The Underworld Cookbook: Reinventing Magic

Months ago, in the GDS2 essay section, applicants were asked to consider an aspect of Magic that could be removed. They wanted to simplify the rules without harming play. There were many responses. Some people talked about removing the maximum hand size. Some discussed not giving players priority during the draw step. The list goes on. That question is interesting but for me it opens up an even larger one.

If Magic were reinvented today, with the full benefit of hindsight, how would it be different?

I have a few thoughts on the matter. I'm sure you'll agree with some and disagree with others; most of all I just want to share my thoughts with you Melvins out there. I should pause right here and warn you: I'm not a professional game designer. I'm not a judge. I'm not even sure I can tell you how the layer system works. What I am is a physicist. I have an eye for elegance and symmetry. Despite having played the game for over fifteen years, I rooted hard for the M10 rules changes from the start; they emphasized grokkability and flavor resonance over old habits and rules loopholes.

I also want to point out that I'm thinking about hypothetical changes that could have been implemented when the game was created, not changes I would want to see implemented right now. The crippling weight of inertia is something that Wizards must consider but I will not; I am going to feel free to talk about changes that would be impossible to implement due to the scope of errata necessary. It's also possible that the differences I discuss would make certain cards unprintable (either on the basis of strength or just because they wouldn't work anymore). But remember, if these were the rules from day one, every card would have been designed to work with them.


Back when I started playing, there were a few more card types than there are now. Namely, we had interrupts and mana sources which were folded into instants over a decade ago with the introduction of the stack.

I can't explain to you exactly how interrupts worked. I know that they were "faster" than instants and that they were the inspiration for the split second mechanic in Time Spiral. They somehow carried the rules baggage necessary for counterspells to work in an age where spells and abilities resolved in batches rather than one at a time. Mana sources are even worse. Starting in Mirage, some interrupts became mana sources (just scroll through the different printings of Dark Ritual). When Sixth Edition came out, they were folded in to instants as well.

My point in all of this is to that Magic doesn't want to have different types for cards that do the same thing.

Artifacts and enchantments basically do the same thing. They can have color or be colorless. They can be creatures. Mechanically, they're identical; they're spells which are cast at sorcery speed and which sit on the table after they resolve with no additional rules baggage (like creatures and planeswalkers have). However, we would never fold them into each other because they are flavorfully distinct. An artifact is a physical construction while an enchantment can be an abstract concept or some embodiment of pure mana. When they are attached to a creature, an artifact is something the creature picks up or puts on while an enchantment changes some characteristic of the creature itself.

I would argue that we can't make the same distinction between instants and sorceries. They're mechanically the same; they have an effect on resolution and then are put into the graveyard. They're flavorfully the same as well; the difference between a sorcery and an instant is the difference between making a creature really big and making a creature really big really fast. We can also see examples in researching, shattering, killing and even charging. Instants are just sorceries that are less restricted. But we have a word for that which we've been using for the better part of five years: flash!

I believe that if Magic were recreated right now, instants would not exist. We'd have either flash sorceries (flash being a supertype) or just sorceries with flash. In terms of symmetry, rules cleanliness, and grokkability it would be fantastic. Spells all would have the same timing restrictions. The rules would support a fast creature or enchantment in the exact same way as a fast sorcery. Plus I think flash sorcery really rolls off the tongue.

The rules baggage is minimal. Flash means "(You may cast this spell any time you could activate an ability)" while slow abilities like those on Dimir Guildmage would just say "Activate this ability only any time you could cast a spell". Mystical Teachings would search your library for a card with flash. In fact, because the game can distinguish between a sorcery with flash and one without, I'm pretty sure only one card would even be functionally changed by this... and who plays that card anyway?

MaRo has expounded on the similarity of instants and sorceries and discussed the subtleties between them on several occasions. One key point he makes is that, if a card works as a sorcery, it should be a sorcery. Being an instant is an added ability, which is why we ought to use an ability word to describe it.

Characteristic-Defining Abilities

Right off the bat you might be asking me what a characteristic-defining ability is. From the comprehensive rules,

A static ability is a characteristic-defining ability if it meets the following criteria: (1) It defines an object's colors, subtypes, power, or toughness; (2) it is printed on the card it affects, it was granted to the token it affects by the effect that created the token, or it was acquired by the object it affects as the result of a copy effect or text-changing effect; (3) it does not directly affect the characteristics of any other objects; (4) it is not an ability that an object grants to itself; and (5) it does not set the values of such characteristics only if certain conditions are met.

For example, changeling is a characteristic-defining ability. The pacts have characteristic-defining abilities to give them color. And any time you see an asterisk for a power or toughness, there's a corresponding characteristic-defining ability.

I feel that characteristic defining abilities are inelegant for several reasons. You have to know that they function in any zone (most of them don't specify that, but it is true). They add some complexity to the layer system, which is complicated enough already. It feels wrong to have a non-number in the power/toughness box. And finally, as far as I can tell, this class of ability is not necessary.

