At yesterday's SCG Legacy Open in Columbus, there was a GRV situation where there was a point of discussion about whether a backup was appropriate or not, and several of us on staff disagreed over the correct resolution. I wanted to present it here as a scenario for discussion, because I'm curious to see where the consensus ends up on this one.

In a Competitive REL Legacy tournament, Player A casts a Liliana of the Veil, despite not having the ability to produce two black mana. Player B controls a Griselbrand, and while he has no cards in hand, he knows the top two cards of his library because of a Brainstorm that resolved in his previous turn. In response to the Liliana of the Veil, he activates Griselbrand's ability, and draws seven cards. He is about to cast a spell in response to the Liliana, when both he and Player A realize that Player A does not have the ability to cast Liliana of the Veil.

We know that this is a GRV for Player A and FTMGS for Player B. The question is, do we back up? And if so, what do we do to backup the game?

Edited Casey Brefka (Jan. 20, 2014 03:15:37 PM)

I don't think we back up here - basically because we can't get all the way back to where we were. We can't verify the 2 cards that should be on top.

I don't like the idea of backing up unless I can get the game back to where it was.

I'm with Mark on this one for the same reason; no backup.

Edited Eric Paré (Jan. 20, 2014 03:31:40 PM)

Mark: You can't always get the game back to where it was and that usually
is not the criterion used for weather to back up. The question is more
about the disruption to the game. There are often situations where a drawn
card is backed up by putting a random card back on top of the library. It's
not exactly where the game was (unless the card that was on top happens to
be the randomly picked card), but can generally be considered
non-disruptive in some situations. I won't say all, because there are times
where a draw, and particularly multiple draws are too disruptive to back up
when a lot of other actions have occured.

This situation is made a bit more awkward by the Brainstorm. If we back up,
we're going to randomly put the 7 cards now in the player's hand back on
top, the two cards the player Brainstormed to the top will be somewhere in
there, and the player now knows the Top 7 cards of their deck, though not
the order. Because the Brainstorm cards aren't known to the other player,
we also can't shuffle away the 5 extra cards that got seen if we chose to
follow the Looking at Extra Cards fix. The fact the player has the
Griselbrand in play and is probably going to draw 7 cards at some point
anyways is irrelevant to the situation.

I believe that backing up is disruptive enough to say that we leave the
game state as-is.

-George FitzGerald
L2, Sarasota, FL

The options here are to not back up, and leave a Liliana on the field or put the seven (Player B's whole hand) back at random. I don't really like either of these remedies. We should only back up when there is not “too much disruption to the course of the game”. I feel that backing up here would change the course of the game since the player put those two cards back with Brainstorm for a reason (the precise reason isn't relevant). The only way we could do it would be to put them back randomly so we will almost certainly bury those cards under a few others.

Originally posted by George FitzGerald:

Mark: You can't always get the game back to where it was and that usually
is not the criterion used for weather to back up.

To back up: “Each action taken is undone until the game reaches the point immediately prior to the error.” If we can't get to that point, we can't back up (no partial backups allowed). I would argue that “whether it's possible to backup” is just as important as “has too much damage been done/would it be too disruptive”.

Originally posted by George FitzGerald:

The fact the player has the
Griselbrand in play and is probably going to draw 7 cards at some point
anyways is irrelevant to the situation.

It could be relevant actually. Player B does not have to draw through those 7 cards first to get to the others. He can crack a fetchland or Ponder to shuffle his library to reset the top cards in hopes of drawing better answers if needed during the game.

Except that strategy decisions are not relevant to our decision to back up
or not.

Mark: You also missed this part of the IPG that talks about back-ups
involving draws.

“Cards incorrectly placed in hand are returned to the location in the zone
from which they were moved (if the identity of the incorrectly drawn card
is not known to all players, a random card is returned instead).”

Your assertions that to back-up you must arrive perfectly at the game state
prior to the GRV are incorrect.

No cards were “incorrectly drawn” here though, were they? Or does the IPG mean that any cards drawn after the GRV are “incorrectly drawn” (for the purposes of rewinding)?

Marc, it does. The game post-GRV is essentially all “incorrect” if we
are rewinding - every action after the GRV is from a doomed timeline,
sometimes the darkest timeline.

On 1/20/14, Mark Mc Govern <forum-8028-9d4b@apps.magicjudges.org> wrote:
> No cards were “incorrectly drawn” here though, were they? Or does the IPG
> mean that any cards drawn after the GRV are “incorrectly drawn” (for the
> purposes of rewinding)?
>
> ——————————————————————————–
> If you want to respond to this thread, simply reply to this e-email. Or view
> and respond to this message on the web at
> http://apps.magicjudges.org/forum/post/48552/
>
> Disable all notifications for this topic:
> http://apps.magicjudges.org/forum/noemail/8028/
> Receive on-site notifications only for this topic:
> http://apps.magicjudges.org/forum/noemail/8028/?onsite=yes
>
> You can change your email notification settings at
> http://apps.magicjudges.org/profiles/edit


–

~Rob McKenzie
Magic Judge Level III
Minnesota

I don't think this changes things, but from the scenario it seems like Liliana is still on the stack and hasn't entered the battlefield yet. The only question would be is according to the CR if the casting of a spell was illegal you'd undo all the actions for casting that spell.

Of course I'd agree that in this case you'd have to back up and undo all the actions or not (since thats according to the IPG).

Interesting one indeed. My personal feeling is that, while both outcomes (allowing the Liliana to be cast off incorrect mana vs. randomizing the top 7 cards on the opponent's library) are disruptive to the game state, the latter one is less disruptive; if the player wants the 2 cards off his Brainstorm, he can just activate Griselbrand again, whereas there is no logical way to justify a Liliana being cast off incorrect mana. As a result, I'd be inclined to back up here.

I would do a backup here, nothing really tells me otherwise in this scenario. The backup is 7 random cards from NAP hand to top of his library, Liliana back from the stack to AP's hand.

The point is GRV+FTMGS means both AP and NAP are responsible for the incorrect game state and NAP can blame himself for having different cards in his topdeck.

On a side note I'd feel very bad looking at that Liliana illegally resolve with a fix so easy at hand.

Someone mentioned the disruption of the Griselbrand player having seen an extra 5 cards, and if we rewind, he can now decide to maybe crack a fetch and shuffle before using Griselbrand to draw another 7.

This seems to be one of those places where you think “well, then I guess the Lili shouldn't have been cast, eh?”

We could go on for quite some time counting the what-ifs with regard to which cards the person saw or wanted.

It seems to me that any advantage B gets from a rewind is a result of A making a mistake. And any disadvantage B gets is a result of not maintaining the game.

The first question in the Additional Remedy after ‘did we catch it in time’ is ‘is the rewind too disruptive?’ In trying to understand “too”, I'm not really interested in how disruptive not-rewinding would be. That seems to me to be taking the game state into consideration too much.

So when answering “is rewind too disruptive?”, I think the answer no. Ask for a rewind.