Short Form Lizard

360 Memnites

June 17, 2026 | by Liz

Background

So as you might suspect I talk to a lot of people who have a lot of opinions about Magic: The Gathering, and specifically cube draft. On some random day in early 2025, there was some discourse about frameworks people use when designing cubes and what makes a "good" cube. And by chance a funny phrase caught my eye:

"...as cube is a hobby (and not say art), any theory of cubing is only useful insofar as it leads to more enjoyable game play..."

And to be fair, I do think this (and all the other context I omitted) synthesized to a good point. But immediately my mind floated about the idea of a cube that would fail to satisfy both "cube as hobby" and "cube as art" camps of design philosophy. What spawned was 360 Memnites.

What is 360 Memnites

360 Memnites is a hypothetical Magic: the Gathering cube draft environment that consists of 360 copies of the card Memnite (In pursuit of making this post as accessible as possible, I've tried to write in such a way that you don't really need to know what any of these words mean). Memnites (This is what I'm going to call the cube from now on it saves me 4 keystrokes) is meant to be the worst possible experience I could imagine. There are no drafting descisions. Every game is deterministic. It's not fun. No one should play Memnites. No one should think about Memnites. If you believe these facts to be true, for your own serentiy I encourage you to stop reading now.

There's more?

Okay so I lied a little bit. Memnites is actually really fun to think about (though I still hold no one should play Memnites, I do not belive it would be fun to play). I'm going to walk you through the mind-bending rabbit hole I spun myself down and hopefully you find some of it funny and/or enjoyable.

Deckbuilding Part I: How many memnites should I play?

So for those unfamiliar with Magic: The Gathering, or the draft format, you recieve 45 cards (by some method that is not worth explaining for the purposes of this post) and then need to build a deck of at least 40 cards. Is there any particular reason we should play 40 copies of Memnite? All 45 copies? Some number in between?

Perhaps unsurprising to Magic players, you should play all 45. Combat is, if you're just playing with 40-45 copies of Memnite, not a factor since they can just block each other and cancel out any damage. If players play with fewer, they risk unnecessarily losing to the following game rule:

121.4. A player who attempts to draw a card from a library with no cards in it loses the game the next time a player would receive priority...

Basically, whoever runs out of cards first, loses. Unfortunately, this does create deterministic games. Since the player going first doesn't draw a card from their deck on their first turn, the second player will draw all their cards first.

Except...

Gameplay Part I: The Mulligan Rule

In Magic, you oftentimes draw a hand of cards to begin the game, and it sucks. Luckily, there's a rule that lets us get a new hand of cards (for a price)!

103.5. Each player draws a number of cards equal to their starting hand size, which is normally seven. (Some effects can modify a player’s starting hand size.) A player who is dissatisfied with their initial hand may take a mulligan. First, the starting player declares whether they will take a mulligan. Then each other player in turn order does the same. Once each player has made a declaration, all players who decided to take mulligans do so at the same time. To take a mulligan, a player shuffles the cards in their hand back into their library, draws a new hand of cards equal to their starting hand size, then puts a number of those cards equal to the number of times that player has taken a mulligan on the bottom of their library in any order. Once a player chooses not to take a mulligan, the remaining cards become that player’s opening hand, and that player may not take any further mulligans. This process is then repeated until no player takes a mulligan. A player can take mulligans until their opening hand would be zero cards, after which they may not take further mulligans.

In order to illustrate how a player could use this strategy, and all future strategies described in this post, I will have imaginary players Aurelia and Bruna play each other. Aurelia will always be the starting player.

Aurelia draws seven cards and decides to keep.
Bruna draws seven cards and decides to mulligan. Bruna likes her hand of six cards and keeps.
When the game begins, Bruna has one more card in her library. This means Aurelia will draw through her library before Bruna, and will lose to rule 121.4.

Now, clearly Aurelia has a counter strategy of also taking a mulligan, but lets see how that plays out.

Aurelia draws seven cards and decides to mulligan.
Bruna, seeing that Aurelia took a mulligan, decides to keep her seven cards.
Aurelia, confused, keeps six cards.

Aurelia casts 6 copies of Memnite.
Bruna draws a card and casts 8 copies of Memnite.
Aurelia draws a card and casts 1 copy of Memnite.
Bruna draws a card. She attacks Aurelia with 8 memnites. Aurelia blocks 7 of them, but takes one damage and goes to 19 life. Bruna casts 1 copy of Memnite. Bruna now controls 2 copies of memnite, and Aurelia controls none.

I won't trouble you with the rest of the entire game, but rest assured, this is the first time we have considered combat to be a viable strategy, and Bruna is able to fully take advantage of it. Given these two scenarios, no matter what descision Aurelia makes with her opening hand, Bruna will win the game.

