's the harm?" There's no other harm except that pile-shuffling takes time, and thus it's a complete waste of it. So why waste time on a procedure that achieves nothing useful?

Some people argue that pile-shuffling is useful to count your cards (in order to verify that no card ended up in the wrong place during a game, and that there aren't any extra cards). Ok, that's fine. However, especially tournament players often pile-shuffle several times. These extra shuffles do not have even this excuse.

The main reason why people pile-shuffle is because they feel that normal mash-shuffling doesn't "separate" cards enough from each other. They feel that pile-shuffling separates the cards from each other in the deck, and thus increases randomness and evens out the deck more (especially in terms of land distribution). But this is once again borderline mana-weaving: If you feel that your mash-shuffling is not randomizing the deck enough, and thus you feel the need to pile-shuffle to compensate, then you are borderline mana-weaving, ie. borderline cheating. If you argue that your mash-shuffling is enough to randomize the deck afterwards (and thus undo any possible mana weaving), then you are arguing that the pile-shuffling was indeed useless and a waste of time.

I find it curious how defensive people get when these things are brought up. Many, if not most, players will vehemently defend pile-shuffling no matter what arguments against its usefulness are presented. I have yet to find a player that simply agrees.

(Often the conversation will quickly start going in circles. When you argue that pile-shuffling effectively amounts to stacking/manaweaving your deck, they will argue that the mash shuffling afterwards will undo any such stacking. When you then point out that they just confirmed that the pile-shuffling achieved nothing useful because of the mash shuffling, they will insist that it was useful because it separates clumps of lands etc. In other words, you can point out, they are indeed stacking/manaweaving with the pile shuffling. And in circles the conversation will go...)

If you don't think pile-shuffling is all that slow, watch this video. (Note that the several pile-shuffles the player does eg. starting at 2:25 in the video have zero benefit in terms of deck randomization because of the rather massive amount of mash shuffling done in addition to them.)

Wednesday, September 10, 2014

An oddly-worded replacement effect

In Magic, replacement effects are very recognizable because they typically use the word "instead". In fact, the vast majority of them conform to the template:

"If (something) would (do something), (do something else) instead."

For example, Rest in Peace says:

"If a card or token would be put into a graveyard from anywhere, exile it instead."

This form is so ubiquitous that when one gets used to it, a card like Knight of the Holy Nimbus sounds very strange.


It says: "If Knight of the Holy Nimbus would be destroyed, regenerate it."

Note how it almost conforms to the standard template, except that it does not use the word "instead". In fact, it doesn't seem to conform to any standard template for replacement effects. These templates are defined in rule 614.1, and are (paraphrasing):
  • Effects that use the word "instead".
  • Effects that use the word "skip".
  • Effects that read "[This permanent] enters the battlefield with ...", "As [this permanent] enters the battlefield ...", or "[This permanent] enters the battlefield as ..."
  • Effects that read "As [this permanent] is turned face up ..."
None of those templates seem to fit Knight of the Holy Nimbus. So what exactly is it?

It turns out that this is a special wording defined in rule 701.12b. Said rule effectively states that

"If [this permanent] would be destroyed, regenerate it."

means

"If [this permanent] would be destroyed, instead remove all damage marked on it and tap it, and if it's an attacking or blocking creature, remove it from combat."

The "instead" is implied in this particular template.

Monday, August 18, 2014

Casting an opponent's commander

Question: In the Commander game format, is it possible to construct a situation where you can cast an opponent's commander from the command zone?

Of course the answer is "yes" (and the image below spoils the answer), but the interesting rules question is why it's possible.

One way to achieve that is to make the opponent's commander go to the top of that opponent's library (for example with Excommunicate), and then attack with an unblocked Daxos of Meletis. After that, even if your opponent decides to put the commander in the command zone instead of exiling it, you can nevertheless cast it.


This is a little quirk of the rules of the game (and is in fact something that WotC fully acknowledges as a valid thing to do. In fact, a recent rules change clarifies the situation, making it even clearer.)

The key reason why this is possible is because the effect uses the wording "that card", rather than for example "the exiled card". (For example the same would not work with something like Oblivion Ring, which says "when Oblivion Ring leaves the battlefield, return the exiled card to the battlefield under its owner's control." Because it uses that wording, if the card is not in exile when the ability resolves, it won't do anything. This is delineated in the rules of the game.)

