Absolutely. And I would hope everyone agrees on that part.
The point of nailing down how draws work is so that when a player is choosing between two decks the better choice is the one which achieves more KOs rather than the one which turns most of its losses into draws via some kind of stalling process.
Not really, that's the 4-1 rule. The debate here is about turning draws into wins; to go with the example that began this all, I feel that a deck that beats a Time Vault/Key deck 6-0 through Pithing Needle or other method that allows them to actually beat them to death should get a higher score than one that just has a counterspell for their win-con.
If it's hard to get a draw off someone (because any kind of repetitive game state instead becomes a loss), matchups become more 3-3 and 6-0... and 6-0's are easier to get.
It wouldn't be, but I think you've missed my point. If decks have a fallback position in which they force draws when they can't win then what you get is lots of 2-2s, not lots of 6-0s (and 3-3s).
4-1 has never been a common scoreline in XCB and probably never will be.
Private Mod Note
():
Rollback Post to RevisionRollBack
--
(I'm on on this site much anymore. If you want to get in touch it's probably best to email me: dom@heffalumps.org)
Forum Awards: Best Writer 2005, Best Limited Strategist 2005-2012
5CB PotM - June 2005, November 2005, February 2006, April 2008, May 2008, Feb 2009
MTGSalvation Articles: 1-20, plus guest appearance on MTGCast #86!
<Limited Clan>
Okay so I just spent my entire evening discussing this with Psly4mne and an L2 friend of eirs, and then discussing it further with some varyingly-helpful people in #mtgrules and #mtgjudge (<3 Brain), and I think I can, at last, present a coherent analysis of how these situations are currently covered by the rules.
Unfortunately it does not line up well with what we want for XCB, about which more later.
This is going to reference section 716 of the CR heavily, so I'm just going to spoil the whole thing right here:
716. Taking Shortcuts
716.1. When playing a game, players typically make use of mutually understood shortcuts rather than explicitly identifying each game choice (either taking an action or passing priority) a player makes.
716.1a The rules for taking shortcuts are largely unformalized. As long as each player in the game understands the intent of each other player, any shortcut system they use is acceptable.
716.1b Occasionally the game gets into a state in which a set of actions could be repeated indefinitely (thus creating a "loop"). In that case, the shortcut rules can be used to determine how many times those actions are repeated without having to actually perform them, and how the loop is broken.
716.2. Taking a shortcut follows the following procedure.
716.2a At any point in the game, the player with priority may suggest a shortcut by describing a sequence of game choices, for all players, that may be legally taken based on the current game state and the predictable results of the sequence of choices. This sequence may be a non-repetitive series of choices, a loop that repeats a specified number of times, multiple loops, or nested loops, and may even cross multiple turns. It can't include conditional actions, where the outcome of a game event determines the next action a player takes. The ending point of this sequence must be a place where a player has priority, though it need not be the player proposing the shortcut.
Example: A player controls a creature enchanted by Presence of Gond, which grants the creature the ability "{T}: Put a 1/1 green Elf Warrior creature token onto the battlefield," and another player controls Intruder Alarm, which reads, in part, "Whenever a creature enters the battlefield, untap all creatures." When the player has priority, he may suggest "I'll create a million tokens," indicating the sequence of activating the creature's ability, all players passing priority, letting the creature's ability resolve and put a token onto the battlefield (which causes Intruder Alarm's ability to trigger), Intruder Alarm's controller putting that triggered ability on the stack, all players passing priority, Intruder Alarm's triggered ability resolving, all players passing priority until the player proposing the shortcut has priority, and repeating that sequence 999,999 more times, ending just after the last token-creating ability resolves.
716.2b Each other player, in turn order starting after the player who suggested the shortcut, may either accept the proposed sequence, or shorten it by naming a place where he or she will make a game choice that's different than what's been proposed. (The player doesn't need to specify at this time what the new choice will be.) This place becomes the new ending point of the proposed sequence.
Example: The active player draws a card during her draw step, then says, "Go." The nonactive player is holding Into the Fray (an instant that says "Target creature attacks this turn if able") and says, "I'd like to cast a spell during your beginning of combat step." The current proposed shortcut is that all players pass priority at all opportunities during the turn until the nonactive player has priority during the beginning of combat step.
716.2c Once the last player has either accepted or shortened the shortcut proposal, the shortcut is taken. The game advances to the last proposed ending point, with all game choices contained in the shortcut proposal having been taken. If the shortcut was shortened from the original proposal, the player who now has priority must make a different game choice than what was originally proposed for that player.
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.
Example: In a two-player game, the active player controls a creature with the ability "{0}: [This creature] gains flying," the nonactive player controls a permanent with the ability "{0}: Target creature loses flying," and nothing in the game cares how many times an ability has been activated. Say the active player activates his creature's ability, it resolves, then the nonactive player activates her permanent's ability targeting that creature, and it resolves. This returns the game to a game state it was at before. The active player must make a different game choice (in other words, anything other than activating that creature's ability again). The creature doesn't have flying. Note that the nonactive player could have prevented the fragmented loop simply by not activating her permanent's ability, in which case the creature would have had flying. The nonactive player always has the final choice and is therefore able to determine whether the creature has flying.
716.4. If a loop contains only mandatory actions, the game is a draw. (See rules 104.4b and 104.4f.)
716.5. No player can be forced to perform an action that would end a loop other than actions called for by objects involved in the loop.
