One of the things that separates the Internet from other forms of media is its interconnectivity. When I read a book, there is no easy way for me to give my opinion to the author. But on the Internet, you click a button and the dialogue begins.

Several weeks ago I wrote a column called “When Cards Go Bad” wherein I explained the reason why bad cards exist. I got a huge response to my column, most of it quite positive. But there was a common thread of complaint that I received. A good respresentative of this complaint is this letter from Ted Bedwell:

Mr. Rosewater,

First, let me say that I really enjoy reading "Making Magic." It is one of the most interesting and insightful columns on the site. I've been out of town and offline and am just catching up on my reading. I was really happy to see that you posted and responded to Nathan Woodall's letter in such great length and detail.

My playgroup has often talked about this very issue. Your response to the question "Why are there bad cards?" was very clear. However, I think you failed to address Nathan's key point. Which , in my estimation, is "Why are there bad RARE cards?" Many of us join Nathan in feeling robbed when we dish out our money for a booster and the one rare we get sucks. I understand and agree that there can only be so many good cards, but shouldn't they be clustered around the rares?

My compatriots are all convinced it is simply the capitalist system at work. The more crappy rares you print, the more boosters folks have to buy to fill the demand for the good rares. Now that I am gainfully employed, I'm finding that buying singles on the Internet is becoming far more cost effective than buying boosters and starters. I wish this was not the case, but I think crappy rares are the main reason why I'd rather pay 5-10 bucks for the one I need as opposed to paying Wizards $3 for the chance to get nothing of any value at all.

Just thought I'd give you a bit of feedback. I'm sure you're already sick of hearing from all of us on this.

Thanks for spending the time to read my mail. Keep up the great work!

Ted Bedwell
Columbia, MD, USA

My response:

Ted,

Thank you for your feedback. I’m glad that you feel strong enough about this issue to take the time to write to me. In tackling the topic of “bad” cards, I glossed over the rare card issue that Nathan and you both brought up. In this column, I was hoping to go into greater detail on how cards end up rare. As with the “bad” card column, this one will be slightly longer than usual. If you aren’t interested in slogging through my lengthy explanation, feel free to jump ahead here to see my argument in short "sound bite" form.

One of the hardest aspects of this topic for R&D is that the public asks for two things that are very much at odds. First, they want more “good” rares. That is obviously the issue that you wrote in about so I’ll be touching upon the topic in a moment. And second, they want us to stop making rare cards that don’t feel rare to them. Examples players often use for this category are Invasion’s Absorb and Odyssey’s Call of the Herd. Accomplishing both these contradictory tasks simultaneously might be impossible.


Players want "no bad rares," but they don't want "all the good cards to be rare." There's no easy compromise.

To understand the reason for “bad” rares, we have to take a look at how rares are made. So how does a card become a rare? (I feel an odd compulsion to explain this in a "Schoolhouse Rock" video.) In general, cards are not designed with rarity in mind. The designer creates the card and it is not until he starts compiling the set that the best rarity for each card is figured out. The one exception to that rule is that designers are often asked to fill in holes in either design or development. When filling a hole, the designer has to create a card to fill the appropriate rarity.

Imagine you’re a designer and you’ve designed a bunch of cards for a set you’re working on. It comes time to compile the first draft of your set. You begin by assigning rarities. What qualities do you look for when deciding what cards are rare?

1) Complexity

One of the great balancing acts of Magic is keeping a wide spectrum of players happy. On one end we have the advanced player; they take great enjoyment in delving into the complexity of the game. On the other hand, we have the novice players; they are just dipping their toe into the world that is Magic. How do we keep the first group happy without scaring away the second?

Both groups are very important. The advanced players spend the most money on the game and participate heavily in organized play events. The novice players are the game’s future. Everyone starts the game as a novice. To ignore them is to doom Magic’s long term health.

Luckily, the solution to this problem rests in their buying habits. There is a high correlation between the level of play skill and the number of cards purchased. By making complicated cards rare, we give advanced players access to them but limit their exposure to the less experienced player.

Complex cards can be “good” or “bad” (I put these terms in quotes because, as I explained in my “bad” cards column, a card’s quality is very subjective) . Since all complicated cards are rare, that means any “bad” complicated card by definition must be rare. An example of a complicated card rated low in recent surveys (and thus “bad” to some subset of players) would be Odyssey’s Nefarious Lich. Nefarious Lich is simply too complicated for R&D to put anywhere other than rare.

As a final note for this category, I feel a need to point out that while complex cards need to be rare that does not mean that rare cards need to be complex. There is this misconception that simple cards shouldn’t be rare. One only needs to look at basic set staples like Birds of Paradise, Wrath of God, or Lord of Atlantis (all cards from Alpha, incidentally) to see the place simple cards fill in the rare slot. As you read through the following categories be aware that each of them requires, in different ways, simple cards to be printed as rares.

2) Rules Complication

This category works similarly to the last one. We want advanced players to get their hands on cards with complex rules interactions while keeping novices from them as long as possible. While complexity and rules complication can overlap at times, they are two distinct qualities.

Sometimes cards that are simple in concept prove very complex in rules interaction. Good examples are Tempest’s Humility and Urza’s Destiny’s Opalescence. Other cards that are difficult to understand have no rules confusion. An example would be Apocalypse’s Dead Ringers. (This card is also an example of a template gone horribly wrong.)

Like complex cards, cards with rules complications can be either “good” or “bad.” If the rules interactions are complicated enough, the card needs to be rare regardless of its quality.

