#2 - "A slice of pie" (A custom card design guide: the color pie)

(The banner is my own elaboration. The original image is by Holly Chaffin and is released in the public domain.)

The Lion’s Lair #2

“A slice of pie”

(A custom card design guide: the color pie)

(This article was originally published here.)



The article index is always updated with the latest content.


In this issue, I’m going to talk a bit about the color pie and keyword mechanics in general, then I’ll present the result of a work I’ve done, summarizing in two tables all printed mechanics (keyworded and non-keyworded) in the game and which colors each of them can be found in. Finally, I’ll quote an important lesson that any custom card designer should always remember.

I’ll assume everyone has a basic knowledge about the color philosophies, the color combinations, and what an evergreen mechanic is. If you are not familiar with these concepts, you can find a lot of material about them in Mark Rosewater’s column “Making Magic”.


In the MCC rubric, we can find the concept of the color pie under “Viability”. In fact, the first question in that section asks:

How well does this card fit into the color wheel?

“Color wheel” is just another way to call the color pie. I’ll always call it “color pie” as that’s its most common name.

Knowing how well a custom card fits into the color pie obviously implies a good knowledge of the color pie itself. You can’t design something that fits well in an environment that you don’t know well enough to start with. The color pie is so important that it’s included in what MaRo calls “the Golden Trifecta”, which means the three inventions by Richard Garfield that are the basis of Magic. Just remove one of those, and Magic isn’t the same game we all know and love anymore. Those are:

• The Trading Card Game genre.
• The mana system.
• The color pie.

I’m not going to talk about the TCG genre, as that’s a broader topic that I don’t have the competence to talk about. We will get to the mana system in a future issue, when we’ll talk about costing cards and effects. We’re going to talk extensively about the color pie in just a second.

Many flavors of pie

The color pie has many facets. There is a philosophical color pie, and MaRo often talks about color philosophies, but that’s not the topic I want to talk about. Of the many flavors of pie, the one we as custom card designers are the most interested about is the mechanical color pie. That is defined not only by what mechanics and effects go into which color, but also by how those are distributed among the colors that can get them. In fact, there is a concept, which I will refer to as “grade” for convention, not having been able to find an official name for it, that express this distribution of an ability among the colors. We can define the grade of a color in an ability as “how often and at which minimum rarity that color usually gets that ability”. Three possible grades exist:

Primary – A color that’s primary in an ability gets it all the time, very frequently at common (for effects that can be at common), and can very easily grant it to any other thing. Example: blue is primary in flying.
Secondary – A color that’s secondary in an ability gets it some of the time, sometimes at common (for effects that can be at common) but more often at uncommon, rare, or mythic rare, and can sometimes grant it to other things, often depending on which ability we are specifically considering. Example: red is secondary in trample.
Tertiary – A color that’s tertiary in an ability gets it very rarely, almost never at common, and cannot grant it to anything. Example: green is tertiary in haste.

More than one color may be primary in an ability (for example: red and white are both primary in first strike). The same can be said for secondary and tertiary.

An ability does not need to have all primary, secondary, and tertiary colors. It can have only primary colors (for example, black and green are both primary in deathtouch, and deathtouch has no secondary or tertiary colors). It can also have a primary color, a tertiary color, and no secondary color.


A final thing to understand before we come to the real meat of this article is the difference among the types of keyword mechanics. There are three of them, each with its own rules and formatting. Here they are, with examples:

Keyword abilities – These are what we commonly think of when we think of a keyword. These are specific words that take the place of some specific rules text, which can be reported in reminder text.

How to format keyword abilities

They are NEVER followed by a period.

Big Flier 5U
Creature – Bird (U)

Big Flier 5U
Creature – Bird (U)

They are NEVER capitalized unless they are the first word at the beginning of a line.

Haste Giver 2R
Creature – Goblin (U)
Other creatures you control have Haste.

Haste Giver 2R
Creature – Goblin (U)
Other creatures you control have haste.

If the keyword ability is made up of multiple words (for example first strike) and is at the beginning of a line, only the first of those words is capitalized.