Example: A player controls Seal of Cleansing, an enchantment that reads, "Sacrifice Seal of Cleansing: Destroy target artifact or enchantment." A mandatory loop that involves an artifact begins. The player is not forced to sacrifice Seal of Cleansing to destroy the artifact and end the loop.
716.6. If a loop contains an effect that says "[A] unless [B]," where [A] and [B] are each actions, no player can be forced to perform [B] to break the loop. If no player chooses to perform [B], the loop will continue as though [A] were mandatory.
The result of my initial conversation with Psyl4mne and the L2 was basically, "Oh dear, the CR seems to be horribly broken." This is because 716.3 is worded very, very poorly in at least three ways. This is kind of interesting, but not the point of this post; if you're curious about any of this I am happy to talk in PM.
The result of my conversation with mostly Brain in #mtgjudge was that we need to apply common sense and make a best guess as to what was intended by the messily-worded 716.3. In particular, e suggested that the rule was intended to read "non-mandatory independent action", and that in cases where the active player appears to be undefined, it means the player who had priority when the gamestate first repeated.
I find these suggestions entirely reasonable, despite them distinctly not being supported by the technical language of the CR. The technical language of the CR doesn't behave anything like how anyone thinks real Magic does either, so.
Adopting these interpretations leads to the following outcomes for the cases we've discussed:
Mandatory/Mandatory => draw as always
Optional/Mandatory => this is not a fragmented loop, and thus the CR does not care that the Optional player wants to continue it forever. E does so, and the game is a draw per XCB rule 1.11e.
Optional/Optional => this IS a fragmented loop, and the player who had priority when the gamestate first repeated will have to do something different.
Note that Mogg's Spark Troops vs. Gizzled Leotau example does NOT fall into the Optional/Optional case. This loop contains only mandatory actions - in fact, it contains no actions at all, since passing priority is not an action. The way Mogg described it, the Shock Troops player is attacking, but there's no particular reason for them to. The essence of the loop is just them sitting there passing the turn and not choosing to break the loop. Per 716.5 they do not have to, so the game is, somewhat counterintuitively, a draw under 716.4.
The rules are pretty unclear about whether situations like bateleur's Isochron Scepter/Time Stop example are loops, because in that situation the Time Stop player can cast it at random different points in either player's turn and it is not clear whether that is "a set of actions repeating indefinitely" (716.1b). A strict reading might lead you to thinking that "repeating" implies "in the same order", but I think common sense probably demands that we assume they meant to say that if the gamestate ends up in an identical sate, it's still a loop if very slightly different actions were taken to get it there. That would make this situation Optional/Optional, and whoever had priority when the gamestate first repeated would have to do something different. If that was the Time Stop player, I believe that they will have to choose not to cast Time Stop and take damage from Platinum Angel. If it was the Angel player, I believe they would have to choose to allow Research to create a token.
This is a change from how XCB has been in the past, but I don't think it's a necessarily bad one. However, there are some really unfortunate results:
Time Vault/Key is Optional/Mandatory.
Orochi Leafcaller is Optional/Mandatory.
The CR sees them as the same, and calls them both infinitely long games, which XCB turns into a draw.
The IPG sees them both as slow play, and gives a game loss to the Optional player.
We want them to be different, but the CR and IPG both think they are the same. The only avenue for argument against this that I see involves taking 716.3 at its word, that is, ignoring the common-sense conclusions that Brain and I came up with, which creates basically utter chaos.
So, if we want XCB to stay roughly how it's been, I think we need to redefine Optional/Mandatory loops that involve the Mandatory player performing actions to be fragmented, AND modify 716.3 such that fragmented loops result in a draw.
What the exact definition of "performing actions" is now becomes an issue. And, I'm afraid, a pretty nasty one:
If we are too loose in our definition, then we risk infinite mana + infinite untaps + Echo Chamber being a draw in the way that we don't want Leafcaller to be. Therefore, it seems like simply making a choice called for by an effect should not count as "performing an action".
However, we can construct a scenario in which that is all that a player does that still meets our intuitions about matches that should be draws:
AP will always attack NAP to 0. Then if NAP ever fails to choose "Creature" for Matrix, e will untap Seat of the Synod, triggering Mesmeric Orb and putting a non-Platinum Angel card on top of eir graveyard and losing from being at 0.
To me, this case is clearly in the "we would have ruled this to be a draw before anyone thought about this issue" category. The only thing that seems to distinguish it from the Echo Chamber example is that turns pass. Therefore, to keep XCB functioning identically to how it has been, I propose the following rule 1.7, bumping current XCR rules 1.7-1.11 up to 1.8-1.12:
[b]1.7.[/b] Fragmented loops that span multiple turns result in a draw. (See CR 716.3)
* 1.7a. An "action" is anything a player does other than passing priority or carrying out the instructions of an effect e does not control. Example: Putting a mandatory triggered ability on the stack is an action. Losing two life when your opponent targets you with Brush With Death is not.
* 1.7b. A "choice" is a decision made by a player during the resolution or casting of an effect. Example: Choosing "Artifact" for your opponent's Storage Matrix is a choice, as is choosing a target for an opponent's Evangelize. Choosing which creatures to attack with is not.
