Before GP Vancouver, some of us came up with some questions to help prep our local judges for the event. This question was proposed, but we weren't able to come to an agreement on the best answer:

Andi casts Serum Visions and then looks at the top 3 cards of their library. When a judge asks why they did that, they respond, “I mostly play Legacy; I guess my hands assumed I was casting Ponder…”

What is the infraction and fix?

(I'll let some discussion happen before posting the solutions we ended up with.)

My initial reaction is Hidden Card Error for this particular problem. The player has created a new set of three cards, though he has not added them to his hand, so they “are part of a distinct set intended by the player.” It feels like GRV in some ways as the player is explicitly resolving the spell differently, that is, Ponder instead of Visions, rather than, say, Anticipating and looking at 4 instead of 3. However, in this situation, I think that HCE provides us the most elegant fix with the least potential for abuse, based on the following:

“If the error put cards into a set prematurely and other operations involving cards in the set should
have been performed first, the player reveals the set of cards that contains the excess and his or
her opponent chooses a number of previously-unknown cards. Put those cards aside until the
point at which they should have been legally added, then return them to the set.”

In this case, then, the “set” should have 1 card, to be drawn by the player, followed by a second set of two scries. NAP can set aside the two cards for the scry, because AP was eventually going to have knowledge of these three cards. If we were to apply the more common HCE fix, shuffling away the 2 excess cards, then NAP gets to know what card was drawn, but AP essentially gets a free shuffle before scrying two new cards, which I don't really like.

This certainly seems to fall under HCE.

Originally posted by IPG 2.3:

This infraction does not apply to simple dexterity errors, such as when a card being pulled off the
library sticks to another card and is seen or knocked off the library. The cards themselves must
be part of a distinct set intended by the player.

The former does not apply, so this is not LEC. The latter does apply, so HCE seems to fit. We really only apply GRV if another GPE does not fit more appropriately.

The AP should be drawing one card, but currently has a set of three. Allow NAP to put 2 of them back into the library (shuffling the randomized portion), put the remaining card into AP's hand as the drawn card off Serum Visions, then finish resolving Serum Visions.

Edited Matt Braddock (Feb. 26, 2017 02:08:59 AM)

This is LEC. A rules or dexterity error that causes a player to look at cards starts as LEC. Then you test the scenario further to see if something makes it HCE or a GRV. (Bear with me; this is long)

Originally posted by MIPG 2.2:

A player takes an action that may have enabled them to see the faces of cards in a deck that they were not entitled to see…

A player can accidentally look at extra cards easily and this infraction handles situations where a dexterity or rules error has led to a player seeing cards in a library that they shouldn’t have.

What we have is a rules error that has caused someone to see the faces of cards they should not have.

Now we test for HCE/GRV

Originally posted by MIPG 2.2:

Once those cards have joined another set, the infraction is handled as a Hidden Card Error or Game Rule Violation.

The key word in this statement is set.
HCE also rules out anything that has not joined another set.

Originally posted by MIPG 2.3:

This infraction only applies when a card whose identity is known to only one player is in a hidden set of cards both before and after the error…

The cards themselves must be part of a distinct set intended by the player.

In order for a group of cards to be a set, they must be 1) physically distinct, and 2) defined by a game rule or effect.

Originally posted by MIPG 1.5:

A set is a physically distinct group of cards defined by a game rule or effect. It may correspond to a specific zone, or may only represent a part of a zone. A set may consist of a single card.

When resolving the first part of serum visions, the effect says draw a card. This means move the card from one zone to another. It is physically distinct, but not in a grouping defined by a game rule or the effect of a spell or ability. (The second part of Serum Visions does define a physically distinct group of cards to perform an action on which would be a set, but he looked at those cards before that set was defined)

When drawing a card, it is considered to be in the library until it touches the next group of cards defined by a game rule or effect. The action of moving it does not put it into a “subset.”

Originally posted by MIPG 1.5:

Cards are considered to be part of a set until they join another set. There is no in-between state.

