I'm wondering if anyone -- particularly someone who is a Magic Judge -- would be interested in helping me edit an article I'm writing for no site in particular.

The focus of the article is to discuss "loops," the way that Magic deals with unbounded or infinite situations. While there are some other pieces out there that lightly touch on these issues, there is no good comprehensive resource for someone who wants to understand the precise way that the Magic rules handle these situations.

I see it come up most frequently in discussions about things you CANNOT do, like run the Four Horseman in a competitive tournament. This looks like one of the "third rails" in Magic discussions, and one that makes Judges run for the hills in particular. I've asked them. I sometimes feel like there was some major religious war in the past on the Judge forums and they've all taken an oath to never really explain the loop rules and the reasons behind them on a deep level because, if they do, they believe it will start a big argument.

I don't think just telling people "It does not work. Your question has been answered, Thread locked." is a good response, though. Far better to walk people through the Comprehensive Rules, the Tournament Rules, and the IPG to explain exactly what's going on and why it makes sense. That's what I've tried to do.

Nevertheless, I'm not a Magic Judge, and I'd like input from someone who is in case I'm missing something. I'm pretty sure my analysis of the written rules is correct. What I do not have, though, is access to the judge's forums and a history of the issues.

So, if anyone is interested, I've linked the article in draft form here:
https://www.dropbox.com/s/xrhayfa2q7nw98s/Loop Article DRAFT 7-28-16.pdf?dl=0

If you are a judge or have information relevant to the issues I'm discussing here, let me know. I'd love to make this a single, final (as of 2016 rules, anyway) exegesis on everything you need to know about loop rules.

In addition to other suggestions, or telling me if I'm just wrong, I'm keenly interested in the following:

1. Is my understanding of how MTGO handles loops correct?
2. What rule causes abilities like Grindstone to stop once they cannot do anything -- i.e. when the library is empty?
3. Are there judge forum posts or rulings that are still applicable today and are relevant to the issues I'm discussing? Is there a place I can cite to them?
4. It looks to me like "advancing the game state" as a concept actually has nothing to do with modern loop rules. Is that an older term? Or is it something that comes out of a judge's prior rulings?

Thanks in advance for any help you can offer.

Its a noble undertaking you've undertaken here Max and I applaud efforts like this. But at the same time, efforts like this concern me. There is a reason judges don't give a concrete answer (aside from the inexperienced ones just not knowing) - circumstances. There are certain issues like the mandatory loop with no way to break it that, like you said in the article, are self evident and straight forward. But beyond that - each game state is unique and could possibly affect a game situation.

I'm not certain if you are just trying to present a single comprehensive resource for research purposes (2 thumbs up if that's the case) or trying to present a single comprehensive 'answer' to questions (2 thumbs down if that is the case). Aside from a mandatory loop with no way to break (which as you said in in the article is pretty self evident) each game is unique presents multiple ways a loop may not fit into a single 'end all, be all' answer. If you aren't going for that, maybe its just a tone I added myself while reading.

I'm not a judge but I can help with a few of these questions.

1. Your understanding seems pretty spot on. The article does a very nice job of reconfirming and explaining various loops. Under the mandatory loops I would suggesting using the term 'infinite loop'. It may not be an official term. But its a term that comes up often when discussing loops and is widely used by players to describe the worldgorger example you use. You could then add in that if there is another target, its not truly and infinite loop. I only bring this up because many readers will be using these terms and it may tie your information directly to what a reader may be researching.

2. Rules for grindstone - from your article, "The G
   rindstone’s ability never stops by its own terms, but the deck being milled will eventually run out of cards. When that happens, the effect stops resolving because there’s nothing left in the library to act upon."

It doesn't stop resolving because there is nothing left to act upon. It stops resolving because the condition to make it repeat (2 cards sharing the same color) are not met. The rule that applies is: 608.2a If a triggered ability has an intervening “if” clause, it checks whether the clause’s condition is true. If it isn’t, the ability is removed from the stack and does nothing. Otherwise, it continues to resolve. See rule 603.4

Good point on using the term "infinite" loop. I think I would still use "mandatory," perhaps with some tweaking to link it directly to the prior term, because the category I was trying to grapple with was loops that go on of their own accord, without the player making a choice.