Swift Knight 2W
Creature – Human Knight (C)
First Strike

Swift Knight 2W
Creature – Human Knight (C)
First strike

When in a list, keyword abilities are separated by commas, again without a period at the end.

Keyword Beast 3GG
Creature – Beast (U)
Hexproof; reach; trample.

Keyword Beast 3GG
Creature – Beast (U)
Hexproof, reach, trample

The order of keyword abilities in a list is supposed to be a standard order, but there are some inconsistencies in that. You can see that in cards like Akroma, Angel of Wrath, Akroma, Angel of Fury, Sphinx of the Steel Wind, and in general all cards with multiple keywords.

Keyword actions – The Comprehensive Rules define those as “specialized verbs”. From this we can deduct two things:
First, they don’t have their usual English meaning, but they have a specific meaning in Magic. That’s why they are “specialized”.
Second, they are verbs! So we must use them as such, and we must remember that they are inserted within a sentence, that as such MUST end with a period. They can even make up a sentence just by themselves, but as it’s a sentence, it needs to end with a period. That’s not Magic, that’s just English grammar! See Wake the Reflections for an example. Note that there is a period after the keyword action “Populate”, which is capitalized just because it’s at the beginning of the line, and not because it’s a keyword. In fact, in Augury Owl scry is not capitalized, because it’s not the first word of the sentence. Also, note how the word “scry” is just inserted within the flow of a sentence with a period at its end.
Also, because keyword actions are verbs, they must be correctly conjugated. See Prey Upon for an example: the verb “to fight” is conjugated at the third person (with an “s” at the end), because the subject is “Target creature you control” and not you as a player. It’s the creature that performs the fighting action, not you.

Ability words – These are just a marker, putting together cards with a similar functionality. They have no rules meaning. That’s why they are in italics, just like reminder text and flavor text, but, at the contrary of those, they are followed by a long dash, after which there is some rules text. Always remember that being in italics is synonym with having no rules meaning in Magic and vice versa. The reason why forecast is not in italics on Magic cards and, say, metalcraft is, is because the former has rules associated to it (the fact that you can use that ability only during your upkeep and only once each turn), while metalcraft does not. It just puts together cards that care about you controlling three artifacts. In fact, forecast is a keyword ability, while metalcraft is an ability word. Also, always remember that ability words cannot be referenced in rules text, at the contrary of keyword abilities and keyword actions. Again, this is because ability words have no rules meaning. For example, you can’t say “Creatures with landfall you control get +1/+1”, because landfall is an ability word and thus can’t be referenced in rules text.
Pay extra attention to imprint: in the past, it was a keyword ability, but now it has received errata to become an ability word. In fact, on cards from original Mirrodin block, the word “imprint” is printed in normal font, while on those from Scars of Mirrodin block it’s in italics. If I, as a custom card designer, want to design a custom card with imprint today, I need to put it in italics and treat it as an ability word, because that’s what it is today.

Note that devotion is normally considered a keyword but it does not fall into any of the three categories above. In fact, it’s treated separately in the Comprehensive Rules, because it’s something different. It does not substitute rules text, instruct you to perform some kind of action, or put together similar cards. It’s a numeric variable, which is a different and separate thing.

In summary, how do you recognize the different kinds of keyword mechanics? Just remember this:
A keyword ability is a noun or adjective that substitutes some rules text and does not have a period at the end.
A keyword action is a verb that is inserted within a sentence that has always a period at the end.
An ability word is always at the beginning of the line, written in italics and followed by a long dash, after which there is some rules text.

Turning tables

What I’m going to do now is to take the Comprehensive Rules (KTK Edition, as it’s the most recent one at the time of this writing) and go to “Section 7 - Additional Rules”, which is where you can find the rules for all keyword actions and keyword abilities ever printed, and to rule 207.2c, which is where you can find the list of all ability words ever printed (only the list, remember they have no rules associated to them). From there, I’ll compile a list of all existing keyword mechanics, excluding the generic ones like “tap”, “sacrifice”, or “discard”. For each of them, I will note in which colors it can be found and which colors are primary, secondary and tertiary in that ability. To determine that, I used this Blogatog post by MaRo, where applicable. For each keyword not available there, I looked at how many cards were printed with that ability in each color and at what rarities, so in this case the results are necessarily less well defined, but enough to get an idea of how that ability is distributed among the five colors anyway.

