The four-of limit was not part of the game originally. When Magic first began, the deck construction rules allowed you to play as many copies of any one card as you liked. If you wanted to play 20 Black Lotus, then that’s what you did.

Since this allowed complete degeneracy and Wizards wanted to push organized play, the 4-card limit was added about half a year after the release of Alpha.

But what if this limit did not exist? What would a format like Modern look like?

To make this question more specific and to simplify calculations, I’ll lay down the following rules:

• You can only use cards from Modern-legal sets.
• You have to play exactly 60 cards.
• You can play as many copies of any one card as you like.
• There are no sideboards.
• There are no mulligans.
• Matches are best-of-one.

To get some first inspiration for decks in this no-limit Modern format, we can look at decks from the real Modern. For instance, an Eldrazi deck seems appealing if you can play 20 Eldrazi Temple. And Infect would get a lot better if you can simply select a collection of Glistener Elf, Mutagenic Growth, and Pendelhaven that can win consistently on turn 2. Or how about a finely tuned combination of Puresteel Paladins, Bone Saws, Mox Opals, and Grapeshots that might win on turn 1?

But these decks would still be far too slow.

The real gauntlet starts with this.

The Chancellor Deck

Welcome to no-limit Modern! With this deck, you’re guaranteed to drain your opponent for 21 before they even have a chance to play their first land.

But this deck is not unbeatable. The triggers go on the stack at the beginning of the first upkeep, and opponents can respond with a win of their own.

The Ripple Deck

With seven Chancellor of the Dross triggers on the stack, you can respond by rippling your way to victory. To choose the ideal mix between Simian Spirit Guides and Surging Flames, you have to consider the probability of starting with at least two Simian Spirit Guides in your opening hand and the likelihood of keeping the ripple chain going.

The 31-29 mix is optimal in the sense that it maximizes the turn-zero kill probability when mulligans are disallowed. My simulations indicate that you’re 90.5% to win in the first upkeep.

But I’m just getting started! As it turns out, these degenerate combo decks have their weaknesses too.

Hatebears

Surging Flame can’t target players when Leyline of Sanctity is on the battlefield, and seven Chancellor of the Dross triggers are easily defeated by gaining 2 life in response with Nourishing Shoal. Note that Chancellor of the Dross’s ability does not target, so Leyline doesn’t work against that combo.

Other alternative hate cards are Mindbreak Trap and Providence, but Mindbreak Trap doesn’t stop a second ripple attempt, and the Providence trigger when you’re on the play doesn’t resolve until after the Chancellor triggers have already put you down to -1 life. Leyline of Sanctity and Nourishing Shoal don’t have these problems, so they are the more suitable tools to stop both combos reliably.

As the win condition, I chose Bassara Tower Archer because you can pitch it to Nourishing Shoal, because it can’t be targeted by Surging Flame (which could otherwise be an issue), and because a 2-power 2-drop fits the deck name perfectly. Style points matter too.

The reason for filling your deck with more than a handful of Archers is because the Ripple Deck can still potentially win by hard-casting Simian Spirit Guides. You need enough Archers to ensure that you don’t randomly lose to Ape beatdown. (The threat of Ape beatdown also explains why there are slightly more Nourishing Shoals than Leylines.)

The numbers in this deck list were ultimately chosen to maximize the win probability against a field of 50% Chancellor decks and 50% Ripple decks. My simulations indicate that this Hatebears list wins 90.9% of the games against Chancellor and 82.6% of the games against Ripple.

But this Hatebears deck can literally never beat the following deck.

The Memnite Deck

Since the Hatebears deck has 13 cards that trade for Memnite and 47 cards that are irrelevant, this is a literal 100% matchup. But would anyone really dare to run this Memnite deck in a format dominated by turn-0 kills? You’re bringing a bunch of 1/1s when you can just lose on your first upkeep!

Metagaming and Equilibria

Ultimately, which deck to choose depends on the metagame. I find it helpful to conceptually think in terms of levels, which often exist in regular formats as well.

• Level 0: The Chancellor deck and the Ripple deck. They are linear, focused, and inherently powerful. Relatively speaking, the Chancellor deck is more consistent but slower, whereas the Ripple deck is less consistent but faster.
• Level 1: The Hatebears deck. It has all the answers to reliably beat the decks from Level 0.
• Level 2: The Memnite deck. By not bothering with answers against the Level 0 decks, you can construct a deck that is extremely well suited to beat Level 1.

As in most formats, the metagame is cyclical: The best way to beat Level 2 is to return to Level 0. It’s a classic example of aggro (Level 2) beating control (Level 1), control (Level 1) beating combo (Level 0), and combo (Level 0) beating aggro (Level 2).

The metagame at a tournament is always a bit of a guessing game, but it is possible to find an equilibrium. To do so, you need the matchup matrix. I already mentioned several win rates for various matchups earlier, and the other ones are easily determined too. Combined with the assumption that matches are best-of-one, you get the following.

Using this, you can find the following equilibrium: If 0% of the field would play Chancellor, 38.4% of the field would play Ripple, 36.6% of the field would play Hatebears, and 25.0% of the field would play Memnite, then each non-Chancellor deck has an overall 50% win rate against this metagame.

So if these four specific deck lists are the only ones you can register and there is enough time for the metagame to settle into this equilibrium, then you shouldn’t take the Chancellor deck to a single-round tournament, but each of the other three decks would give you an equal chance to win. So as weird as it may sound, showing up with 60 Memnites would be fine.