* 1.7c. Loops in which exactly one player takes only mandatory actions are fragmented.
* 1.7d. Loops in which exactly one player takes no actions, but does make choices, are fragmented if and only if the loop spans multiple turns. (This rule supercedes 1.7c)
* Updated example to use Collapsing Borders instead of Honden, so that NAP doesn't perform any actions in the loop.
* Updated 1.7d to specify that it supercedes 17.c (which it already did, but it was confusing).
* Updated 1.7b and 1.7a to include picking targets for Evangelize and Echo Chamber as choices.
Good stuff. I generally like your approach to resolving this, although I'm not sure the wording is quite right (if only because it's extremely technical and I would prefer something that's watertight against us having missed some subtlety in the CR).
Time Vault/Key is Optional/Mandatory.
Orochi Leafcaller is Optional/Mandatory.
The CR sees them as the same, and calls them both infinitely long games, which XCB turns into a draw.
If I'm understanding correctly, that contradicts Eli's answer I posted here. Seeing as Eli's a L3 judge I'm inclined to be a little cautious of anyone contradicting his ruling.
Private Mod Note
():
Rollback Post to RevisionRollBack
--
(I'm on on this site much anymore. If you want to get in touch it's probably best to email me: dom@heffalumps.org)
Forum Awards: Best Writer 2005, Best Limited Strategist 2005-2012
5CB PotM - June 2005, November 2005, February 2006, April 2008, May 2008, Feb 2009
MTGSalvation Articles: 1-20, plus guest appearance on MTGCast #86!
<Limited Clan>
I don't think I'm contradicting him exactly - I think he's just slightly conflating the CR with the MTR/IPG. It's certainly the case that games that would go on indefinitely do not do so in tournaments, and that you are required (by the tournament rules) to propose a shortcut or receive a slow play infraction. This is also a part of the tournament rules that it makes a lot of sense for most casual players to adopt as well, since even they tend not to have literally infinite time on their hands So I think Eli was just answering "what actually happens", rather than what is specified by the CR alone.
I still think that the best ruling (independent of the rules) is that being able to stall for an indefinite number of *turns* is a draw (which covers things like having a Thrill-Kill Assassin vs. Nether Spirit - that is, if you stop attacking you die, if they stop blocking they die), but being able to stall within a single turn with short-cuttable actions is not (barring an actual mandatory actions based loop, obviously).
This, I feel, matches the XCB definition of optimal play ("take as many turns as possible") and how it's worked to date.
Given that by RAW, this is essentially something that comes down to tournament rules, not the CR, and I personally feel that "if your deck can't cleanly kill them, it's not as good as a deck that CAN, and your score should reflect this".
The "we both sit there" option of Shock/Leotau is a stable no non-mandatory actions state, but we've had things very different from that: a player playing a creature every turn only to get it Karakas'd, for example; with the Karakas player either being obliged to take a hit, or the casting player obliged to stop casting it, you get a result instead of a draw (assuming this is something like Leyline/Karakas/Cenn's Tactician vs. a creature that they can't afford to allow to stay in play).
This CR wrangling is... nice, I'm sure, but by the CR people would deck out every XCB game. Our priority should be making a good set of XCB rules, not suddenly switching to slavishly following the CR for everything we haven't explicitly spelled out as a departure.
I'm a little confused, but I *think* that you might not have read the end of my post correctly.
I am fairly sure that my proposal (which is a significant departure from the CR both as written and as practiced) gives you the exact outcomes you want, namely, no change from how XCB has always worked (but the ability to disambiguate situations that we might not have ever thought about before).
"Play a creature every turn only to get it Karakas'd" is Optional/Optional and thus fragmented and thus a draw under my proposal's rule 1.7.
I don't think I'm contradicting [Eli] exactly - I think he's just slightly conflating the CR with the MTR/IPG. It's certainly the case that games that would go on indefinitely do not do so in tournaments, and that you are required (by the tournament rules) to propose a shortcut or receive a slow play infraction. This is also a part of the tournament rules that it makes a lot of sense for most casual players to adopt as well, since even they tend not to have literally infinite time on their hands So I think Eli was just answering "what actually happens", rather than what is specified by the CR alone.
I'm on this same page, which is exactly why I quoted the Slow Play rules from the IPG up at the top of this thread. My reading of the CRs is that nothing in them stops you from using Orochi Leafcaller over and over until the sun explodes. The thing that stops you in tournament play is in the bit in the IPG about how the shortcut rules are not optional in a tournament. The thing that stops you in casual games is societal pressure, probably.
I'm pretty convinced that Eli's answer is correct only in the case of games that are operating under the MTRs in addition to the CRs.
I'm a little confused, but I *think* that you might not have read the end of my post correctly.
I am fairly sure that my proposal (which is a significant departure from the CR both as written and as practiced) gives you the exact outcomes you want, namely, no change from how XCB has always worked (but the ability to disambiguate situations that we might not have ever thought about before).
"Play a creature every turn only to get it Karakas'd" is Optional/Optional and thus fragmented and thus a draw under my proposal's rule 1.7.
You posted a wall of text. I still don't know what it says, and am getting a headache from trying to work it out.