Originally posted by MIPG 2.2:

Cards are considered to be in a library until they touch cards in another set.

So, since a rules error has been committed which has lead to cards being seen, but those cards were not added to another set, the best match for this scenario under current policy is LEC.

Originally posted by David Rockwood:

When resolving the first part of serum visions, the effect says draw a card. This means move the card from one zone to another. It is physically distinct, but not in a grouping defined by a game rule or the effect of a spell or ability. (The second part of Serum Visions does define a physically distinct group of cards to perform an action on which would be a set, but he looked at those cards before that set was defined)

I don't believe this error is LEC, despite the fact that the cards did not enter a zone as defined by serum visions. First, what's important is that the cards come to occupy a distinct set, not necessarily changing zones. For example, looking at the top 4 cards with Anticipate is still HEC, despite the fact that only one of those card ever actually leaves the library as far as game rules are concerned. In that example, there is an extra card in a set (the cards from the library being looked at), and the opponent cannot confirm which card is “extra” so we cannot apply the LEC fix.

The same applies in this scenario: serum visions' controller knowingly and willfully created a set of three cards. We cannot apply LEC because the fix does not work. Granted, we should not consider how to fix a scenario before deciding what the penalty is, so consider also

Originally posted by IPG 2.4:

The cards themselves must
be part of a distinct set intended by the player.

(Emphasis mine). The player here clearly intended to “resolve ponder” and, in doing so, create a set of three cards to look at, just like Anticipate.

Thus I am confident this cannot be an LEC, regardless of whether Serum Visions itself causes a zone change / creates a specific set. Indeed, as David noted, the player would have eventually created a set (the scry 2) but “he looked at those cards before the set was defined.” I would argue that, in fact, the set becomes defined when the player removes the cards from the top of the library.

*Edited for grammar error.

The definition of a set requires that a game rule or effect creates it though. (See above definition) I don't believe a player can just create a set through their intention. To draw a parallel, here is a (slightly modified) real situation that happened.

Someone forgets that they attacked with a Wharf Infiltrator instead of a Grim Flayer. They say trigger and I say ok. They immediately pick up the top three cards of their library and look at them. (They don't add them to the hand or any other grouping of cards.) I immediately point out which creature attacked and we straighten things out and call the judge over.

This is also LEC. The fact that there was a group of cards touching each other did not “create a set” because there was no effect in the game to create it. The fact that something that never existed would have caused a set to be created doesn't change this.

Serum Visions resolves top down, so the player should have drawn a card. Instead, they “resolved” something that didn't actually exist, and looked at cards they were not entitled to see.

To take this further, if we apply the HCE fix, there have to be excess cards in a set. We take the excess cards in a set and put them where they should be. In this case, that would mean taking them from the library, and putting them into the random portion of the library. (Remember, cards are considered not to have moved until they touch cards in another set.) If we apply LEC, the effective difference is shuffling the third card as well. Shuffling all cards back makes more sense than giving the opponent an advantage as “punishment” for this mistake.

My argument follows:

The philosophy behind HCE is to prevent a player from gaining an advantage by “losing” an extra card in a set it shouldn't be in. I.e. if a player actually draws an extra card, we give the opponent the option to choose which card is “extra.” That is because it is impossible to know which cards were there before, and which cards were not. That is because the position of the card that should not be in the hand is not public information, and is not correctable with public information. Therefore, it is impossible to get back to a correct game state. The next best thing is to allow the opponent to choose which card is extra to eliminate the potential for gaining an advantage from this mistake.

Originally posted by MIPG 2.3 - Hidden Card Error:

A player commits an error in the game that cannot be corrected by only publicly available information

In the original case, none of the cards are mixed with other groups of cards that should exist. The HCE fix makes less sense in the OP, because the error is correctable with public information. Both players can agree, “Those are the three cards that were seen.” The game state before the error was a Serum Visions on the stack, and three random cards on top of the library. If we apply the LEC fix, we shuffle the three cards into the library. Now we have a Serum Visions on the stack with three random cards on top of the library. We have another correct game state. There is no need for the more punitive HCE fix, because there are no cards “lost” where they shouldn't be.