Because Daxos uses the wording "that card", if the exiling is replaced with the card going to another public zone, the effect will keep track of it (until it changes zones to somewhere else, after which it loses track of it because it's now a different object.) Note that it won't be able to track it if the exiling is replaced with going to a hidden zone.

So the next logical question is: What about the commander mana tax? Do you have to pay 2 extra mana for each time it has been cast from the command zone?

You don't pay the owner's mana tax. Instead, you have your own mana tax imposed for that particular commander (and which is independent from the mana tax of your own commander.) In other words, the first time you do this, there is no tax. However, if you manage to repeat the same feat with the same commander during the same game, you'll now have to pay 2 extra mana to cast it (and so on.)

Monday, August 4, 2014

Damage replacement and redirection

Suppose you control a Force Bubble and a planeswalker (it doesn't matter which for the intents of this question; I'm putting Jace in the image just as a random example). Your opponent casts a damage spell, like Lightning Bolt, targeting you. Question: Is it possible for your opponent to redirect that damage to the planeswalker or not?


An intuitive answer would be that "of course not; the Bubble is replacing the damage with something else, so there's nothing to redirect to the planeswalker." However, it's not that simple. This is actually a bit contrived part of the game.

The Bubble's ability is a replacement effect. The redirect-damage-to-planeswalker rule is also a replacement effect (this is defined so in the rules of the game. It doesn't really matter that the replacement effect is optional, it still applies here.) Thus there are two replacement effects trying to modify the same damage. So what happens?

The rules state in this case that the affected player (ie. in this case you) decides the order in which the two replacement effects are applied. Thus you have two options:

If you decide that the Bubble's effect is applied first, then the damage is replaced with putting counters onto the Bubble. After that the planeswalker-redirection effect does nothing because there's nothing to replace anymore.

However, you could also choose to have the redirection effect apply first (in most cases this wouldn't be sensible, but in some cases it could well be.) What happens then?

The effect in question now asks your opponent whether he or she wants to redirect the damage to the planeswalker. If your opponent does, then the damage is redirected, and then Bubble's effect does nothing (because likewise there's nothing to replace anymore).

However, if your opponent does not want to redirect the damage, now the Bubble's replacement effect is applied, and in this case there is something to replace, so the damage is replaced with putting counters on the Bubble.

So this is an interestingly contrived situation. To recapitulate, these are all the options:
  1. You decide to apply Bubble's effect first. The damage is replaced with counters on the Bubble, and the other effect then does nothing.
  2. You decide to apply the redirection effect first. Now your opponent chooses whether to redirect the damage to the planeswalker or not.
    1. Your opponent decides to redirect, thus the planeswalker gets the damage and nothing else happens.
    2. Your opponent decides not to redirect. Then the Bubble's replacement effect is now applied, and it takes place.

Monday, June 9, 2014

Phasing out Banisher Priest

Suppose you control a Banisher Priest enchanted with a Splinter Twin. You make a token that's a copy of Banisher Priest, and exile some creature with it. Then someone phases the token out using Reality Ripple. Question: Will the exiled creature return to the battlefield?


Banisher Priest has the following ability: "When Banisher Priest enters the battlefield, exile target creature an opponent controls until Banisher Priest leaves the battlefield" (emphasis mine.)

An object that phases out is not considered to change zones nor leave the battlefield.

A token that phases out ceases to exist as a state-based action. As far as I know, it's undefined by the rules whether "ceases to exist" constitutes "leaving the battlefield".

Likewise it's apparently not defined whether Banisher Priest's ability (or any of such new-style temporary exiling abilities) will end when the object they are tracking phases out. (After all, they are looking for the event "leaves the battlefield", and this event never happens.)

It seems to be that (as of writing this post) this is yet another situation where the rules of the game are incomplete, as they do not clearly and unambiguously define what should happen.


Update: It seems that the official interpretation from Wizards is that the creature does not return and will remain exiled indefinitely.

Friday, June 6, 2014

Humble God

Assume that there's a Humility on the battlefield, and you control Heliod, God of the Sun. Question: Will Heliod be indestructible? But more interestingly: Will your other creatures have vigilance?