I second what WhammeWhamme said. I don't want to need to know all that terminology and technical jargon to know how to resolve my time walk/anurid scavenger deck. Really it seems like this is a bunch of hoopla over something that had never ever actually been a problem. We follow the optimum actions guideline, and things make intuitive sense. The ABT comp rules seem to have ballooned some time when I wasn't looking, and if I were a newcomer, I'd be rather out off trying to parse what is and is not allowed.
These are not rules that anyone but a few moderators need to understand fully. However, they absolutely DO have to exist, or else rounds will just come down to inconsistent moderator judgment calls.
Assuming my proposal or something like it is adopted, essentially nothing will change except that the results we already expect will have some justification.
We're having this conversation because the rules as they existed previously were not enough to cover situations that actually came up, and it significantly derailed at least two weeks of game threads. Much better to settle it here, thoroughly, so that future games can flow more smoothly.
I've been somewhat loath to add to the list of proposed rules changes floating around purely because there are a number of reasonable options already floating about depending on how we want certain situations to go. However, I think if we agree that this problem is solved in paper magic by the slow-play rules from the IPG, perhaps we can solve our problem with a version of that rule rather than a modification of the CRs. If I take my inspiration from:
Quote from IPG »
It is also slow play if a player continues to execute a loop without being able to provide an exact number of iterations and the expected resulting game state.
I might just make this one addition (my change is in bold)
1.11. Each player plays one match against each other player.
1.11a. Each match has two games.
1.11b. Each player is the starting player for one game in each match.
1.11c. Games are played with perfect information; players know the identities of face-down cards and cards in hidden zones, and players know which decisions have been made by other players.
1.11d. Games are played optimally. The best outcome for a player is to win the game and the worst outcome is to lose the game.
1.11e. If a player's optimal course of action is to execute a loop indefinitely without passing priority on an empty stack, that player must instead choose a finite number of iterations and then allow the game to move into the next phase.
1.11f. If a game would continue indefinitely, then the game is a draw.
I think this particular tweak lines up with most of my preferences (vault/key is still a draw, Orochi Leafcaller is not a draw, and all my Drain vs. Gain matchups are draws), but there are other versions of the same tweak that can have other results. I think even Bateleur's preferences for vault/key to not draw can also be captured with a version of this rule, although a clean version of that wording escapes me right now.
To be clear - I don't really have a horse in this race. My feelings about what should be a draw are subjective and I've only just rejoined this community a few weeks ago. I'd much rather have a clear and concise rule than a rule that leads to exactly the set of results that feel 'right' to me.
I'd be happy with Vault/Key being a draw if it means we get a clean, comprehensible rule.
Private Mod Note
():
Rollback Post to RevisionRollBack
--
(I'm on on this site much anymore. If you want to get in touch it's probably best to email me: dom@heffalumps.org)
Forum Awards: Best Writer 2005, Best Limited Strategist 2005-2012
5CB PotM - June 2005, November 2005, February 2006, April 2008, May 2008, Feb 2009
MTGSalvation Articles: 1-20, plus guest appearance on MTGCast #86!
<Limited Clan>
I still think that the best ruling (independent of the rules) is that being able to stall for an indefinite number of *turns* is a draw (which covers things like having a Thrill-Kill Assassin vs. Nether Spirit - that is, if you stop attacking you die, if they stop blocking they die), but being able to stall within a single turn with short-cuttable actions is not (barring an actual mandatory actions based loop, obviously).
The results of the poll suggest that most people feel this way. What do people think of the following rule?
"If a loop containing at least one optional action would be repeated indefinitely during a turn, instead each player proposes a number of times for that loop to repeat and the loop is repeated for the greatest number proposed. No player is required to make a choice that would end a loop that crosses multiple turns."
Rule 1.11, with the proposed rule:
1.11. Each player plays one match against each other player.
1.11a. Each match has two games.
1.11b. Each player is the starting player for one game in each match.
1.11c. Games are played with perfect information; players know the identities of face-down cards and cards in hidden zones, and players know which decisions have been made by other players.
1.11d. Games are played optimally. The best outcome for a player is to win the game and the worst outcome is to lose the game.
1.11e. If a loop containing at least one optional action would be repeated indefinitely during a turn, instead each player proposes a number of times for that loop to repeat and the loop is repeated for the greatest number proposed. No player is required to make a choice that would end a loop that crosses multiple turns.
1.11f. If a game would continue indefinitely, then the game is a draw.
I don't think this quite does it, because it leaves open the possibility of a loop that BOTH players want to continue (because either one breaking it will be a loss) that occurs entirely within one turn.
This is rarer than the cross-turn case, but it can happen. I think the rule you want is that if *only one player* has an optional action in a loop that doesn't span turns, they have to pick a number.
I don't think this quite does it, because it leaves open the possibility of a loop that BOTH players want to continue (because either one breaking it will be a loss) that occurs entirely within one turn.
I think the rule you want is that if *only one player* has an optional action in a loop that doesn't span turns, they have to pick a number.
I think that rule could be interpreted as meaning that the rule didn't apply to any case in which an opponent had any options beyond passing priority.
Updated proposed wording:
"1.11e. If a loop containing at least one optional action would be repeated indefinitely during a turn, then any player may propose a number of times for that loop to repeat instead. If a player does, then each other player may propose a different number and the loop is repeated for the greatest number of times proposed instead. No player is required to make a choice that would end a loop that crosses multiple turns."