For abilities that need to function when the card is not on the battlefield, you can just specify that. Make Transguild Courier's reminder text into rules text and you've got it. This change only takes half a dozen words, and doesn't come up very often.

For creatures, just replace all asterisks with zeroes then give them boosts instead of setting their values. For example, Tarmogoyf could just as easily be a 0/1 with "Tarmogoyf gets +1/+1 for each card type among cards in graveyards." For Aquamorph Entity, after you change the asterisks into zeroes, you don't even have to change anything else!

Creature Types

Come on, dude, let it die.
There are too many creature types. I'm just going to put that out there. If you take a peek at the most recent Mistform Ultimus Watch I think you'll agree. Having a good number of creature types adds variety to the game but beyond a point it's just silly.

Some types have to exist. Classical fantasy staples like Demons, Dragons, Elves, Giants, Goblins, Halflings, Merfolk, Ogres, Spirits, and Zombies are the bread and butter of the game. The creature type Human was controversial when introduced, but it's necessary for symmetry. The race-class model currently being used by Wizards is nice so we can include a bunch of occupations as well. Soldier, Knight, Wizard, Shaman, Rogue, Cleric, and Warrior are all fantastic. There are a few classes I feel are redundant; I would fold Barbarian and Berserker into Warrior, for example, and fold Monk and Druid into Cleric and Shaman. Furthermore, we have many types which are needlessly specific and exist only for the benefit of a few fringe cards. Graveborn? Try Zombie. Sand? Elemental. Cephalids, Clams, Nautiluses, Octopi, Slugs, and Squids all ought to be Mollusks!

In my opinion, the introduction of a new creature type should demand a compelling reason; either it should be flavorfully or mechanically relevant. Flagbearer and Eye are types which justifiably appear only on a few cards. Bringer and Volver are not. They could just as easily be, for example, Avatars.

There is something to be said for the awkwardness of tribal being a card type rather than a supertype but I haven't really got the rules credentials to discuss that in detail. I'll just say that the Supertype-Type-Subtype setup Wizards has established is very powerful and grokkable. I hope there's some good reason that tribal isn't just a supertype with the additional rule saying noncreatures with the supertype tribal may have creature subtypes; that's certainly how it reads on a card. [Ed: There is, but it's far deeper than most players can ever care about.]

Summoning Sickness

Most of the things I'm talking about would be ways to reinvent the Magic architecture while changing gameplay only in corner cases. On the other hand, here I'm going to talk about something which would pretty drastically change the game. The entire pricing scale of creatures would have been developed differently, sorcery-speed removal would get worse, and Control Magic effects would be better.

I'm talking about summoning sickness. If Magic were reinvented today, I'm not convinced it would exist.

Think about it. It's ingrained in you that creatures can't attack on the first turn, but why is this? From a flavor perspective, what's the difference between having summoning sickness and being tapped? Either one could be taken to mean that the creature is busy, distracted, asleep, etc. Why do we need mechanical redundancy? We have the same problem on the fast end. Haste usually means something is fast or angry, but first strike, flash, and vigilance are all used to mean essentially the same thing.

And that's assuming that there is a clear in-world flavor to haste. Sometimes it just means that Wizards wanted you to be able to use the ability right away!

Without summoning sickness, the rules are cleaner. As long as a creature is untapped, it's allowed to attack. Explaining Magic to a new player just got one sentence easier. And tons of cards no longer need to grant haste to make sense, temporary resurrection and temporary stealing come to mind immediately.

And you know what? If there's a creature that shouldn't be allowed to attack right away, just make it come into play tapped! You could even go one step further and come up with a keyword; call it flinch or something: "(Tap this permanent as it enters the battlefield or changes controllers.)"

The elimination of summoning sickness is a drastic change. Trying to implement it now that the game is established would be unthinkable. But given the chance to do so, I would argue that it should not have been included in the first place.

Instruction Formatting

Have you ever noticed how helpful red is? When you cast a kill spell, it gives you an instruction. "Kill that guy." Counterspells say "get rid of that nonsense." Even green puts you to work: "Search your deck." Not red. Lightning Bolt says, "Let me deal with that for you. I'll do that damage. I wouldn't want you to stress yourself."

It's the case across the entire game; most spells give you instructions, while burn spells instruct you that they will be dealing some damage on your behalf. It's inconsistent.

You might think it would be nice if all spells were so helpful; "Doom Blade destroys target nonblack creature" sounds all right. "Counterspell counters target spell" is a little awkward, but it does make it inescapably clear what's going on (for example, that it's a blue card countering your spell, not a player). But it's all downhill from there. How would you feel if Divination said, "Divination instructs you to draw two cards"?

I don't see why we can't just have Lightning Bolt say, "Deal 3 damage to target creature or player."

Spell Fizzling

In general, when you cast a spell or activate an ability you expect it to have the effect printed on the card. But sometimes someone throws a wrench into the works by sacrificing a creature, giving something shroud, or even eliminating a player. Then you have to know two things.