(Note that the Oracle text of Humility is: "All creatures lose all abilities and are 1/1." Humility doesn't make anything a creature, nor does it remove any types.)

There are four possible combinations here depending on which one entered the battlefield first, and whether your devotion to white is less than five or not.

The first question depends solely on your devotion to white. If it's less than five, the question is whether Humility removes the ability that makes Heliod a non-creature or not. Since Humility only affects creatures, and Heliod's ability may make it a non-creature, this might at first seem like a chicken-and-egg problem. (If Humility removes the ability, then there's nothing to make it a non-creature, but if the ability makes it a non-creature, then Humility won't affect Heliod.)

However, the seeming problem is resolved thanks to the layering system. Heliod's ability is evaluated in layer 4 (the layer where type setting and removing abilities are evaluated). This happens before evaluating Humility's ability, which is done in layer 6 (the layer where ability adding and removing is evaluated.) Thus Heliod's ability trumps Humility's, and thus if your devotion to white is less than five, Heliod won't be a creature, Humility won't affect it, and it will be indestructible. If your devotion to white is five or more, then Heliod will be a creature and Humility will remove all of its abilities and make it 1/1.

The more complex question is whether your other creatures will have vigilance.

If your devotion to white is less than five, and thus Heliod is not a creature, then it depends on timestamps. This is because there are now two effects that add or remove abilities (and thus are in the same layer.) If Heliod entered before Humility, then your other creatures won't have vigilance, but if Heliod entered after, they will.

Something more complicated happens if your devotion to white is five or more, and thus Heliod is a creature. This is because now the two abilities are dependent: They exist on the same layer, and the existence of one depends on the other. Thus in this case Humility's effect is evaluated first regardless of timestamps, and thus no creatures will have vigilance.

Monday, May 26, 2014

Drawing while casting

There's a lesser-known rule in Magic the Gathering that says:
401.5 If a spell or ability causes a card to be drawn while another spell is being cast, the drawn card is kept face down until that spell becomes cast (see rule 601.2h). While face down, it’s considered to have no characteristics. The same is true with relation to another ability being activated. If an effect allows or instructs a player to reveal the card as it’s being drawn, it’s revealed after the spell becomes cast or the ability becomes activated.
What this means is that if a card is drawn during the process of casting a spell (the process of casting a spell starts by the player revealing the card he or she is going to cast, and ends with its costs having been successfully paid, and involves calculating said costs in between), the drawn card is kept face down and considered to have no characteristics until the spell has been fully cast.

This rule doesn't come into play very often. To my knowledge there currently exists one single card in the entire game where this rule applies (and if I'm not mistaken, this rule was in fact added precisely because of this single card). The card in question is Chromatic Sphere.


Because the ability in Chromatic Sphere is a mana ability (it does not target and it adds mana to the player's mana pool) it can be activated when paying the cost of a spell, ie. during the casting process of said spell.

What the rule is thus saying is that if you activate Chromatic Sphere while casting a spell, the drawn card will be kept face down and considered to have no characteristics until the spell has been successfully cast (ie. it enters the stack.)

Now, an interesting question is: What happens if, for example, the drawn card is now discarded, still during the casting process of the spell? What if the drawn (and then discarded) card is, for example, Progenitus?

So, consider this situation: You start casting a spell. When it comes time to pay its mana cost, you activate the Chromatic Sphere, drawing a card (which in this case happens to be a Progenitus). You then activate the ability of a Bog Witch and discard the card you just drew. The question is: Will Progenitus be shuffled into the library or not?


The Progenitus card will be face down and without characteristics during this whole process (because the abilities of both the Chromatic Sphere and the Bog Witch are mana abilities, and activated here during the casting process of a spell). Since the discarded card has no characteristics, will its replacement effect happen?

(Note that the "reveal Progenitus" part doesn't exist any more than the replacement effect itself, if the card has no characteristics, so it can't be used as an argument for the replacement effect applying.)

The jury is still out on this one. It seems that this is an unclear situation, ie. a situation where the rules may not give an unambiguous answer. Some arguments can be made for the replacement effect happening, and other arguments for it not happening. The rules don't seem to give a definitive answer.

(In a tournament setting this would in practice be up to the judge to decide.)