While we're working on the rules, can I offer a suggestion that all of rule 1.11 be moved and renumbered to be somewhere near the top of 2.x (the tournament rules section)? 1.11 basically lays out the structure of a match in XCB, which is something that paper Magic does in the Tournament Rules rather than the Comprehensive Rules.
This isn't really that important and I could even accept the argument that 1.11 is important for new players to see, understand, and even focus on. But my inner Melvin screams that 1.11 doesn't belong in 1.x at all, so I thought I'd at least share the thought.
Although it's no longer relevant to the wording of the XCB rule, I'd like to share the response that I received today from the CI team about the question I sent them:
Also, I've sent the following question to the Cranial Insertion team, to get further clarity around how some of these sequences are handled by the regular rules of Magic:
"I'm trying to get a better understanding of the rules regarding shortcuts, and here's the scenario that's giving me trouble:
Player A controls Recycle and Shock Troopers. Player B controls Recycle and Grizzled Leotau. Neither player has any cards in his or her hand or graveyard, and neither player controls any other permanents. Both players are at 3 life. Is Player A required to sacrifice Shock Troopers (Player B wins), is Player B required to attack with Grizzled Leotau (Player A wins), or is neither player required to do anything (the game is a draw)?
Basically, I see a loop consisting of Player A attacking and Player B blocking. The loop would be broken if Player A sacrificed Shock Troopers or if Player B attacked with Horned Turtle. It also would be broken if Player A chose not to attack with Shock Troopers, but that would create a new loop consisting of neither player attacking. I don't think that attacking with Shock Troopers is a "game event" as referred to in rule 716.2a. That leaves 716.3 to cover this situation. However, each player is the active player at some point during the loop, and both players are involved in the loop, so it's not clear to me which player - if any - is required to make a different choice.
As a follow-up question, rule 716.2a refers to "the outcome of a game event", but I don't see anywhere that this phrase is defined. I assume this refers only to cases where unknowns are involved, such as what will happen when a player resolves Ponder. Is this interpretation correct?"
Answer:
Hi Robert, thanks for writing in.
In the first example, the loop is being held up by actions that the
players perform. Since those actions aren't mandatory, the game won't
automatically be a draw due to that loop. This means that the loop
must be broken by some player taking a different action than the one
that keeps the loop going. If no player wants to do this, then the
game forces the active player to do this. Since the loop spans
multiple turns, the active player is whoever was the active player
when the loop began.
If the loop is broken by the players choosing not to attack, the
resulting loop doesn't require any actions to keep going, so it would
end the game in the draw. This is a special case of a loop that
consists of only mandatory actions. Players can choose to break such a
loop, but they don't have to.
As far as the phrase "the outcome of a game event" is concerned, it's
simply an English phrase that means what it says.
--
Carsten Haese
If I remember correctly, this is basically what Personman discovered, too. I'm still not entirely happy with the lack of definition on outcome. Having more mana after tapping Elvish Mystic certainly feels like an outcome of the event that involves tapping the Mystic, but that's clearly not what's meant by the rule. So it goes.
Not really, that's the 4-1 rule. The debate here is about turning draws into wins; to go with the example that began this all, I feel that a deck that beats a Time Vault/Key deck 6-0 through Pithing Needle or other method that allows them to actually beat them to death should get a higher score than one that just has a counterspell for their win-con.
If it's hard to get a draw off someone (because any kind of repetitive game state instead becomes a loss), matchups become more 3-3 and 6-0... and 6-0's are easier to get.
I don't think that's a good thing.
It wouldn't be, but I think you've missed my point. If decks have a fallback position in which they force draws when they can't win then what you get is lots of 2-2s, not lots of 6-0s (and 3-3s).
4-1 has never been a common scoreline in XCB and probably never will be.
(I'm on on this site much anymore. If you want to get in touch it's probably best to email me: dom@heffalumps.org)
Forum Awards: Best Writer 2005, Best Limited Strategist 2005-2012
5CB PotM - June 2005, November 2005, February 2006, April 2008, May 2008, Feb 2009
MTGSalvation Articles: 1-20, plus guest appearance on MTGCast #86!
<Limited Clan>
I suggest an approval voting system for said poll.
4th place at CCC&G Pro Tour
Chances of bad hands (<2 or >4 land):
21: 28.9%
22: 27.5%
23: 26.3%
24: 25.5%
25: 25.1%
26: 25.3%
Unfortunately it does not line up well with what we want for XCB, about which more later.
This is going to reference section 716 of the CR heavily, so I'm just going to spoil the whole thing right here:
716.1. When playing a game, players typically make use of mutually understood shortcuts rather than explicitly identifying each game choice (either taking an action or passing priority) a player makes.
716.1a The rules for taking shortcuts are largely unformalized. As long as each player in the game understands the intent of each other player, any shortcut system they use is acceptable.
716.1b Occasionally the game gets into a state in which a set of actions could be repeated indefinitely (thus creating a "loop"). In that case, the shortcut rules can be used to determine how many times those actions are repeated without having to actually perform them, and how the loop is broken.
716.2. Taking a shortcut follows the following procedure.
716.2a At any point in the game, the player with priority may suggest a shortcut by describing a sequence of game choices, for all players, that may be legally taken based on the current game state and the predictable results of the sequence of choices. This sequence may be a non-repetitive series of choices, a loop that repeats a specified number of times, multiple loops, or nested loops, and may even cross multiple turns. It can't include conditional actions, where the outcome of a game event determines the next action a player takes. The ending point of this sequence must be a place where a player has priority, though it need not be the player proposing the shortcut.