Edited David Rockwood (Feb. 27, 2017 05:45:43 AM)

In relation to the sets being made or not, in the IPG section for HCE we see example B is scrying 2 cards instead of 1 (which isn't moving them into a different zone) and also:

“For example, if a player resolves Collected Company, picks up three cards with one hand and then four cards with the other, the last drawn set of four cards should be used for the remedy, instead of the full set of seven cards.”

These cards are still technically in the library, and they have not touched his hand of cards (as he is using both hands for the CoCo). But these are still defined as sets in the IPG.

Because of that, I would agree with Michael. If we accept the 3 cards are a set, which I am for this, then they were put there prematurely, and since he will see all 3 cards anyway, the opponent can put 2 aside and have the caster draw the last card, then scry the two that were put aside.

Alternatively, you could see the three cards as being the set of cards he was drawing through the first line of Visions. In that case, this paragraph covers is:

“If a set affected by the error contains more cards than it is supposed to contain, the player reveals the set of cards that contains the excess and his or her opponent chooses a number of previously- unknown cards sufficient to reduce the set to the correct size. The cards chosen are treated as excess cards.”

The set was meant to be one card but is three. The fix is the same, opponent chooses two cards and returns them to the top of the library, since they'll both be scried anyway.

That is how I see it, would like to see if there is a definite answer.

Based on the most recent policy update, this seems pretty clearly to fall under HCE. The guidance we got in the update was to look at the intention behind picking up the extra cards. If the offending player intended to pick up that many cards, then it's HCE. If they accidentally grabbed extra cards, it's LEC.

The fact that there was no game rule specifically asking AP to perform actions on the top several cards of the library just yet is immaterial, since there is a effect-defined set that actions are being performed on (the single card that should be drawn per serum visions) that currently has too many cards in it. The fact that AP did not intend the cards to be drawn because they were mistaken about the effect creating the set is also immaterial.

In fact, I'd be inclined to not shuffle back any of the three cards, since serum visions specifies action on the top three cards of the library (draw the first, scry the next 2) and we still have the top three cards of the library in a specific set. It is much cleaner to reveal the three cards, have NAP choose which one is the top card, and then have AP scry the remaining 2.

In applying LEC and HCE, we should try to avoid giving players additional chances at the identity of cards they are supposed to have access to in the first place. This is why we have the “set aside” remedy in HCE for cards drawn prematurely. Once you've seen those cards, we don't want to shuffle them away to further mess with the gamestate, we only want to re-randomize (by shuffling away) cards that are really in excess of the cards you were supposed to have access to.

My ruling would be: HCE (warning) - reveal the three cards. Have NAP choose which one was the top card and add it to AP's hand, then have AP finish resolving serum visions by performing a scry action on the remaining two cards.

Originally posted by David Rockwood:

This is LEC. A rules or dexterity error that causes a player to look at cards starts as LEC … What we have is a rules error that has caused someone to see the faces of cards they should not have.

I disagree. This is not a rules error or dexterity error, and it falls under HCE.

“Set” just means ~“group of cards we can identify.” That last bit of Philosophy on HCE says

Originally posted by IPG:

For example, if a player resolves Collected Company, picks up three cards with one hand and then four cards with the other, the last drawn set of four cards should be used for the remedy, instead of the full set of seven cards.

In this situation, the player picked up three cards – that's the set.

Originally posted by Andrew Keeler:

My ruling would be: HCE (warning) - reveal the three cards. Have NAP choose which one was the top card and add it to AP's hand, then have AP finish resolving serum visions by performing a scry action on the remaining two cards.

I agree wholeheartedly. HCE is about trying to minimize the advantage the player could have gained, and this fix does that.

There's been some good discussion here, and it helps us all think about the distinctions between L@EC, GRV, and HCE.