Tuesday, May 13, 2014

Nylea, Blind Obedience

Assume your opponent controls Blind Obedience, your devotion to green is four, and you play Nylea, God of the Hunt. Question: Will it enter the battlefield tapped or not?


Nylea is a creature spell on the stack. (Well, technically speaking an enchantment creature spell, but that's not important here.) Its ability that makes it not be a creature if your devotion to green is too low doesn't come into effect until it's on the battlefield.

In fact, if we look at the whole process of the spell resolving, there isn't a single moment where the effect kicks in and it stops being a creature: As soon as it hits the table, your devotion to green will be five, and thus the ability does nothing. There isn't even a single moment where Nylea isn't a creature.

Because of all the above, the intuition would be that it enters the battlefield tapped. However, that's not so. (Many players will have the intuition that it enters untapped and, while this is the correct answer, might not know the exact reason.)

This is once again a question of how an "enters-the-battlefield replacement effect" works for a resolving permanent spell. What's happening here is this:

As the Nylea spell starts resolving, it will look if any ETB replacement effects will affect it. In this case there's one such effect that may potentially affect it: That of Blind Obedience. Said effect affects artifacts and creatures, and thus what we are looking for is the question: "Is this an artifact or creature when it hits the board?" If the answer is yes, then Blind Obedience's replacement effect will take place, else it won't.

So we "look forward in time" to see how Nylea would look on the battlefield (to see if it's an artifact or a creature there.) There are certain things that are taken into account when doing this. One of those things are abilities on the card that affect the permanent itself (rather than an ability that affects a group of permanents with certain characteristics), and Nylea has such an ability: "As long as your devotion to green is less than five, Nylea isn't a creature."

In other words, we look if Nylea would be a creature on the current battlefield, taking into account that ability. And the answer is: No, it wouldn't be a creature. And the reason for this is that your devotion to green at this moment is four, and thus the ability stops it from being a creature.

This is quite a curious situation. "Devotion" only looks for permanents that are already on the battlefield. It doesn't see this "hypothetical permanent" that we are handling here (for the purposes of determining whether it's a creature or not.) Nylea isn't on the battlefield yet, and thus doesn't change your devotion to green.

Thus the full chain of events is: The Nylea spell starts resolving, we "look forward in time" to see how it would look like on the battlefield, we see that it wouldn't be a creature (because your devotion to green is four), thus Blind Obedience does not affect it, and then it finishes resolving and enters the battlefield untapped as a creature (because now your devotion to green is five.)

Sunday, April 20, 2014

"Game state" in Magic

One easily gets the impression that the Comprehensive Rules of Magic: The Gathering read like a legal document that defines the game with mathematical precision, to the utmost detail. However, this isn't always so.

One particular example is the concept of "game state". This term is used many times in the rules but, quite surprisingly, is not defined anywhere, not even vaguely or in passing. The exact definition of the concept can become important especially when the rules talk about "the same game state" (ie. when the "game state" is the "same" before and after an event.)

Practical example: Assume that it's your turn and you have a combo that allows you to increase the power of a creature without limit, and your opponent has a combo that allows them gaining an unlimited amount of life. If you say "I increase the power of my creature by a million", your opponent may respond to it with "I gain two million life". Are you then allowed to further increase the power of your creature (thus allowing your opponent to further increase their life total and so on, potentially ad infinitum)?

The answer is no: You can't repeat this process forever. The following rule forbids it from being continued:
716.3. Sometimes a loop can be fragmented, meaning that each player involved in the loop performs an independent action that results in the same game state being reached multiple times. If that happens, the active player (or, if the active player is not involved in the loop, the first player in turn order who is involved) must then make a different game choice so the loop does not continue.
This means that you cannot further increase the power of your creature (at least not without doing something else first.) Note, however, the bolded part.

Surprisingly, the concept "the same game state" is not explicitly defined anywhere in the rules.

It cannot mean "the game state is identical" because in this particular example it's not: The power of a creature has changed, as well as the life total of a player. Clearly something has changed, and thus the game state is not identical to what it was before. However, officially this rule still applies to this situation (and thus you are not allowed to repeat the combo before doing something else first.)

So what does "the same game state" mean? It's undefined by the rules, and therefore up to a judge to decide. One can come up with an intuitive definition (ie. something that the people who wrote that rule probably had in mind), but this is nevertheless up to a judge to decide.