Example: A player controls a creature enchanted by Presence of Gond, which grants the creature the ability "{T}: Put a 1/1 green Elf Warrior creature token onto the battlefield," and another player controls Intruder Alarm, which reads, in part, "Whenever a creature enters the battlefield, untap all creatures." When the player has priority, he may suggest "I'll create a million tokens," indicating the sequence of activating the creature's ability, all players passing priority, letting the creature's ability resolve and put a token onto the battlefield (which causes Intruder Alarm's ability to trigger), Intruder Alarm's controller putting that triggered ability on the stack, all players passing priority, Intruder Alarm's triggered ability resolving, all players passing priority until the player proposing the shortcut has priority, and repeating that sequence 999,999 more times, ending just after the last token-creating ability resolves.
716.2b Each other player, in turn order starting after the player who suggested the shortcut, may either accept the proposed sequence, or shorten it by naming a place where he or she will make a game choice that's different than what's been proposed. (The player doesn't need to specify at this time what the new choice will be.) This place becomes the new ending point of the proposed sequence.
Example: The active player draws a card during her draw step, then says, "Go." The nonactive player is holding Into the Fray (an instant that says "Target creature attacks this turn if able") and says, "I'd like to cast a spell during your beginning of combat step." The current proposed shortcut is that all players pass priority at all opportunities during the turn until the nonactive player has priority during the beginning of combat step.
716.2c Once the last player has either accepted or shortened the shortcut proposal, the shortcut is taken. The game advances to the last proposed ending point, with all game choices contained in the shortcut proposal having been taken. If the shortcut was shortened from the original proposal, the player who now has priority must make a different game choice than what was originally proposed for that player.
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.
Example: In a two-player game, the active player controls a creature with the ability "{0}: [This creature] gains flying," the nonactive player controls a permanent with the ability "{0}: Target creature loses flying," and nothing in the game cares how many times an ability has been activated. Say the active player activates his creature's ability, it resolves, then the nonactive player activates her permanent's ability targeting that creature, and it resolves. This returns the game to a game state it was at before. The active player must make a different game choice (in other words, anything other than activating that creature's ability again). The creature doesn't have flying. Note that the nonactive player could have prevented the fragmented loop simply by not activating her permanent's ability, in which case the creature would have had flying. The nonactive player always has the final choice and is therefore able to determine whether the creature has flying.
716.4. If a loop contains only mandatory actions, the game is a draw. (See rules 104.4b and 104.4f.)
716.5. No player can be forced to perform an action that would end a loop other than actions called for by objects involved in the loop.
Example: A player controls Seal of Cleansing, an enchantment that reads, "Sacrifice Seal of Cleansing: Destroy target artifact or enchantment." A mandatory loop that involves an artifact begins. The player is not forced to sacrifice Seal of Cleansing to destroy the artifact and end the loop.
716.6. If a loop contains an effect that says "[A] unless [B]," where [A] and [B] are each actions, no player can be forced to perform [B] to break the loop. If no player chooses to perform [B], the loop will continue as though [A] were mandatory.
The result of my initial conversation with Psyl4mne and the L2 was basically, "Oh dear, the CR seems to be horribly broken." This is because 716.3 is worded very, very poorly in at least three ways. This is kind of interesting, but not the point of this post; if you're curious about any of this I am happy to talk in PM.
The result of my conversation with mostly Brain in #mtgjudge was that we need to apply common sense and make a best guess as to what was intended by the messily-worded 716.3. In particular, e suggested that the rule was intended to read "non-mandatory independent action", and that in cases where the active player appears to be undefined, it means the player who had priority when the gamestate first repeated.
I find these suggestions entirely reasonable, despite them distinctly not being supported by the technical language of the CR. The technical language of the CR doesn't behave anything like how anyone thinks real Magic does either, so.
Adopting these interpretations leads to the following outcomes for the cases we've discussed:
Mandatory/Mandatory => draw as always
Optional/Mandatory => this is not a fragmented loop, and thus the CR does not care that the Optional player wants to continue it forever. E does so, and the game is a draw per XCB rule 1.11e.
Optional/Optional => this IS a fragmented loop, and the player who had priority when the gamestate first repeated will have to do something different.
Note that Mogg's Spark Troops vs. Gizzled Leotau example does NOT fall into the Optional/Optional case. This loop contains only mandatory actions - in fact, it contains no actions at all, since passing priority is not an action. The way Mogg described it, the Shock Troops player is attacking, but there's no particular reason for them to. The essence of the loop is just them sitting there passing the turn and not choosing to break the loop. Per 716.5 they do not have to, so the game is, somewhat counterintuitively, a draw under 716.4.
The rules are pretty unclear about whether situations like bateleur's Isochron Scepter/Time Stop example are loops, because in that situation the Time Stop player can cast it at random different points in either player's turn and it is not clear whether that is "a set of actions repeating indefinitely" (716.1b). A strict reading might lead you to thinking that "repeating" implies "in the same order", but I think common sense probably demands that we assume they meant to say that if the gamestate ends up in an identical sate, it's still a loop if very slightly different actions were taken to get it there. That would make this situation Optional/Optional, and whoever had priority when the gamestate first repeated would have to do something different. If that was the Time Stop player, I believe that they will have to choose not to cast Time Stop and take damage from Platinum Angel. If it was the Angel player, I believe they would have to choose to allow Research to create a token.