Until recently - i.e., the creation and clarification of HCE - I might have considered it a GRV, and applied the L@EC remedy as part (or all) of the fix. That was then…

This, now, is HCE. The player has a set of cards with too many in it, and - here's two key indicators! - the opponent can't fix it, and the player intended to pick up that many cards. (Note that, if they tried to pick up 3 but got 4, this would still be HCE in this example - because they weren't supposed to grab 3, either.)

As I stated in another thread, an example of rules error (“…situations where a dexterity or rules error has led to…”) is paying B, not U, to cast Ponder.

Originally posted by David Rockwood:

In order for a group of cards to be a set, they must be 1) physically distinct, and 2) defined by a game rule or effect.
-and-
The definition of a set requires that a game rule or effect creates it though.

That's not actually what the definition of HCE says about a set; in fact, you quoted the relevant reference: “The cards themselves must be part of a distinct set intended by the player.” In this example, the player created a set of three cards by removing them from his library, and that's what he intended to do - pick up three cards. I see where people are getting that understanding, but that's not how it's meant to be read; the player created the set by intentionally picking up three cards.

d:^D

P.S. - apologies to David Poon, for perhaps dismantling his intention of sharing their conclusions… ;)

Edited Scott Marshall (March 1, 2017 11:31:16 PM)

Originally posted by Scott Marshall:

That's not actually what the definition of HCE says about a set; in fact, you quoted the relevant reference: “The cards themselves must be part of a distinct set intended by the player.” In this example, the player created a set of three cards by removing them from his library, and that's what he intended to do - pick up three cards. I see where people are getting that understanding, but that's not how it's meant to be read; the player created the set by intentionally picking up three cards.

Let me try and understand then.

The definition of set in MIPG section 1.5 says “A set is a physically distinct group of cards defined by a game rule or effect.” My interpretation was that a “set intended by the player” was “a physically distinct group of cards defined by a game rule or effect intended by the player.”

Are there two definitions of set, or am I interpreting the first wrong?

Edited David Rockwood (March 1, 2017 11:58:28 PM)

Ah, I see where the disconnect begins. I'll recommend some clarification in that definition.

The player created a set via the effect he thought he was resolving - Ponder, instead of Serum Visions. Even though the card's effect doesn't match what he did, he intended to pick up a set of 3 cards, and did so.

d:^D

Can I have a reminder on which clarification of HCE stops this from beeing GRV? This would have been my first guess because we try to look at the first thing going wrong here which was not drawing a card for the visions or we could also say that there still would be a mistake if the player would only look at two cards and this error seems to be very GRVish to me.

Edit: Short clarificarion of what I meant above: Even if we do the HCE-fix on the scry-set of cards and fix it so there are only two cards remaining, we still have the problem that there was no card drawn by the player for Serum Visions first part.

Edited Markus Dietrich (March 2, 2017 11:04:22 AM)

Originally posted by Markus Dietrich:

Can I have a reminder on which clarification of HCE stops this from beeing GRV? This would have been my first guess because we try to look at the first thing going wrong here which was not drawing a card for the visions or we could also say that there still would be a mistake if the player would only look at two cards and this error seems to be very GRVish to me.

Edit: Short clarificarion of what I meant above: Even if we do the HCE-fix on the scry-set of cards and fix it so there are only two cards remaining, we still have the problem that there was no card drawn by the player for Serum Visions first part.

The error happened when AP “drew” the second and third cards, instead of scrying them separately. We're not using the HCE remedy to recreate the set (of 2) for AP to scry, we're recreating the set (of 1) for AP to draw. It just so happens that the 2 other cards are ones he should be performing another action on immediately after.

Originally posted by Matt Marheine:

Can I have a reminder on which clarification of HCE stops this from beeing GRV?

The idea behind “or a rules error”, in the wording of HCE, is for situations where a rules error occurs, the opponent could notice & stop it at that point, and then cards are drawn or looked at, etc.

In this example, the first (only) indication that the opponent has of something going wrong is when the cards are already being looked at.

d:^D