Deckbuilding Part II: Basic Lands

So there's another rule I've kinda ignored up until this point.

905.1d After the draft and all actions that may be taken during or after the draft, all the cards a player has drafted become that player’s card pool. The player builds their deck from only these cards and any number of basic land cards...

So we don't just have to build our decks with Memnites. Basic lands, in this format, are essentially blank cards, but they may help prevent us from drawing all of our cards from our deck. Lets see what happens when Aurelia adds a single basic land to her deck.

Aurelia draws seven cards and decides to keep.
Bruna, seeing that Aurelia's deck is exactly one card taller than the last game, weighs her options...
If Bruna decides not to mulligan, she will always be able to at least match the number of Memnites Aurelia plays. If Aurelia draws the land too early, Bruna can win by combat damage one point at a time. (Win Probability: 25/46 or ~54.3%
If Bruna decides to mulligan once, she will never be able to push any damage through, and her deck is the same size and Aurelia's, so she will assuredly draw from an empty deck. (Win Probability: 0%)
If Bruna decides to mulligan twice, she is safe from drawing from an empty deck, but Aurelia might be able to beat her in combat unless she draws the land too early. (Win Probability: 26/46 or ~56.5%)

Clearly, the best option for Bruna is to mulligan twice, and she is still favored. But Aurelia has gone from winning no games ever, to winning ~43.5% of them. That's a great conversion rate for just including a single additonal card.

What if we consider the opposite scenario, where Bruna is the one playing the land?

Aurelia, seeing that Bruna's deck is exactly one card taller than the last game, weights her options...
If Aurelia decides not to mulligan, Bruna will always have at least 7 memnites, and sometimes 8 memnites in play, which is plenty of blockers. If that wasn't enough, Bruna also has more cards in her library, which means she can't lose to drawing from an empty deck. (Win Probability: 0%)
If Aurelia decides to mulligan, Bruna can then decide to mulligan with her, returning to scenario 1 with one fewer card. They can do this all the way to zero cards, and Aurelia's chances never increase. (Win Probability: 0%)

This means playing 1 land increases Aurelia's chances of winning if Bruna isn't playing a land, and doesn't change Bruna's chances of winning if Aurelia isn't playing a land, but if Aurelia and Bruna ever decide to switch who goes first, being the player with the land is obviously better. Surely this won't have drastic consequences.

Deckbuilding Part III: Simulation

Ok, so we are about to hit a huge problem. We're able to derive the probabilities of Aurelia and Bruna winning in the cases that one of them is playing one land, and maybe even if both of them are playing one land, but when we start asking "what happens if one of them plays two or more lands" all Hell breaks loose!

So I will not be doing those calculations for this article so I can remain at least somewhat sane. If I ever find there is a way to simplify these really complex calculations, I'll be sure to give this post an update. Instead, I took a slightly different approach...

I forced my computer to play 100 million games of Magic: The Gathering.

The script simulates a really rudimentary version of the game, and tries each permutation of:

• Aurelia and Bruna play somewhere between 0 and 7 basic lands.
• Aurelia and Bruna mulligan in such a way that either player could have up to 7 more cards than their opponent in their starting hand.

The computer then plays 100 thousand games of each configuration, and reports the winrate of each player. Unfortunately, that leaves us with a bunch of weird outputs (I've even cleaned it a bit and removed most of the simulations that provided deterministic results on the tail ends). But now that we have all this data on what decks win against other decks, how do we actually implement this in gameplay?

Gameplay Part II: Data-driven Decision Making

Let's take a look at some of these weird results (feel free to gloss over them though, they're honestly quite boring and difficult to read). I've gone ahead and categorized the matchups into four categories.

Mutual Benefit

45 vs 45+2
  with a mulligan diff of -3
  ftp with 0 lands wins 0/100000 games
  draw player with 2 lands wins 100000/100000 games
  ftp losses to combat: 100000
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of -2
  ftp with 0 lands wins 24197/100000 games
  draw player with 2 lands wins 75803/100000 games
  ftp losses to combat: 75803
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 24197
  ----------------------
  with a mulligan diff of -1
  ftp with 0 lands wins 0/100000 games
  draw player with 2 lands wins 100000/100000 games
  ftp losses to combat: 17657
  ftp losses to deck: 82343
  draw losses to combat: 0
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 0
  ftp with 0 lands wins 30011/100000 games
  draw player with 2 lands wins 69989/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 69989
  draw losses to combat: 30011
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 1
  ftp with 0 lands wins 75235/100000 games
  draw player with 2 lands wins 24765/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 24765
  draw losses to combat: 75235
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 2
  ftp with 0 lands wins 100000/100000 games
  draw player with 2 lands wins 0/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 0
  draw losses to combat: 100000
  draw losses to deck: 0