This is a change from how XCB has been in the past, but I don't think it's a necessarily bad one. However, there are some really unfortunate results:
Time Vault/Key is Optional/Mandatory.
Orochi Leafcaller is Optional/Mandatory.
The CR sees them as the same, and calls them both infinitely long games, which XCB turns into a draw.
The IPG sees them both as slow play, and gives a game loss to the Optional player.
We want them to be different, but the CR and IPG both think they are the same. The only avenue for argument against this that I see involves taking 716.3 at its word, that is, ignoring the common-sense conclusions that Brain and I came up with, which creates basically utter chaos.
So, if we want XCB to stay roughly how it's been, I think we need to redefine Optional/Mandatory loops that involve the Mandatory player performing actions to be fragmented, AND modify 716.3 such that fragmented loops result in a draw.
What the exact definition of "performing actions" is now becomes an issue. And, I'm afraid, a pretty nasty one:
If we are too loose in our definition, then we risk infinite mana + infinite untaps + Echo Chamber being a draw in the way that we don't want Leafcaller to be. Therefore, it seems like simply making a choice called for by an effect should not count as "performing an action".
However, we can construct a scenario in which that is all that a player does that still meets our intuitions about matches that should be draws:
AP controls Storage Matrix, Recycle, Mesmeric Orb, Raging Bull, Collapsing Borders, and one of each basic land. NAP is at 2 life and controls Recycle, an Arrested Volrath's Shapeshifter copying the Platinum Angel on top of eir graveyard, a tapped Seat of Synod, a Forest, a Mountain, and a Plains.
AP will always attack NAP to 0. Then if NAP ever fails to choose "Creature" for Matrix, e will untap Seat of the Synod, triggering Mesmeric Orb and putting a non-Platinum Angel card on top of eir graveyard and losing from being at 0.
To me, this case is clearly in the "we would have ruled this to be a draw before anyone thought about this issue" category. The only thing that seems to distinguish it from the Echo Chamber example is that turns pass. Therefore, to keep XCB functioning identically to how it has been, I propose the following rule 1.7, bumping current XCR rules 1.7-1.11 up to 1.8-1.12:
[b]1.7.[/b] Fragmented loops that span multiple turns result in a draw. (See CR 716.3)
* 1.7a. An "action" is anything a player does other than passing priority or carrying out the instructions of an effect e does not control. Example: Putting a mandatory triggered ability on the stack is an action. Losing two life when your opponent targets you with Brush With Death is not.
* 1.7b. A "choice" is a decision made by a player during the resolution or casting of an effect. Example: Choosing "Artifact" for your opponent's Storage Matrix is a choice, as is choosing a target for an opponent's Evangelize. Choosing which creatures to attack with is not.
* 1.7c. Loops in which exactly one player takes only mandatory actions are fragmented.
* 1.7d. Loops in which exactly one player takes no actions, but does make choices, are fragmented if and only if the loop spans multiple turns. (This rule supercedes 1.7c)
* Updated 1.7d to specify that it supercedes 17.c (which it already did, but it was confusing).
* Updated 1.7b and 1.7a to include picking targets for Evangelize and Echo Chamber as choices.
If I'm understanding correctly, that contradicts Eli's answer I posted here. Seeing as Eli's a L3 judge I'm inclined to be a little cautious of anyone contradicting his ruling.
(I'm on on this site much anymore. If you want to get in touch it's probably best to email me: dom@heffalumps.org)
Forum Awards: Best Writer 2005, Best Limited Strategist 2005-2012
5CB PotM - June 2005, November 2005, February 2006, April 2008, May 2008, Feb 2009
MTGSalvation Articles: 1-20, plus guest appearance on MTGCast #86!
<Limited Clan>
I don't think I'm contradicting him exactly - I think he's just slightly conflating the CR with the MTR/IPG. It's certainly the case that games that would go on indefinitely do not do so in tournaments, and that you are required (by the tournament rules) to propose a shortcut or receive a slow play infraction. This is also a part of the tournament rules that it makes a lot of sense for most casual players to adopt as well, since even they tend not to have literally infinite time on their hands So I think Eli was just answering "what actually happens", rather than what is specified by the CR alone.
I still think that the best ruling (independent of the rules) is that being able to stall for an indefinite number of *turns* is a draw (which covers things like having a Thrill-Kill Assassin vs. Nether Spirit - that is, if you stop attacking you die, if they stop blocking they die), but being able to stall within a single turn with short-cuttable actions is not (barring an actual mandatory actions based loop, obviously).
This, I feel, matches the XCB definition of optimal play ("take as many turns as possible") and how it's worked to date.
Given that by RAW, this is essentially something that comes down to tournament rules, not the CR, and I personally feel that "if your deck can't cleanly kill them, it's not as good as a deck that CAN, and your score should reflect this".
The "we both sit there" option of Shock/Leotau is a stable no non-mandatory actions state, but we've had things very different from that: a player playing a creature every turn only to get it Karakas'd, for example; with the Karakas player either being obliged to take a hit, or the casting player obliged to stop casting it, you get a result instead of a draw (assuming this is something like Leyline/Karakas/Cenn's Tactician vs. a creature that they can't afford to allow to stay in play).