@Khahan said:

Its a noble undertaking you've undertaken here Max and I applaud efforts like this. But at the same time, efforts like this concern me. There is a reason judges don't give a concrete answer (aside from the inexperienced ones just not knowing) - circumstances. There are certain issues like the mandatory loop with no way to break it that, like you said in the article, are self evident and straight forward. But beyond that - each game state is unique and could possibly affect a game situation.

It's certainly my intention just to explain the rules, not to prove a concrete answer. That said, after digging into the subject a bit, it feels like the answer to a particular problem is pretty concrete:

1. Is a loop occurring? If not, we're not in these rules.
2. If so, is the kind of loop that can be subject to a shortcut? If so, do that.
3. If not, then issue a slow play warning.

It seems like any situation would be susceptible to resolution based on this.

Or, perhaps the worry is that as soon as someone thinks they know the precise boundaries for what is not allowed, they try to game the rules by doing something arguably just outside those bounds, but still abusive?

Maybe that deserves a paragraph in the intro just cautioning people that: (1) trying to find a loophole will end in tears; and (2) your local judge is God.

@Khahan said:

It doesn't stop resolving because there is nothing left to act upon. It stops resolving because the condition to make it repeat (2 cards sharing the same color) are not met. The rule that applies is: 608.2a If a triggered ability has an intervening “if” clause, it checks whether the clause’s condition is true. If it isn’t, the ability is removed from the stack and does nothing. Otherwise, it continues to resolve. See rule 603.4

Okay, thanks. That makes sense.

How do the rules handle the ending of the effect of Helm of Obedience and Leyline of the Void, then? The condition for Helm is an "if this, end" rather than "if this, continue." What about that game rules makes Helm stop?

@MaximumCDawg said:

How do the rules handle the ending of the effect of Helm of Obedience and Leyline of the Void, then? The condition for Helm is an "if this, end" rather than "if this, continue." What about that game rules makes Helm stop?

I'm not actually sure what game rule would apply. Here is the ruling from helm of obedience:

"If an effect like that of Leyline of the Void prevents cards from being put into your opponent’s graveyard, the process described in the first sentence of Helm of Obedience’s effect will never stop. Your opponent’s entire library will be exiled, even if X is 1. "

I'm at work right now and its getting a bit busy. If somebody hasn't posted something before I get home I'll try to dig thru the rules and find something specific.

@MaximumCDawg I'm an L2 judge in the Dallas area. If you want to email the article to me I'll take a look (dropbox is blocked at work)
infant.no.1 at g mail

If someone else already has it handled then I look forward to reading it when it is published.

-josh-

@infant_no_1
I linked the dropbox in the opening post, but I'm happy to e-mail it if you'd rather.

I dont know if this fully answers your question about helm/leyline, but here is what I could find that at least seems relevant. All from the comprehensive rules under resolving spells and abilities section:

609
.3. If an effect attempts to do something impossible, it does only as much as possible.
Example:
If a player is holding only one card, an effect that reads “Discard two cards”
causes him or her to discard only that card. If an effect moves cards out of the
library (as
opposed to drawing), it moves as many as possible.

So when helm says to put x cards or a creature into the graveyard it will keep trying to resolve to do that. This is how it mills your library. Once there are no cards to reveal its impossible to reveal anymore so the ability ceases to do anything at all. I could not find anything that specifies, 'when this occurs the spell stops.' Just seems natural to do that. I dont know if something like that does exist in the rules or not. This is just the best I could find.

Yeah, that rule looks like it might be the one that applies. "If an effect attempts to do something impossible, it does only as much as possible" certainly explains why the Helm stops. It tries to mill forever, but it can only mill as many cards in the library as a player has, so it stops.

This raises another interesting area of inquiry, probably worth it's own article: negative values in Magic. Some variables, like life, can be negative. Others, like cards in libraries, cannot be. Some, like power, can be negative but when it's time to actually apply them to an effect, they are treated like they are zero. I bring it up because an affect doing "something impossible" might depend upon what kind of game variable we're talking about. If you need to keep milling libraries down past zero cards, that's impossible and we stop. If you need to keep damaging a player after they're at zero life (Platinum Angel in play, say) then that's fine.

So you can get into a Mandatory Loop causing a draw if you assemble Exquisite Blood + Sanguine Bond against an opponent's Platinum Angel but you don't cause one with Leyline + Helm.