How do we determine optimal play from both players? Well, we start with the listing "mulligan diff of 0". This assumes either both players kept their opening 7, or they mulliganned an equal number of times. A negative diff implies Aurelia mulliganned that many more times than Bruna, and vice versa. At a diff of zero, Aurelia has a ~30% winrate. Aurelia looks at all the lower mulligan diffs and sees she has lower winrates, so it's probably not advantageous to mulligan. She also sees that at all the higher mulligan diffs she has higher winrates, but that relies on Bruna taking mulligans. In this scenario it is best for both players to keep their opening hands.

There are three potentially more interesting scenarios:

Mutual Destruction

45 vs 45+1
  with a mulligan diff of -2
  ftp with 0 lands wins 0/100000 games
  draw player with 1 lands wins 100000/100000 games
  ftp losses to combat: 100000
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of -1
  ftp with 0 lands wins 58658/100000 games
  draw player with 1 lands wins 41342/100000 games
  ftp losses to combat: 41342
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 58658
  ----------------------
  with a mulligan diff of 0
  ftp with 0 lands wins 0/100000 games
  draw player with 1 lands wins 100000/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 100000
  draw losses to combat: 0
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 1
  ftp with 0 lands wins 41271/100000 games
  draw player with 1 lands wins 58729/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 58729
  draw losses to combat: 41271
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 2
  ftp with 0 lands wins 100000/100000 games
  draw player with 1 lands wins 0/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 0
  draw losses to combat: 100000
  draw losses to deck: 0

This is a scenario we've looked at before, but with some match data to back it up. Aurelia sees that she goes from a 0% winrate to a 58.7% winrate by taking a mulligan. Clearly Bruna is not going to allow this and will take a mulligan if Aurelia does, and could mean they do so until both have zero cards in their opening hands. If Aurelia ever keeps, Bruna isn't going to get a better winrate by taking more mulligans than Aurelia, so she should also keep.

Second Actor Advantage

45+1 vs 45
  with a mulligan diff of -1
  ftp with 1 lands wins 0/100000 games
  draw player with 0 lands wins 100000/100000 games
  ftp losses to combat: 100000
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 0
  ftp with 1 lands wins 45546/100000 games
  draw player with 0 lands wins 54454/100000 games
  ftp losses to combat: 54454
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 45546
  ----------------------
  with a mulligan diff of 1
  ftp with 1 lands wins 100000/100000 games
  draw player with 0 lands wins 0/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 100000
  ----------------------
  with a mulligan diff of 2
  ftp with 1 lands wins 43731/100000 games
  draw player with 0 lands wins 56269/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 56269
  draw losses to combat: 43731
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 3
  ftp with 1 lands wins 100000/100000 games
  draw player with 0 lands wins 0/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 0
  draw losses to combat: 100000
  draw losses to deck: 0

Here's another scenario we've looked at before. Aurelia sees that by taking a mulligan, her winrate goes to 0%. Clearly that isn't a good idea, so she will keep her opening hand. Bruna sees that by taking two mulligans after Aurelia keeps brings her winrate down from 45.5% to 43.7%.

First Actor Advantage

45+1 vs 45+1
  with a mulligan diff of -2
  ftp with 1 lands wins 0/100000 games
  draw player with 1 lands wins 100000/100000 games
  ftp losses to combat: 100000
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of -1
  ftp with 1 lands wins 25246/100000 games
  draw player with 1 lands wins 74754/100000 games
  ftp losses to combat: 74754
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 25246
  ----------------------
  with a mulligan diff of 0
  ftp with 1 lands wins 84621/100000 games
  draw player with 1 lands wins 15379/100000 games
  ftp losses to combat: 15379
  ftp losses to deck: 0
  draw losses to combat: 0
  draw losses to deck: 84621
  ----------------------
  with a mulligan diff of 1
  ftp with 1 lands wins 9944/100000 games
  draw player with 1 lands wins 90056/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 90056
  draw losses to combat: 9944
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 2
  ftp with 1 lands wins 77203/100000 games
  draw player with 1 lands wins 22797/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 22797
  draw losses to combat: 77203
  draw losses to deck: 0
  ----------------------
  with a mulligan diff of 3
  ftp with 1 lands wins 100000/100000 games
  draw player with 1 lands wins 0/100000 games
  ftp losses to combat: 0
  ftp losses to deck: 0
  draw losses to combat: 100000
  draw losses to deck: 0

Here we see something a little strange happens. Aurelia notes that taking a mulligan will reduce her winrate from 84.6% to 25.2%. However, If she decides to keep, Bruna could take one mulligan and reduce her winrate even lower to 9%. Since a grind to zero would disadvantage Bruna, it's best for both players if Aurelia takes one mulligan.