So What’s the Best Deck?

I believe the phrase “best deck” is thrown around too loosely in the Magic community. It’s an imprecise, ambiguous term that can be interpreted in various ways. Before using it, we need to choose a definition.

Some writers have used “best deck” to describe the deck that they judge to have the most raw power in the abstract, before metagaming or knowing what other decks are out there. Hall-of-Famer Zvi Mowshowitz, for instance, used these terms when he claimed that “Affinity is the Best Deck in Standard“ in the Mirrodin block era. And R&D’s Adam Prosak claimed that “any given Standard environment has exactly one best deck,” thereby indicating that the best deck is an attribute of merely the set of Standard-legal cards (i.e., the Standard format or card pool) and nothing else.

In my interpretation, these definitions effectively assume that you can give every possible deck a distinct power grade from 0 to 10 and that the deck with the highest power grade in the format is the best. If we apply this definition to the no-limit Modern format, then I would judge that Chancellor has the most raw power in the abstract and is thus the best deck. This is subjective, but it seems reasonable given its guaranteed turn-0 kills.

While this perspective on “best deck” is helpful in identifying potentially strong strategies, it does not mean (as Zvi rightfully pointed out) that the deck with the highest power grade is also the best deck to play. That depends on how the metagame develops. Chancellor, for example, is not likely to win a tournament in the equilibrium metagame.

Another approach—one that I find more useful—is to define a deck to be a “best deck” if, out of all possible decks, it maximizes your match win probability against a given metagame. This means that we’re talking about the best deck for the weekend, or the best deck for a given tournament, or the best deck in Standard right now.

Under this definition, you need the Standard card pool (or equivalently, the set of decks) and a metagame (i.e., a probability distribution over the decks) to determine the best deck or set of best decks. Applying this to no-limit Modern, where every deck is exploitable, the answer to the question “what is the best deck?” is: “that depends on the metagame.”

Where Do We Go From Here?

Of course, the four decks I proposed are not the only decks that are feasible in this format, and the four-deck equilibrium can be attacked by other decks. Just like in real formats, new strategies would be introduced, and existing lists would be tweaked to improve against the metagame.

For instance, if no one is playing Chancellor anymore, then those Nourishing Shoals would quickly be cut from Hatebears. This, in turn, would eventually prompt a return of Chancellor decks. Just like in real formats, the metagame tends to be dynamic and cyclical.

But other tech and tweaks will probably emerge, and getting an edge in the mirror match will also become important.

Focusing on tweaks to my four decks, one of the first things that comes to mind is that the Chancellor deck could add a few Soul Spikes because a hand with 6 Chancellor of the Dross and 1 Soul Spike will be able to dish out 22 damage, allowing you to beat one Nourishing Shoal.

What’s more, it helps in the mirror match. A key rule for the Chancellor mirror is that if a bunch of triggers would go on the stack simultaneously, the ones of the non-active player will resolve first. So if both players reveal seven Chancellor of the Dross, then the non-active player (i.e., the player on the draw) will win. But if the active player (i.e., the player on the play) casts at least one Soul Spike in response, then they would survive and eventually win via decking.

The only downside is that Soul Spike loses to Leyline of Sanctity, which is otherwise useless against the deck. Again, it’s all about your metagame expectation.

The Ripple deck could add some answers to enchantments, such as Demystify or Back to Nature, so that you don’t have to scoop right away to Leyline of Sanctity. Such an addition would lead to a slight reduction in the turn-1 kill consistency, but the increased resiliency is probably worth it.

If we opt for Demystify, then the resulting addition of Sacred Foundry could also help in the mirror match. In a pure mirror of the 31 Simian Spirit Guide + 29 Surging Flames deck, the first player to blink usually loses: If you suppose that both players open with 4 Surging Flame and 3 Simian Spirit Guide (so that each player can cast only one Flame) then the first player to put Surging Flame on the stack will usually die to an opposing ripple chain in response.

But if you have Sacred Foundry in your deck, then you can eventually get to a point where you have 2 Sacred Foundry on the battlefield along with 3 Surging Flame and 4 Simian Spirit Guide in hand. This would allow you to cast three Flames in one turn and beat any opposing hand from the original Ripple deck.

The Memnite deck doesn’t need 60 Memnites per se. With 55 Memnites and 5 Leylines, for instance, you would still beat the Hatebears deck every single time, but you would gain a lot of percentage points against the Ripple deck.

And to improve in the creature mirror, you could replace all those Memnites with Mountains and Goblin Guides, which in turn would be beaten by a collection of Temple Gardens, Loam Lions, and Fleecemane Lions. Before you know it, the arms race would be on. These tweaks illustrate the general wisdom that it’s generally better to go slightly bigger (but not much bigger) than the deck you are trying to beat.

Conclusion

The no-limit Modern format is as degenerate as you might have expected, but the decision of which deck to bring to a no-limit Modern tournament is far from trivial. It made for an excellent setting to illustrate metagaming and leveling approaches.

Yet I didn’t come close to exhausting the set of competitively viable decks in this format. If you can think of a sweet deck choice for a no-limit Modern tournament, then don’t hesitate to share your ideas in the comment section below.

While the deck choice problem is intriguing, playing the actual games in this format would be absolutely horrendous. There’s basically no decision making at all! This drives me to a clear conclusion: the four-card limit was the single best rule ever added to the game. Well done, Wizards!