This CR wrangling is... nice, I'm sure, but by the CR people would deck out every XCB game. Our priority should be making a good set of XCB rules, not suddenly switching to slavishly following the CR for everything we haven't explicitly spelled out as a departure.
I am fairly sure that my proposal (which is a significant departure from the CR both as written and as practiced) gives you the exact outcomes you want, namely, no change from how XCB has always worked (but the ability to disambiguate situations that we might not have ever thought about before).
"Play a creature every turn only to get it Karakas'd" is Optional/Optional and thus fragmented and thus a draw under my proposal's rule 1.7.
I'm on this same page, which is exactly why I quoted the Slow Play rules from the IPG up at the top of this thread. My reading of the CRs is that nothing in them stops you from using Orochi Leafcaller over and over until the sun explodes. The thing that stops you in tournament play is in the bit in the IPG about how the shortcut rules are not optional in a tournament. The thing that stops you in casual games is societal pressure, probably.
I'm pretty convinced that Eli's answer is correct only in the case of games that are operating under the MTRs in addition to the CRs.
You posted a wall of text. I still don't know what it says, and am getting a headache from trying to work it out.
Assuming my proposal or something like it is adopted, essentially nothing will change except that the results we already expect will have some justification.
We're having this conversation because the rules as they existed previously were not enough to cover situations that actually came up, and it significantly derailed at least two weeks of game threads. Much better to settle it here, thoroughly, so that future games can flow more smoothly.
However, something needs to be amended so that Land Rule formats can still end in a draw.
4th place at CCC&G Pro Tour
Chances of bad hands (<2 or >4 land):
21: 28.9%
22: 27.5%
23: 26.3%
24: 25.5%
25: 25.1%
26: 25.3%
I might just make this one addition (my change is in bold)
To be clear - I don't really have a horse in this race. My feelings about what should be a draw are subjective and I've only just rejoined this community a few weeks ago. I'd much rather have a clear and concise rule than a rule that leads to exactly the set of results that feel 'right' to me.
(I'm on on this site much anymore. If you want to get in touch it's probably best to email me: dom@heffalumps.org)
Forum Awards: Best Writer 2005, Best Limited Strategist 2005-2012
5CB PotM - June 2005, November 2005, February 2006, April 2008, May 2008, Feb 2009
MTGSalvation Articles: 1-20, plus guest appearance on MTGCast #86!
<Limited Clan>
The results of the poll suggest that most people feel this way. What do people think of the following rule?
"If a loop containing at least one optional action would be repeated indefinitely during a turn, instead each player proposes a number of times for that loop to repeat and the loop is repeated for the greatest number proposed. No player is required to make a choice that would end a loop that crosses multiple turns."
Rule 1.11, with the proposed rule:
1.11. Each player plays one match against each other player.
BWTeysa, Orzhov Scion
GWRhys the Redeemed
GUKruphix, God of Horizons
GRXenagos, God of Revels
GThrun, the Last Troll
GStompy
This is rarer than the cross-turn case, but it can happen. I think the rule you want is that if *only one player* has an optional action in a loop that doesn't span turns, they have to pick a number.
Good point.
I think that rule could be interpreted as meaning that the rule didn't apply to any case in which an opponent had any options beyond passing priority.
Updated proposed wording:
"1.11e. If a loop containing at least one optional action would be repeated indefinitely during a turn, then any player may propose a number of times for that loop to repeat instead. If a player does, then each other player may propose a different number and the loop is repeated for the greatest number of times proposed instead. No player is required to make a choice that would end a loop that crosses multiple turns."
BWTeysa, Orzhov Scion
GWRhys the Redeemed
GUKruphix, God of Horizons
GRXenagos, God of Revels
GThrun, the Last Troll
GStompy
This isn't really that important and I could even accept the argument that 1.11 is important for new players to see, understand, and even focus on. But my inner Melvin screams that 1.11 doesn't belong in 1.x at all, so I thought I'd at least share the thought.
Question:
Hi Robert, thanks for writing in.
In the first example, the loop is being held up by actions that the
players perform. Since those actions aren't mandatory, the game won't
automatically be a draw due to that loop. This means that the loop
must be broken by some player taking a different action than the one
that keeps the loop going. If no player wants to do this, then the
game forces the active player to do this. Since the loop spans
multiple turns, the active player is whoever was the active player
when the loop began.
If the loop is broken by the players choosing not to attack, the
resulting loop doesn't require any actions to keep going, so it would
end the game in the draw. This is a special case of a loop that
consists of only mandatory actions. Players can choose to break such a
loop, but they don't have to.
As far as the phrase "the outcome of a game event" is concerned, it's
simply an English phrase that means what it says.
--
Carsten Haese
If I remember correctly, this is basically what Personman discovered, too. I'm still not entirely happy with the lack of definition on outcome. Having more mana after tapping Elvish Mystic certainly feels like an outcome of the event that involves tapping the Mystic, but that's clearly not what's meant by the rule. So it goes.
BWTeysa, Orzhov Scion
GWRhys the Redeemed
GUKruphix, God of Horizons
GRXenagos, God of Revels
GThrun, the Last Troll
GStompy