Friday, April 18, 2014

Phasing out equipment

Assume that you control a token creature that has been equipped with an equipment (for the sake of simplicity, let's say for example Flayer Husk is equipped to the token it creates). Then someone casts Reality Ripple targeting the token creature. What happens?


The token creature will be phased out. What happens to the equipment? The following rule specifies what happens:
702.25f When a permanent phases out, any Auras, Equipment, or Fortifications attached to that permanent phase out at the same time. This alternate way of phasing out is known as phasing out "indirectly." An Aura, Equipment, or Fortification that phased out indirectly won’t phase in by itself, but instead phases in along with the permanent it’s attached to.
The equipment phases out alongside the creature. Because it phased out "indirectly", it won't phase in automatically during the player's next untap step.

But what happens to the token when it phases out?
702.25k Phased-out tokens cease to exist as a state-based action. See rule 704.5d.
In other words, both the token and the equipment will phase out, then the token will immediately cease to exist, and... the equipment will never phase in again, because it was phased out indirectly.

The equipment will remain phased out for the rest of the game (unless something specifically phases it in.) A phased-out permanent is treated as though it doesn't exist at all.

This is even more of a "remove from the game" than exiling is...

Thursday, April 17, 2014

Swirl the Mists

Swirl the Mists is an enchantment that says:
As Swirl the Mists enters the battlefield, choose a color word.
All instances of color words in the text of spells and permanents are changed to the chosen color word.
Assume that a player has cast it earlier, choosing white as the color. Then you cast Ulasht, the Hate Seed, which says:
Ulasht, the Hate Seed enters the battlefield with a +1/+1 counter on it for each other red creature you control and a +1/+1 counter on it for each other green creature you control.
Will it look for red and green creatures you control, or white creatures you control?


Ulasht will have the color words in its text box replaced with "white" when it's a spell on the stack, and also when it becomes a permanent on the battlefield. There isn't a single moment, not even for a brief moment, that the color words are reverted back during the resolution of the spell. They are changed to "white" and remain so all the way, without ever reverting back.

This would seem to give a pretty easy answer to the question: It will look for white creatures you control when it's entering the battlefield?

Well, no. That's the incorrect answer. It will look for red and green creatures, as the original text says.

This is one of the most unintuitive, obscure and quirky aspects of the rules of the game. It's so quirky, in fact, that one could justly argue that it's an outright oversight (albeit probably one that would be difficult to fix without breaking something else.)

The thing is, Ulasht's first ability is an ETB replacement effect, and such replacement effects look "forward in time" to see how the permanent would look like when it's on the battlefield (before it actually happens). There's a specific list of things that are taken into account when looking at this hypothetical future permanent, and it so happens that Swirl's static ability is among some of the things that are not included in the list.

This means that the ETB replacement effect will look at how the permanent would look on the battlefield without Swirl's static effect, as if it didn't exist at all. In other words, the hypothetical future permanent will look unmodified, and thus the replacement effect will look for red and green creatures (even though the actual spell, and the permanent it becomes, at no point reverts back to its original wording.)


Edit (Oct 2017): New rules changes with the Ixalan set have changed this. After this rules change Ulasht will look at creature of the color specified by Swirl the Mists. It will get two +1/+1 counters for each creature of that color.

Sunday, March 30, 2014

Face-down permanents and copying

Assume that you cast a morph card face down, let's say for example Willbender, and then it becomes a copy of something else, like a Llanowar Elves (eg. via someone casting Mirrorweave targeting the elves.) What happens to the face-down Willbender?


Normally a face-down permanent is a 2/2 creature with no name, color or any other characteristics. So what happens if an effect makes it a copy of something else?

Quite many players, even very experience ones, will intuitively say that it simply becomes a copy of that something else (ie. in this case it will become an exact copy of Llanowar Elves, ie. a 1/1 green creature with that name, and with its ability.) Their intuition will usually be less sure about the question of what happens to its original morph ability, and whether it can be morphed or not. Some will guess that it retains its morph ability and can be morphed, but once morphed it will still be a copy of Llanowar elves.

Of course it's not that simple, and the situation is a bit more curious.