@MaximumCDawg said:

Yeah, that rule looks like it might be the one that applies. "If an effect attempts to do something impossible, it does only as much as possible" certainly explains why the Helm stops. It tries to mill forever, but it can only mill as many cards in the library as a player has, so it stops.

This raises another interesting area of inquiry, probably worth it's own article: negative values in Magic. Some variables, like life, can be negative. Others, like cards in libraries, cannot be. Some, like power, can be negative but when it's time to actually apply them to an effect, they are treated like they are zero. I bring it up because an affect doing "something impossible" might depend upon what kind of game variable we're talking about. If you need to keep milling libraries down past zero cards, that's impossible and we stop. If you need to keep damaging a player after they're at zero life (Platinum Angel in play, say) then that's fine.

So you can get into a Mandatory Loop causing a draw if you assemble Exquisite Blood + Sanguine Bond against an opponent's Platinum Angel but you don't cause one with Leyline + Helm.

Actually I think these situations are pretty straight forward if you think about them. Lets start w/the aforementioned leyline/helm. We know when its impossible to mill any other cards (the opponent has no more library) the action stops and the opponent loses on his next draw. That's one impossibility and/or negative number situation dealt with.

But what about life? If you are at 0 life, you can take damage and go below. But if you think about the damage comes from another source. The sanguine loop you mentioned has a source dealing X amount of damage. That is wholly different than being at 0 life with platy in play and trying to activate necropotence. Basically you can be put into a negative by an outside force. The negative life is a way to track and really nothing more. But you can't put yourself at a negative or pay what you don't have. In fact, this is dealt with in the comprehensive rules:

107.1. The only numbers the Magic game uses are integers.
107.1a You can’t choose a fractional number, deal fractional damage, gain fractional life, and so on. If a spell or ability could generate a fractional number, the spell or ability will tell you whether to round up or down.
107.1b Most of the time, the Magic game uses only positive numbers and zero. You can’t choose a negative number, deal negative damage, gain negative life, and so on. However, it’s possible for a game value, such as a creature’s power, to be less than zero. If a calculation or comparison needs to use a negative value, it does so. If a calculation that would determine the result of an effect yields a negative number, zero is used instead, unless that effect sets a player’s life total to a specific value, doubles a player’s life total, sets a creature’s power or toughness to a specific value, or otherwise modifies a creature’s power or toughness.
Example: If a 3/4 creature gets -5/-0, it’s a -2/4 creature. It doesn’t assign damage in combat. Its total power and toughness is 2. You’d have to give it +3/+0 to raise its power to 1.
Example: Viridian Joiner is a 1/2 creature with the ability “{T}: Add to your mana pool an amount of {G} equal to Viridian Joiner’s power.” An effect gives it -2/-0, then its ability is activated. The ability adds no mana to your mana pool.
107.1c If a rule or ability instructs a player to choose “any number,” that player may choose any positive number or zero, unless something (such as damage or counters) is being divided or distributed among “any number” of players and/or objects. In that case, a nonzero number of players and/or objects must be chosen if possible.
107.2. If anything needs to use a number that can’t be determined, either as a result or in a calculation, it uses 0 instead.

A lot of what's needed to understand these situations is actually hiding in the tournament shortcuts section 718.

Most importantly:

718.3. 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 is important for using RIP + Helm when someone has a Blightsteel in their library. As it is two competing replacement effects (RIP and Blightsteel) when the robot is milled, the controller of Blightsteel chooses which effect to use. However, as this creates this kind of fragmented loop, the player must state how many times he replaces the effect, and then must make a different choice (i.e. letting it get RFG) afterward. Therefore RIP + Helm mills a player, regardless of what's in their library. Note, that this is not the case for Grindstone + 2 Blightsteel as the player is not making a choice there, just applying the effect, so that becomes a mandatory loop.

Another example of this in Modern is if two people have the Melira/Anafenza + persist creature + sac outlet combo. If active player has infinite damage and non-active player has infinite life, the loop of: "I deal 1 million damage" -> "I gain 1 million life in response" requires the active player to make a different choice and therefore non-active player survives. However, if it is the opposite turn, the infinite damage player wins. This is not relevant for MTGO, per se, as there are no automated loops, but it does show the universal nature of this concept as well some of the issues of putting automated loops into MTGO.

Back to mandatory loops, inaction is always permitted:

718.5. 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.

718.6. 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.