As a quick aside, there are scenarios where a player has an excess of lands in their opening hand (i.e. they won't be able to put it on the bottom of the deck, and will start with a useless card). In these scenarios it may be advantageous to take additonal mulligans if doing so could decrease the probability of a loss to drawing from an empty deck. This is quite a complicated and rare case, so I will not be considering it for our calculations, but that does mean there is some additional margin of error.

Deckbuilding Part IV: The Metagame

So now we know how to play optimally given any two decks, how do we decide what deck we should be playing? Assuming optimal play, we can build a matrix to determine how well positioned a given deck is against other decks in the format. This table will show Aurelia's winrate, but we can get Bruna's winrate by taking the reciprocal.

So we can try to work out a good strategy for both Aurelia and Bruna, but (spoiler warning) it's kinda just going to amount to a complicated game of rock paper scissors...

Aurelia obviously wants to play the deck that has the best chance to win. If we sum up the winrates of each row, we find that playing fewer lands directly correlates to a higher winrate. But Bruna obviously wants to counterpick whatever she things Aurelia will play, and if Bruna is going to counterplay, Aurelia should probably also try to counterplay. If Bruna thinks Aurelia will play 0 lands, she would want to give Aurelia the worst odds of winning and play a deck with 0 or 1 lands. If Aurelia thinks Bruna is playing 0 lands, she should probably play 1 land, but if Bruna is playing 1 land, she should play 2 lands. If Bruna thinks Aurelia is playing 1 land, she should play 2 lands, but if Aurelia is playing 2 lands, she should play 0 or 3 lands... Let me just list out the counterpicks.

For Aurelia to counterpick Bruna...

• Bruna on 0 lands: Aurelia on 1
• Bruna on 1 land: Aurelia on 2
• Bruna on 2 lands: Aurelia on 3
• Bruna on 3 lands: Aurelia on 0

and for Bruna to counterpick Aurelia...

• Aurelia on 0 lands: Bruna on 0 or 1
• Aurelia on 1 land: Bruna on 2
• Aurelia on 2 lands: Bruna on 0 or 3
• Aurelia on 3 lands: Bruna on 0

I'll note here that playing more than 3 lands isn't exactly a winning strategy. Those decks are well positioned against each other sometimes, but they usually get destroyed by the smaller decks.

So unfortunately, there's really no way to solve the meta. If you're going to play to win, you should just pick 0-3 lands and hope you don't get properly counterpicked or get lucky. This is probably healthy for the format if we cared about such things.

Gameplay Part III: The Bruna Bias

With your discerning eye you may have noticed an unfortunate trend for Aurelia - the only time her winrate is >50% is the 0 lands vs 3 lands matchup. This probably means Bruna is unlikely to play 3 lands (though it does make an interesting counterpick strategy since Aurelia is probably equally unlikely to play 0 lands).

The Magic players out there probably know that this winrate bias is somewhat accounted for by the fact that matches in draft are not typically one and done matches. Most common is a best of three format where the player who chooses turn order is randomly decided for the first game, and the loser of each game choses the turn order in the next. This does create a few more decision points than you might expect.

100.4. Each player may also have a sideboard, which is a group of additional cards the player may use to modify their deck between games of a match...
...
100.4b In limited play involving individual players, all cards in a player’s card pool not included in their deck are in that player’s sideboard.

If we remember all the way back to rule 905.1d, a player's card pool also includes as many basic lands as they want, so we can change our decks going into games two and three, but more interestingly, we don't know if we're going first or second when we sit down with our deck in game one. There is still a huge bias toward the player that "wins the coinflip" and chooses to go second in the first game (which is an interesting inversion of typical Magic knowledge that going first is generally better), but the best of three format does smooth out the odds a bit closer.

Conclusion

Wow... you read that whole thing? Thank you! Memnites has been my personal mind maze for over a year now, and finally I have something to point poeople towards instead of infodumping at someone who might be causually interested / placating my interest.

Another fun discussion topic (that I certainly don't have the time or brainpower to think about) is how much more interesting the format becomes when you start adding one or few copies of other cards to the draft like fetch lands, Noxious Revival, or Overrun, or replace the titular Memnite with Wild Mongrel. I'll leave that as a thought exercise for you!

I love thinking / talking about this format, but no matter how much I gush about it, please never attempt to build or play 360 Memnites, I beg you!

If you have further thoughts, questions, or think I made any mistakes along the way, I would love to hear from you!