The following table is the result of this work. Each row corresponds to a keyword mechanic and each column to a color. The numeral “1” means that color is primary in that mechanic, “2” that it’s secondary, and “3” that it’s tertiary. An empty cell means that color does not have that ability.

How to read the table:
• In the “Rules” column you find the reference to the corresponding rule in the Comprehensive Rules (KTK Edition).
• In the “Type” column, “A” means “Keyword action”, “B” means “Keyword ability”, “W” means “Ability word”, and “O” means “Other” (applicable only to devotion).
• In the column identified with a star symbol (“*”), I’ve reported if the mechanic is evergreen (“E”), a block mechanic (“B”, including keywords that appear on regular cards in special sets like Conspiracy), or obsolete (“O”: shroud, which has turned into hexproof; fear, which has turned into intimidate; and banding, which well... is banding).
• The following six columns refer to the five colors in WUBRG order and artifacts (“A”).
• The only exception is annihilator, which can be found only on colorless non-artifact cards, and thus appears with all empty cells in this table.
• The table is ordered listing evergreen mechanics first, then block mechanics, and lastly obsolete mechanics. Within each of these categories, alphabetical order is applied.
• The table is divided into consequential images because of space issues and put into spoiler tags because it’s very long.


We can do the same work for more generic concepts, like “counterspells”, “mana ramp”, or “rituals”. This last word in particular gives me the opportunity to make an important remark: here I’m considering only modern Magic design. Abilities that were once in a color but not anymore now, won’t appear in the table under the former color. For example, “rituals” like Dark Ritual are no longer black. Now they are red, so I will not consider that ability in black but only in red. I tried to remember as many of these more generic kind of spells and effects as I could. I may have missed some of them, in which case please let me know and I’ll add them. “N” stands for “not a keyword”.

Enjoy again!

Weakness is the greatest strength

Up until here we’ve had fun querying databases and tabling out the results. But all this work is not just a pastime, as it reflects a very important concept in MTG design that I would like to be the theoretical lesson from today’s article for everyone that wants to get better as a (custom) card designer: the importance for each color to have weaknesses.

Innovation is a very important and potent tool at our disposal, and in fact it has its own place in the MCC rubric. An innovative card captures the attention of the audience because it does something that has never been done before. This plays into the “surprise” factor of the “comfort, surprise, completion” mantra that MaRo often repeats, and that are the three things that each of us psychologically needs in the games we play, in each activity we do, and I dare to say in our whole life.

But beware! Not all innovation is necessarily good! There is good innovation and bad innovation. For example, if your original and innovative card is such because it undermines or completely undoes what is meant to be a weakness of its color(s), that’s bad! You shouldn’t do that! In the color pie, each color needs weaknesses. If any color can do anything well and efficiently, you have no reason to expand into other colors. In a world where each color can do everything, only monocolored decks would be played, and those decks would all do the same things at the same power level, so each matchup would be exactly 50/50. But most importantly, the difference between those decks would be only nominal. They would all really be the same deck. You could say that only one deck would be played and that all matches would be mirror matches. Does that sound as an exciting world to you? I don’t think so. It would quickly get boring, there would be no metagame and no deck evolution and innovation. Patrick Chapin wouldn’t be called “the Innovator” in that world, because there would be nothing to innovate. The best deck would already be at its best. It’s a static world. No change whatsoever, no dynamics. A lot of comfort would be there, but no surprise. A lot of boredom would also be there. If all games played the same, would anyone be playing anymore after a few games? In principle, maybe some Spikes who don’t like variance at all, and ironically I tend to be in that range as a player. I hate variance, but I understand the need for it in the game. Anyway, even those Spikes wouldn’t have fun: how can you prove something about yourself when each game you play is already scripted to be a coin flip? In a short time, there would be no players anymore, and that is the death of the game.