So, while a Dragon player must choose a different creature to reanimate if they have no win con, the opponent never needs to swords the dragon to prevent the draw.

Hopefully this helps.

So, I'm getting around to editing this piece, but in the meanwhiles I thought of a situation which may be related to loops, but is slightly different from all the categories set forth so far: what if resolution of a single ability requires doing something that is neither infinite nor, strictly speaking, a loop, and yet cannot be allowed to happen because it would ruin the tournament?

Here's one way you could set this up.

Player 1: I combo off with Kikki-Pestermite; short cut to a million tokens.
Player 2: I will do exactly the same thing, making as many tokens as you. As non-active player, I can always win this race.
Player 1: (Thinks: Well, I'll just outpace him next turn on tokens. Gotta attack, though, to avoid getting outnumbered in a few turns) Fine. I'll attack with my million tokens.
Player 2: I block each of your tokens with one of mine. And... I now cast Fighting Chance

What do the Rules say about this? Shockingly little, I think. It's not a loop because you're not returning to the same game state and making a new decision; you're resolving an effect. You can't just do an informal shortcut because the outcome is indeterminate. So, you gotta do it. Only, if you did it, then the whole tournament would end in a catastrophe. So clearly, that cannot happen. But, what can the judges do? Slow Play penalty does not appear to apply on its face (no loops involved, no unreasonable pace of play involved). I suppose someone could whip up a computer program that rolls X number of dice and counts the even numbers or something, but failing that, what part of the Comp. Rules or IPG would authorize stepping in?

@MaximumCDawg I suppose you win the tournament; everybody else goes home. Chances are, you are already at the kitchen table.

@peterflugzeug Pithy, but that's not my point. I'm trying to actually determine how the ruled play out. Perhaps there is a "tournament must go on" general rule somewhere?

@MaximumCDawg The only context where a resolution to this situation has to be strictly defined is in MTGO, as in real-life tournaments a) an ad-hoc resolution would work equally well, and b) as @peterflugzeug said, this situation would never even come up in competitive play. So the ‘general rule’ you are looking for is: The Head Judge of the tournament has the final say in questions regarding rules and tournament operations.

I did some digging to find actual citations for the principals you guys are saying. No doubt you are right as a practical matter, but my goal is to be a little more precise than that. While I realize that the Head Judge is the final arbiter in each tournament, I don't see a rule that says the Head Judge can deviate from published rules or find a new fix in the case of a tournament logistics problem that is not accounted for in the rules.

I'm not convinced its NOT covered by the Rules, though. Here's what I came up with.

Magic Comp Rules

101.3. Any part of an instruction that’s impossible to perform is ignored. (In many cases the card will specify consequences for this; if it doesn’t, there’s no effect.)

This is interesting. Could a judge rule that an effect whose resolution would require more time than allotted for the tournament to be "impossible" and thus simply ignore it? It doesn't seem unreasonable

Then again, we have this:

609.3. If an effect attempts to do something impossible, it does only as much as possible.

Which might temper being able to use a rule to call resolution of an effect calling for a million coin flips to be "impossible."

Additionally, the Magic Tournament Rules go a bit further than the IPG when it comes to Slow Play, adding in the concept of getting stuff done in the announced time limit:

5.5 Slow Play
Players must take their turns in a timely fashion regardless of the complexity of the play situation and adhere to time limits specified for the tournament. Players must maintain a pace to allow the match to be finished in the announced time limit. Stalling is not acceptable. Players may ask a judge to watch their game for slow play; such a request will be granted if feasible.

Still, even this refers to "pace" of play, not the massive nature of a single ability to resolve.

Finally, what if the player has the ability to actually flip a million dice? Perhaps they made an app for that. Well, no, that's probably not going to work because using something other than a coin requires consent of the opponent:

705.3. A coin used in a flip must be a two-sided object with easily distinguished sides and equal likelihood that either side lands face up. If the coin that’s being flipped doesn’t have an obvious “heads” or “tails,” designate one side to be “heads,” and the other side to be “tails.” Other methods of randomization may be substituted for flipping a coin as long as there are two possible outcomes of equal likelihood and all players agree to the substitution. For example, the player may roll an even-sided die and call “odds” or “evens,” or roll an even-sided die and designate that “odds”
means “heads” and “evens” means “tails.

