Warning: The example described here are meant to be at the ends of the Backup spectrum. Most situations won’t be as clear cut and small alterations in the presented scenarios may already affect the decision to backup or not. Take this article as it is: A presentation of some core reasonings you can use before making a ruling of yours!
Last week-end in Birmingham, when handling an appeal, I could realize that the main reason the judge chose to leave the situation as it was is that a Fetchland has been sacrificed since.
While it is extremely important to consider the presence of a Fetchland on the battlefield, I’m under the impression this has turned into the belief that a Fetchland on the battlefield fully denies the possibility of Backing up.
While it is true that the presence of a Fetchland can strongly undermine a backup, it’s important to evaluate how much that is true!
Case 1: Backing up through a drawn card
AP casts Tarmogoyf. NAP controls an Island and a Scalding Tarn and taps both to cast Remand.
Tarmogoyf goes back to AP’s hand, NAP draws a card, AP passes the turn, NAP realizes the mistake as they untap the Scalding Tarn.
Restoration point
First, if you backup, you need to backup to the point Tarmogoyf is on the stack and Remand has been sent back to hand.
Indeed, you can never backup halfway through casting a spell, as this would mean a player may have to target themselves with a Lightning Bolt if they forgot their opponent’s Leyline of Sanctity
Note: While you can’t back to the middle of the announcement of a spell if the error happened while the spell was cast, you can backup to the resolution of a legally cast spell if the error happened while resolving the spell.
Performing the backup
As suggested in the article linked before, here’s the step by step version of the backup.
- NAP taps their lands
- NAP puts a random card back to the top of their library
- NAP puts Remand back into their hand
- NAP untaps their lands.
Consequences of leaving as is
The most common and foreseeable consequence is that NAP may adjust their mana base more easily by sacrificing the fetchland later in the game.
Consequences of the Backup
Everything is straightforward except 2.. Indeed, the card that’s put back to the top of the library is potentially not the one which was there before. Additionally, it is now known to NAP.
And this is the game changer: NAP can now make decisions based on this knowledge:
- Does NAP now want to keep that card and let Tarmogoyf resolve instead of casting Remand?
In which case the original strategic decisions have been strongly affected, and AP now knows that NAP has a Remand in hand. - Does NAP still want to Remand the Tarmogoyf but feels bad because he lost a great card in the process which he previously had in hand?
In which case the quality of NAP’s hand has certainly diminished - Does NAP want to shuffle that card away?
In which case the quality of NAP’s hand has potentially improved
Two of the three consequences of the backup do affect the game state a lot. The last one is harder to evaluate but can do as well. On average, backing up will likely affect the situation more than just leaving it as it is.
It can be argued that two of these three situations are actually in favor of AP and that it “punishes” NAP for making a mistake but we’re not talking about “punishing” a player here, we’re talking about restoring the game state to an organic state.
Note that in Legacy, that’s a tougher call to make since NAP may be able to take more advantage of a Brainstorm later. I’ll set this aside since the MIPG has never handled Brainstorm greatly anyway.
Conclusions
If a card has been drawn since the mistake happened, the potential for the fetchland to strongly affect the game state is so important that leaving the situation as it is is generally safer.
Case 2: Backing up through an already sacrificed fetchland
NAP controls Linvala, Keeper of Silence. AP taps two Forests and Birds of Paradise to cast Tireless Tracker, which resolves. AP then plays and sacrifices a Windswept Heath, searching a Temple Garden, then puts two clues on the battlefield. Finally, they pass the turn. As NAP untaps their permanent, AP calls the judge and points out that they couldn’t tap Birds of Paradise for mana.
Restoration point
The error happened when casting Tireless Tracker. Therefore, the restoration point is just before Tireless Tracker was cast.
Performing the backup
- NAP taps their lands and creatures
- AP removes the two Clues
- AP puts the Temple Garden back into their library
- AP puts Windswept Heath back into their hand
- AP sends Tireless Tracker back to their hand
- AP untaps the two Forests and the Birds of Paradise
Consequences of leaving as is
Without this mistake, AP would not have been able to have the two clues on the battlefield and had they still cast Tireless Tracker, NAP would have had an entire turn to deal with it before clues show up. This means that potentially, AP can draw an additional two cards one turn earlier.
3. is the only non-straightforward step, since it involves reshuffling the library, which changes its order.
However, if the library was randomized before the Fetchland was activated, this is not an issue at all: If both top and bottom cards were unknown before the shuffle, they remain unknown after the shuffle.
Comparison of both options
In this situation, backing up feels completely free despite a Fetchland has been sacrificed. The main reason for this is that the activation of the Fetchland did not affect the game state at all.
On the other hand, the consequences of leaving as is are tangible.
Conclusions
Despite a Fetchland was sacrificed, the fact no card had been drawn makes this backup fairly easy.
The key point here is that nearly every element post backup is exactly as it was before: Random library is still random, every card is where it originally was. The notable exception is that AP revealed some cards from their hand but I usually say that as long as no instants were revealed, that’s not a big deal.
In the latter scenario, things will be a little more complicated to assess if the top or bottom card were known prior to AP activating the Fetchland. The whole question becomes: How much does the reshuffling of this top or bottom card affect the game state?
If the bottom card was known (due to AP having taken a mulligan likely), this doesn’t change much, especially as the player will have the option to sacrifice the Fetchland the turn after, or still this turn if needed!
Even if this was a one-of combo key-card, the odds that this card gets back to top instead of the bottom are extremely low: There are about 50 cards left in the library in this example, so it’s around 2% chances. Even if there were as low as 25 cards in the deck, that would only represent a 4% chance.
If the top card was known, that’s 100% affecting the next draw step, especially if AP doesn’t play and sacrifice the Fetchland in the end (which is the optimal decision if they have another land and something else than the Tracker to cast). It’s very hard to assess the consequences of this from a purely theoretical point of view hence I won’t pursue in this direction.
Kevin Desprez.