Do you want to kill the game? Kill variance! Well, you could actually be interested in this if you’re one of MaRo’s clones! And how can you kill variance? Make every color do everything! That’s why colors need to have weaknesses. That’s why each color has at least some empty cells in the tables above. In fact, the weaknesses of each color are what defines the color pie. For those that don’t know, and you need to know this to be a good (custom) card designer, these are meant to be each color’s main weaknesses (with examples of things getting close to break them):

WHITE: white has the worst card advantage of all colors, to balance the fact that it has the most answers. White has a very hard time drawing cards beyond the one for the turn. In fact, white cards that can give you card advantage (so I’m not counting cantrips, as that’s card parity and white can do that just as any other color) are very powerful, even if they are very conditional. Mentor of the Meek is a powerhouse in a weenie deck. Trust me, I’ve tried that in Duels of the Planeswalkers and I can’t forget how good it was!

BLUE: blue is supposed to have an hard time permanently dealing with threats that have already entered the battlefield. If it fails to counter them, blue has to bounce them (and they will get cast again), put them into the library (either on top, on the bottom, or shuffled into the library, from where they will get drawn again sooner or later), or transform them with spells like Polymorph (leaving behind another threat, usually weaker but that you still have to deal with anyway). Some people (myself included) don’t like this last kind of spells, right because they can be seen as too close to hard removal for blue. The point is that if you have an annoying creature, for example, say, a Dragon, and I Pongify it, I still have to deal with the token, but in the meantime I got rid of the Dragon for good! Even if I deal with the token, it won’t ever return! And on top of that, the token is a much easier threat to deal with, and blue has the means to do that (just think of bounce). This discussion has been going on just a few days ago, for example, when Reality Shift was spoiled.

BLACK: black is the best in destroying creatures, and lately planeswalkers too (Hero's Downfall), but has difficulties dealing with all other card types. In the very early years of Magic, there have been some black cards that destroyed artifacts, for example, but they are considered a mistake now, and anyway you can’t consider anything that was printed so long ago as a precedent for things regarding the color pie as it is today. Also, black’s card draw can’t be paid only with mana, there must be some additional cost not involving mana (pay some life, sacrifice some creatures, exile some cards from graveyards, etc...).

RED: red is supposed to have trouble dealing with enchantments, which is justified flavorfully because they’re not tangible, and creatures with high toughness, because its removal is damage based. This weakness is why MaRo doesn’t like Chaos Warp, a card that lets red deal with opposing enchantments when it’s not supposed to be able to do that at all. Also, while red gets a version of looting sometimes called “rummaging” (from Rummaging Goblin) where you discard first, it can’t get card advantage from such effects. At maximum it should get card parity. The closest red can get to card advantage is the impulsive draw ability, as seen on Chandra, Pyromaster for example.

GREEN: green is supposed to have trouble dealing with opposing creatures outside of combat, and even when it does it must use its own creatures (that’s where the keyword action “fight” comes from). Another very important weakness of green is that it’s not supposed to ever be able to deal direct damage unless it’s dealing it to flying creatures. That’s why MaRo (and I too) don’t like Hornet Sting. Also, green is not supposed to get flying unless very infrequently in high profile cycles, and again, that’s why MaRo (and I too) don’t like Hornet Queen and Hornet Nest. It looks like if hornets are involved, something bad is going on! Finally, while green is allowed to have card draw, it must be somehow tied to creatures. Remember that Harmonize is from Planar Chaos, and you should never ever use that set as a precedent for anything involving the color pie.

Remember: just like in real life, everyone has weaknesses. And sometimes, weakness is the greatest strength!

Signing out

This article was meant to give some guidelines about how the mechanics are distributed in the color pie. These are just guidelines, because, remember: Magic is a game that breaks its own rules! There are exceptions to almost anything, and almost anything is fair game if it really makes sense in the design of the card or the set as a whole. Now I hope I won’t see mechanics out of color anymore while judging in the various contests!

I’ve decided to post one article a week on Saturdays for the time being. Next time I’ll talk about the concepts of simplicity and elegance in a card.

Until next time,