So, I wonder if this is an appropriate way to address this unusual situation in my article:

*LOOPS AND DELAYED TROUBLE

A loop that itself does not cause a problem could lead to one down the road. It is possible to construct circumstances in which players are not in an infinite mandatory loop, but due the results of an earlier unbounded loop, arrive at a sequence of actions that cannot be accomplished within the time permitted for a tournament.

For example, perhaps two players have both made a million copies of Pestermite from Kiki Jiki, one attacks the other, the defending player blocks each token with their own, and then the defending player plays Fighting Chance. How many Pestermite tokens survive the exchange? To figure that out, Fighting Chance asks the players to flip a million coins, which cannot be done before the tournament is over.

The problem here is that the initial loop was determinate, and therefore it was proper to execute it as a short cut, but now there is an indeterminate effect that wants to piggy-back on the result of the earlier determinate loop.

Obviously, you are never going to be allowed to start flipping coins until tournament organizer closes the venue. As a practical matter, the Tournament Organizer will use his or her authority to stop the problem. What would the solution be? Since you're neither executing a loop nor playing at a slow pace, the Slow Play rules probably are no help. There is a part of the Comprehensive Rules that might suggest the solution, however. Under Comp. Rule 101.3, any action that is impossible is ignored. Since it is impossible to flip a million coins, then the Tournament Organizer might very well simply announce that Fighting Chance resolves with no effect. *

So, I was doing my revisions and I ran into a situation involving Petals of Insight. Basically, I'm wondering how Petals+Omniscence is compatible with the Rules requirement that the opponent may “either accept the proposed sequence, or shorten it by naming a place where he or she will make a game choice that is different from what is being proposed.” Magic Comp. Rules § 718.2b (2016).

Here's what I have now:

"Whether a loop is Indeterminate is not controlled entirely by the application of the shortcut rules, not simply because there is something random involved. For example, consider a player who casts Petals of Insight while controlling Omniscence. Petals allows them to rearrange their deck, three cards at a time. Can the player announce a shortcut to put their library in a specific order of their choosing?

"Yes, they can. Unlike with the Four Horseman example above, the player casting Petals can say with absolute certainty that, if they cast Petals 410,758 times, that player will have arranged the deck in any way they like. The mathematics proving this is a bit beyond the scope of this article. The number of iterations is finite and fixed, and the outcome is deterministic. This is not theoretical; Judge Gavin Duggan has issued exactly this ruling on the matter.

"Still, it may be possible to construct board states where a loop involving Petals runs afoul of the shortcut Rule’s requirement that the second player be allowed to interrupt it. Suppose the opponent has a Field of Dreams in play, and wishes to stop the loop when Emrakul is on top of the library so he or she can exile it with an instant. Will Emrakul ever be on top before the library has been completely rearranged? That depends on where it happens to be. If Emrakul’s position from the top of the library is a multiple of four, she will end up on top at some point. If not, then the opponent can carefully arrange each group of three cards so that she never is revealed. Since, in this situation, it is impossible to know whether the loop can be interrupted without actually doing it, this would seem to prohibit the shortcut."

@MaximumCDawg said in Editors Wanted - Loop Article:

So, I was doing my revisions and I ran into a situation involving Petals of Insight. Basically, I'm wondering how Petals+Omniscence is compatible with the Rules requirement that the opponent may “either accept the proposed sequence, or shorten it by naming a place where he or she will make a game choice that is different from what is being proposed.” Magic Comp. Rules § 718.2b (2016).

Here's what I have now:

"Whether a loop is Indeterminate is not controlled entirely by the application of the shortcut rules, not simply because there is something random involved. For example, consider a player who casts Petals of Insight while controlling Omniscence. Petals allows them to rearrange their deck, three cards at a time. Can the player announce a shortcut to put their library in a specific order of their choosing?

I don't see this as being incompatible with each other. If I have omniscience in play and petals of insight in my hand I can, as they state, arrange my deck however I want. Lets go with their numbers so I announce I'm going to do this 410,758 times. You as my opponent says, "Ok, on attempt 50 Im going to mindbreak trap your petals." Technically I only get 50 because I lose the ability to rearrange after that and the other 360,758 attempts never hit the stack.

If you say, "Ok, after attempt 50 Im going to activate grindstone and mill you." You get to that 50, I get milled and, after seeing what goes to my yard, can start the process again. If you have an effect that causes me to reshuffle, we go back to the mill event resolution. The game state has changed so I can activate the loop again however many times I want.

