Let’s Talk Mulligans

During the banned list announcement last week, we brought up mulligans. In case you missed it, here’s what we had to say:

There will be no rules changes this round, but given the success of the mulligan tests in Vancouver this year we are taking a close look at the
Commander mulligan. Our unique “EDH mulligan,” aka, Partial Paris, has always been a great tool to help ensure that each game of Commander was fair and
memorable as possible, but it has always had its downsides as well. We believe that the Vancouver mulligan might fill many of the same roles as Partial
Paris while alleviating some of those downsides. We’ll be making our final decision on whether to switch in the next cycle, so until then any input
that players can provide will be read and appreciated!

You can expect something definitive in the next announcement in January, but the message to take away here is that the days of the Partial Paris mulligan
are numbered. The main issue with the Partial Paris is that it allows a certain amount of hand sculpting. While that might be desirable in a competitive
environment, it’s not so much for us. We adopted it a long time ago when it seemed the best choice to give players the opportunity to start with a playable
hand; one of the problems is that it allowed players to craft an explosive start. Getting a starting hand that gets you into the game is our main goal.
Nothing is worse than wanting to play and not being able to. When you want to be a spectator, Magic can be great to watch. When you want to be an active
participant, spectating is the nut low. So where is the solution? Let me float a few ideas-but before I do, I’ll remind everyone that this is the
equivalent of me thinking out loud. Although we have thrown around a few things, none of these necessarily has the inside track on what we’re going to do
(although some don’t in truth have that much of a chance, but they’re good to frame the argument). I’m just trying to see what sticks to the wall. As
always, your opinions are welcome. If you have a great idea that’s not included below (or a variation thereof), ship it. We are happy to listen to
reasonable suggestions.

Before we move on, I want to tell all you metal fans that if you haven’t tuned in on Iron Maiden’s new album Book of Souls, you’re missing out.
It’s 90 minutes of classic-style Maiden-think Powerslave with nearly 30 years better production value. Give it a listen.

I’ll operate under four basic criteria. First, we want to give players the best possible chance of having a playable starting hand. Second, we want to
ensure that in doing so, we’re not engendering or encouraging the kind of hand crafting that leads to early turn kills. This is a tall order. As much as
possible (and I understand this is a little idealistic), I’d like to see the mulligan go from a purely strategic decision (“does this hand give me an
explosive early lead in the game?”) to a playability decision (“can I at least be in the game with this grip?”). My personal preference (and this isn’t
necessarily shared by everyone on the RC) is that the only evaluation is “can I play it or not?” It’s a Boolean instead of an evaluation. Your mileage is
certainly welcome to vary. The real thing is that we just want folks to be playing Commander. That means both not having to watch because you can’t play
anything and not having to watch because someone else can play everything. Third, shuffling 99 cards between every mulligan
decision is pretty sucky. It wastes time that you could be instead playing. Shuffling isn’t completely off the table, but it’s definitely a thing. Not
shuffling (and I mean shuffling back in everything at the end) does two things. It inhibits hand sculpting for early combos because once a combo piece has
passed by, you can’t get it in a subsequent hand. Based on the liberal nature of the mulligan, however, it’s going to give you a better insight into what’s
left in the deck, which might lead to a different set of problems. Finally, the rule has to be both easily written and easily understood. Magic players are
reasonably clever, so we’re reasonably sure you can handle some complexity, but we don’t want you to have to have flowcharts, critical hit tables,
semaphores, or understand Morse Code to make sense of it (Side note: in honor of Jack Black’s band in High Fidelity “Kathleen Turner Overdrive,” I
wouldn’t mind having a band called “The David Morse Code”). I’ll also operate under the assumption that simply keeping Partial Paris is not an option. So
what’s on the table?

Do Nothing

In this option, the official mulligan rule of Commander is the official mulligan rule of Magic. The multiplayer option would kick in, so you would go
7-7-6-5 and so on, with a scry 1 if you take less than seven. The upside to this is that it’s easy, both from a comprehension standpoint and a
rules-writing one. No one has to learn anything new, do anything that they’re not doing in other formats, and no one has to add anything to the Magic
Comprehensive Rules. The downside is that it might not be enough. In a 100-card singleton deck (although I suppose you’re actually drawing from just 99),
variance is pretty high. While we enjoy a reasonable amount of variance as one of the hallmarks of the format, it might simply be too much. Someone that
understands the math a little better than I do might be able to speak to that more intelligently. Shuffling in between would apply, since that’s what you
do in the official rules.

Set Number Pattern

This is any option that stipulates a specific number of cards you draw based on how many times you’ve mulliganed-like 7-7-6-6-5-5 and so on, or
7-7-7-6-6-6, etc. This could be done with the shuffle or the no shuffle option (although, again, the shuffle is pretty dreary). The upside is that you’re
more likely to get a playable hand. The downside is that you have to remember which number you’re on. “Did I take 6 twice or just once?”
Additionally (and again, here is where a math person could weigh in), 7-7-6-6 might not be statistically significant from 7-7-6-5, meaning that it might
matter in so few cases that it’s not worth the trouble to write a rule over.