I had some good conversations over the past few weekends with Patrick Chapin (congrats on the induction, Pat!) about the rules for infinity in Magic and, specifically, how they interacted with the Four Horsemen deck. I’m not going to hash out the reasons for the existence of the non-determinate looping rules – there’s an excellent reddit thread on that – but talking to him exposed some misperceptions on both player and judge sides that I think are worth clearing up.

For the record, I actually think the Horseman deck is pretty cool. I don’t play (non-Commander) constructed because the fundamental goal – getting a consistent performance out of your deck – is not the sort of thing I look for out of Magic play, but I admire the Swiss-watch way in which the deck fits together and what it’s trying to do. It’s an ingenious brew, and anyone who thinks these rules were put in place to kill it is off base.

There has been, it seems, some bad communication from the judge community. Some players have been told that playing Monolith and Orb together is “illegal”. It’s not, and if you’ve been giving that impression, please don’t. Both cards are perfectly fine to play, and some other combos, such as Lab Maniac, make use of it in legal ways. Taking it a step further, activating them to mill your deck, even if you know there’s an Emrakul in it, is also fine. Problems arise when the deck demonstrates that it can end up where it started, not before. Don’t jump the gun on this.

There are also persistent questions about what qualifies as “the same game state”, and both players and judges have misapplied this. It’s not something that we could realistically codify; there are just too many variables. In general, though, returning to the same state means that the physical state of the game *as it relates to the loop* has reverted back to a previous state. So, if the iteration of the loop popped a Narcomeba into play, or did a point of damage to the player, that’s a change to the game state that’s directly related to the loop in question and the game is clearly advancing. If the player tapped an island halfway through just to “change” the game state, that’s not meaningfully advancing the game, as it’s not part of the looping process. Stuff does happen during a lot of horsemen loops, and stepping in prematurely is incorrect. By far the most common (only?) way to actually achieve a same game state is to hit Emrakul, shuffle (state A), start up, then hit Emrakul again before any other piece and shuffle, returning you to state A. That is the point at which the judge should be intervening.

One complaint that is valid but misguided is that the loop itself isn’t really slow, and it’s unfair to make it Slow Play when stuff like Eggs or High Tide can make you sit there and pray for death while not being Slow Play. This is true, and that’s due to a quirk in how the IPG is laid out and updated. The reasons that these loops are not allowed is a tournament structure concern and how it interacts badly with Magic’s infinity rules, but not strictly a slow play concern.

So why is it handled there? Fundamentally, because we want judges to treat it exactly the same way that they would treat slow play. Most situations involve a judge coming in, determining an infraction and applying a penalty. But Slow Play is different. It’s incredibly rare for a judge to just come in and *wham*, Warning. That’s because we understand that players will tank on occasion and may not even realize it. That’s not a problem. So we solve that by having the judge issue a gentle reminder to the player that their pace isn’t what it needs to be, and only resort to warnings if the player makes a habit out of it. It’s a kinder, gentler approach for a problem we don’t want to bring the hammer down on. And that’s how we want to handle these loops. It isn’t a problem that the player started the loop. When they get to a game state reset position, there’s no need for a penalty at this point, just a friendly reminder from a judge that what they’re trying to do isn’t really compatible with the structure of a Magic tournament and they need to do something else. It’s only when they keep trying that we should be issuing Warnings.

From a marketing standpoint, perhaps it should be a separate infraction (“Non-Game-State-Advancing Loop Execution”, bleah) so that it doesn’t carry the stigma of Slow Play (which is a bit weird, because Slow Play feels to me like one of the least stigmatizied infractions out there, and we’re not even issuing penalties) or open us up fighting against it with the idea that it isn’t a lower-case slow play issue. The downside, of course, is that it’s adding an extra half page to the IPG for an incredibly rare situation, which means there’s more to learn, more to be tested on, software to update, etc, etc. Those things aren’t without cost, so when something can be slid in, we’ll strongly consider doing it, even if it means the name isn’t quite right. For years, we treated missing your draw for the turn as a Missed Trigger, even though it wasn’t even a trigger, and if I could go back and change one thing in the IPG, I never would have called it a Player Communication Violation!

And remember, please be diplomatic with a player who discovers this rule at an unfortunate time. It takes a fairly restrictive set of circumstances to find the deck and not find out about the problem, but we’ve seen that it can happen, and with a card pool the size of Magic’s, it’ll probably happen again with a different deck at some point.