Remember, this is all sorcery speed, so you are really doing this 1 spell at a time. You aren't putting 410,000 copies on the stack and having them resolve one at a time. You are taking an action in full - announcing, placing it on the stack, both passing priority, repeat - 410,000 times. That makes a big difference in how this plays out.

I think if petals of insight were an instant you could get into some stickier situations. If petals were an instant your opponent could say, "put them on the stack." Ok, after the last one I mindbreak trap you. So you lose petals and can't repeat. Then he could say, "I let all but the last 50 resolve, then I extirpate you."

Now what does the judge do? You have 50 copies. Does he let you short cut or do you have to take the time to resolve each one?

@MaximumCDawg

"Still, it may be possible to construct board states where a loop involving Petals runs afoul of the shortcut Rule’s requirement that the second player be allowed to interrupt it. Suppose the opponent has a Field of Dreams in play, and wishes to stop the loop when Emrakul is on top of the library so he or she can exile it with an instant. Will Emrakul ever be on top before the library has been completely rearranged? That depends on where it happens to be. If Emrakul’s position from the top of the library is a multiple of four, she will end up on top at some point. If not, then the opponent can carefully arrange each group of three cards so that she never is revealed. Since, in this situation, it is impossible to know whether the loop can be interrupted without actually doing it, this would seem to prohibit the shortcut."

This is a really interesting concern and would probably end up preventing the opponent from shortcutting until they've gone through the deck once. In my mind, this puts a lid on a lot of complaints about loops because it's a precedent that the opponent needs to provide a legitimate concern to a 'mostly deterministic' loop which I think is fair, but not done in the case of something like 4 horsemen. That being said, while the length of the process to order the deck is unknown until one run through is done and the deck order is known, you know that it will always be finite. In the end it becomes a matter of shuffling vs not shuffling being the only difference maker.

However, I do note that they accepted a proof of the loop method having an upper bound in lieu of actually demonstrating a loop, which I am able to make for the shuffling case, but I doubt that any of the judges would care enough to hear it. It's pretty clear why the current decision stands and it's because the combo is complicated enough to be untenable to explain/argue with an opponent that doesn't know.

When you're flipping N coins you're going to get a variance of around the square root of N. For a million coin flips you could just flip one coin and the winner of that flip gets 1,000 more creatures than the average result.

I saw an article at some point making that argument but I don't recall where/who

I think maybe Obeyline should draw the game. The card's Gatherer rulings literally say the effect never ends

@Khahan I follow you, but that's not where I have a concern.

When you have a shortcut, you propose going to a particular, definite state. The opponent is supposed to be able to then either accept the shortcut or stop it at a specific point, and similarly identify the particular, definite state when it is interrupted. If the opponent has no ability to see the cards you are going through with Petal, your discussion is on point. The opponent can say, "I will interrupt you after the 50th casting of Petals." From the opponent's perspective, you have a somewhat randomized deck. If your opponent wishes to do something here, like cast a spell with Storm 50, that's all good. Even if the opponent wishes to mill you, they're still milling random cards, from all players perspective.

The problem, for me, comes if the opponent gets information about your top deck. Because, at this point, the opponent WANTS to stop you when your win condition ends up on top, but CANNOT specify the number of iterations it will take for that to occur. It seems precisely the same problem as Four Horseman: you lose if Emrakul is a multiple of four cards from the top, and the opponent can mill it at instant speed, and you win otherwise, but you cannot tell if it is there in advance.

In the absence of a shortcut, the player cannot simply keep executing the loop either, per the slow play rules. The "loop" here is Petals casting and returning to your hand.

That's why I'm having a hard time understanding why the Petals ruling is ok.

@ajfirecracker said in Editors Wanted - Loop Article:

When you're flipping N coins you're going to get a variance of around the square root of N. For a million coin flips you could just flip one coin and the winner of that flip gets 1,000 more creatures than the average result.

I saw an article at some point making that argument but I don't recall where/who

I think maybe Obeyline should draw the game. The card's Gatherer rulings literally say the effect never ends

I'm not writing this article to suggest new mathematical approaches to problems. I'm trying to describe how the Rules actually apply. Since there's nothing in the Rules that allows you to "estimate" the standard variance of coin flips, there's no basis to do that. I mean, your judge can do whatever, but this approach has nothing to do with the Rules.