What actually happens is that it's the down-facing side (ie. the front of the card) of the permanent that becomes a copy of Llanowar Elves, not the up-facing side. The permanent will still be a 2/2 creature with no other characteristics even though it's now a copy of Llanowar Elves.

Since the down-facing side has no morphing ability now, it can't be morphed (for as long as it remains a copy of Llanowar Elves.) If Mirrorweave had instead targeted some other creature with a morph ability, then you could morph your face-down creature using the morph cost of that other card (if you can pay it.) However, when it turns face up, it will still be a copy of that other card.

The face-down permanent could even become a copy of a non-creature (for example if Mirrorweave had targeted an animated land.) While the down-facing side would become a copy of that non-creature card, it would still be a 2/2 creature with no other characteristics.

The reason for this is that being "face down" is a state of the permanent in the exact same way as being tapped or phased out. Being in the "face-down state" implicitly means that the permanent is a 2/2 creature with no other characteristics (regardless of what the actual card is.) It would work even if a permanent, like a land, would somehow become turned face down by some effect: It would become a 2/2 creature regardless of what it originally was.

However, note that copying in the other direction is special. In other words, if Mirrorweave had targeted the face-down permanent, then every other creature would have become a 2/2 creature with no other characteristics (rather than a copy of Willbender.) This is because the rules of the game specifically define the mechanic to work like that. (More exactly, the rules state that the visible characteristics of the face-down permanent, ie. "2/2 creature" become its copyable values.)

Tuesday, January 28, 2014

Mana abilities and Caged Sun

I discussed in my previous blog post when abilities are so-called mana abilities and when they are not, and how this can in some rare circumstances make a difference. Here is a rather unique problem related to this topic:

Is the triggered ability of Caged Sun a mana ability or not?


To understand this question we need to first fully understand what a "triggered mana ability" is. This is defined by the following rule in the Comprehensive Rules:
605.1b A triggered ability without a target that triggers from activating a mana ability and could put mana into a player's mana pool when it resolves is a mana ability.
The following rule makes it clear when a triggered ability is not a mana ability:
605.5a An ability with a target is not a mana ability, even if it could put mana into a player's mana pool when it resolves. The same is true for a triggered ability that could produce mana but triggers from an event other than activating a mana ability, or a triggered ability that triggers from activating a mana ability but couldn't produce mana. These follow the normal rules for activated or triggered abilities, as appropriate.
So, in summary, a triggered ability is a mana ability if it produces mana, does not target, and it triggers from activating a mana ability. If the ability targets, or it triggers from a non-mana ability, then it's not a mana ability even if it produces mana.

(A relevant difference between the two types of abilities is that a mana ability does not use the stack, and instead it resolves immediately when it triggers. It cannot be responded to. A non-mana ability uses the stack and can be responded to.)

So the question is: Is the ability of Caged Sun a mana ability or not? (Does it use the stack? Can it be responded to?) The ability in question reads:
Whenever a land's ability adds one or more mana of the chosen color to your mana pool, add one additional mana of that color to your mana pool.
Normally a mana-producing ability of any land is a mana ability, so it fulfills all the requirements for the ability of Caged Sun to be a mana ability.

However, the million dollar question is: What happens if a land has a non-mana ability that produces mana?

There exist currently no land cards that have such abilities. However, it's possible to create this kind of situation via gameplay. (One possibility is to, for example, make Deathrite Shaman into a land while retaining all of its abilities. This is possible by using certain combinations of cards.)

Technically speaking if a non-mana ability of a land produces mana like this, the ability of Caged Sun will trigger and it will not be a mana ability in this case according to the strict interpretation of the rules (and thus will use the stack.) In other words, whether the ability of Caged Sun is a mana ability or not depends on what's triggering it.

This is, as far as I know, not the intent of the card, or the game's rules in general. An ability ought to always be either a mana ability or not, rather than switching between the two depending on the situation. In other words, it ought to be a fixed characteristic of the ability that doesn't depend on anything else.

Probably for this reason the official ruling by WotC is that the ability of Caged Sun is always a mana ability, regardless of the situation (at least as of writing this post.) The interesting thing is that the comprehensive rules do not support this interpretation.

As of writing this post there is no disambiguating rule in the comprehensive rules to settle this question, but many people predict that it might be added there in the future.