On resolution, a spell or ability will do as much as it can. If there's a part of the effect that cannot be carried out, skip it.

If a spell or ability has targets and they are all illegal at resolution, the spell or ability is countered.

From a strictly mechanical perspective, the first rule suffices. The second rule just adds complexity without enriching gameplay. If Runeclaw Bear is sacrificed in response to Doom Blade, it really doesn't make a difference whether the spell is countered or just does everything it can (which is nothing). If your opponent has no cards in hand, you can still draw a card off of Brainbite. But we have an entire rule stating that if they sacrifice their artifact before Smash resolves, you don't do as much as you can. The whole spell fizzles.

There are some cases where having a spell fizzle makes flavor sense. For example, look at Sorin's Thirst. "You're drinking the blood," you may say "If there's no blood, how are you gaining life, smartypants?" Well, if that's really so important to you, we can use three magic words: if you do.

Deal 2 damage to target creature. If you do, gain 2 life.

This has the added benefit of preventing the life gain if the damage is prevented, for example by Master Healer. Boom, next topic.

Phase Order

This is something very familiar to experienced players, but undeniably complicated. There are phases and steps (twelve in total!) all of which go in a particular order... and most of which are glossed over in everyday play. I think that if Magic were reinvented today the structure of the turn would be dramatically simplified.

Wizards in on record saying that the upkeep is awkward and that the maximum hand size rule is under consideration for removal. I think we can probably all agree that it's awkward that the end step is not the last step of the turn. And cleanup... don't even get me started on that nightmare.

I won't claim that this is perfect, but here's a possible turn structure.

Beginning of Turn
Clean up damage and until end of turn effects from the previous turn.
Untap all permanents you control.
Draw a card.
If applicable, triggers go on the stack.
Abilities can be activated.

First Main Phase

You can even still Condemn a creature after it's dealt damage!
If applicable, triggers go on the stack.
Land may be played, spells may be cast, and abilities may be activated.

Combat: Declare Attackers
Declare attackers.
If applicable, triggers go on the stack.
Abilities may be activated.

Combat: Declare Blockers
Declare blockers.
If applicable, triggers go on the stack.
Abilities may be activated.

Combat: Deal Damage
Deal damage.
If applicable, triggers go on the stack.
Abilities may be activated.

Second Main Phase
If applicable, triggers go on the stack.
Land may be played, spells may be cast, and abilities may be activated.

End of Turn
If applicable, triggers go on the stack.
Abilities may be activated.

There are a few functional changes here. One big one is that upkeep triggers (now beginning of your turn triggers) happen after you draw. Dealing with upkeep triggers before drawing adds good tension to the game, but also adds awkwardness and rules overhead. Additionally, many players instinctively draw after untapping (I know you don't, but new players are valuable too). Being forced to sacrifice something as a result of that reflex is very frustrating.

You'll also notice that damage wears off at the beginning of your opponent's turn, not the end of yours. This doesn't change much; without a maximum hand size cleanup applies symmetrically to all players. However, in rare cases there will be a difference in how this plays since here nothing is allowed to trigger during cleanup. For example, under the old rules, if you use a Giant Growth to save your Solemn Simulacrum from your opponent's Black Sun's Zenith, it will die, trigger, and give you a tapped land during your opponent's cleanup step. Then they'll have another cleanup step. Then it'll be your turn and the land will untap. Here, he's too late. The death trigger doesn't go on the stack until after you've untapped. But if that's the cost to get rid of a step that sometimes shows up more than once I'm totally comfortable with it. And in any case, this simplifies the rules architecture in a way that will almost never affect gameplay.

The Beginning of Combat and End of Combat steps have been eliminated. They don't actually do anything. If you're going to tap down my team, you can at least have the decency to do it during my main phase. This makes it a little harder to tap down creatures with haste, which I like. They're supposed to be a pain in the butt for control players.

Closing Thoughts

Perhaps this will draw complaints that the differences I'm talking about would create a dumbed down game; these are the same sorts of complaints that were widespread when damage was removed from the stack in M10.

First of all, that didn't dumb down the game. It actually made the game more skill intensive by removing what was typically an obvious play and leaving players with a choice. On the other hand, damage on the stack did make it easier to make blowout plays against new players who didn't understand the rules fully.

That's not the point I'm trying to make here. What I'm saying is that Magic is a game with a lot of depth and complexity. I think if it were reinvented today with the knowledge of how long it would go on, the amount of work that would be put into new sets, and the sort of mechanics that would be explored, the basic rules architecture would be simpler.

Look at Go. It's got arguably the simplest rules of any game, but is outrageously complex due to the number of options a player has during play. It's also widely regarded as one of the best games ever created. That's what Magic wants to be. The massive number of choices confronting a player during deck construction and play provide plenty of complexity; were Magic reinvented today with that knowledge, I think the rules architecture would be simpler to create a more elegant - but still very complex - game.


Posts Quoted:
Clear All Quotes