3) Wordiness

The default font size of an average Magic card is 9 point. But sometimes a 9 point font doesn’t fit. For these occasions, Editing allows us to shrink the font size down as small as 7.5 point. We refer to anything with a font size below 9 point as having “microtext.” One of the standing rules in R&D is that only rare cards can have mircotext. Thus, if a card is wordy enough that it requires microtext (and once again the quality of this card is irrelevant), R&D is obligated to cut some words out of it (thus allowing Editing to enlarge the font size) or make it rare.

Please be aware the wordiness and complexity, while often interconnected, are not the same thing. An example of a wordy card that isn’t too complex is Alter Reality from Torment. Because the card had reminder text for both flashback and a rules reminder (to remind you that the effect is permanent), this rather simple card had seven lines of rules text.


Huge fatties have always been rare.

4) Big Creature or Effect

R&D learned long ago that players like things big: big nasty creatures (like Mirage’s Phyrexian Dreadnought) and big powerful spells (like Invasion’s Obliterate). Since these don’t seem like the kind of thing you should be running into all that often, we make these cards rare to maintain their specialness.

This is one of the categories that often puts simple cards into the rare slot. Large creatures or big wide sweeping effects need not be complicated. Examples would be Torment’s Angel of Retribution and Odyssey’s Traumatize. A lot of big creatures and spells are aimed at a section of audience that R&D nicknames "Timmy" (expect a longer explanation of R&D’s player types in a future column). Since Timmy doesn’t care as much about the overall power level of the card (the average Timmy doesn’t play in tournaments), R&D doesn’t worry about making all of these cards tournament level. Tournament players thus might see these as “bad” rares, but there exists an entire subset of players that get excited by them.

5) Splashy

Size isn’t everything. Some very cool creatures (such as Planeshift’s Doomsday Specter) aren’t big. And some spells, while limited in effect, are very unique (such as Odyssey's Tainted Pact). R&D believes these smaller cards with special abilities and effects deserve rare slots.

I am often asked, for instance, why we chose to make Planeshift’s Terminate common while making Apocalypse’s Vindicate rare. The answer is that “destroy target permanent” is a special ability. In eight years and over 6000 cards, we’ve only made two cards with that ability (the other being Desert Twister, at twice the mana cost). “Destroy target creature” by comparison is a common occurrence that shows up multiple times in every expansion. Some of you, I’m sure, will write in to me to point out that Desert Twister was uncommon. While that’s true, I feel a need to point out that one card is seldom a precedent.

Just as with the big creatures and spells, this category creates cards that are very subjective. While some players are excited by a card that does something never before done, others dub it “bad” if it doesn’t see any play in tournament-level decks.

6) Narrow

One of the trademarks of Magic design is the narrow card. These are cards that very efficiently do a particularly specific function. If a deck exists that wants this effect, the card will see play. If no deck finds a use, it sits in the binder. Narrow cards sit on the fence between being “good” and “bad.” Donate (from Urza’s Destiny), as an example, was pushed out of numerous designs by the development team for being “too narrow.” As it is now the key card in one of the best decks inExtended, it has obviously found a niche.

Narrow cards are designed primarily for constructed. As such, they most often have no role in limited. This being the case, R&D chooses to put narrow constructed cards that have no limited application into the rare slot. Other classic examples of cards like this are Ice Age’s Despotic Scepter and Mirage’s Lion's Eye Diamond that went on to find a perfect fit in a powerful deck.

7) Disruptive to Limited

One of the challenges of designing Magic is that we are actually designing cards for different formats -- each with different power levels -- all at the same time. The power level of sealed, for example, is below that of draft, which is in turn lower thanBlockConstructed which is lower thenStandardwhich is lower thanExtendedwhich is lower thanType 1.

The biggest problem is this: Limited formats are at a significantly lower power level than constructed formats. The power level is so distinct between the two that R&D actually has to design cards specifically for each group (which isn’t to say that we can’t make some that overlap both). This, of course, causes a number of problems. The largest of which is, how does R&D keep the more powerful constructed cards from being ruining limited?

The answer to the problem is to make troublesome cards rare to minimize their effect on limited. Be aware that these are not just tournament-level cards but also cards with weird effects that are interesting in “fun” constructed but overpowered in limited. The reason this category may make “bad” rares is that there exist cards overpowered in limited that might not be tournament worthy. In general these are the “fun” constructed cards I mentioned above. Examples would be Invasion’s Darigaaz, the Igniter or Odyssey’s Aboshan, Cephalid Emperor.

8) No Room in Uncommon

Now we get to the categories that are a little less obvious to the outside world. As I said earlier, cards tend be designed first and assigned a rarity later. As a rule of thumb, cards fall into one of the following categories:

  • Must be common
  • Could be common or uncommon
  • Must be uncommon
  • Could be uncommon or rare
  • Must be rare
  • Could be common, uncommon or rare (this almost shouldn’t be a category, but it happens once in a blue moon)

Often during design or development, one rarity gets overloaded. When this happens, R&D adjusts by forcing some of the variable cards up or down in rarity. A common occurrence is that we run out of space in uncommon (remember this is the place to put slightly more complicated cards that we want to see played in limited) and have to force cards to common and rare. Thus, some cards that potentially could be uncommon (and might be “bad”) end up as rares due to lack of space.

9) Part of a Cycle

A cycle is a group of cards that are connected mechanically and often thematically. The most common use of cycles is to create five cards one in each color. A good example would be the burst cycle from Odyssey (Life Burst, Aether Burst, Mind Burst, Flame Burst, Muscle Burst). One of R&D’s rules is that cards of a cycle must all be of the same rarity. While we break this rule on occasion (much as we break any rule on occasion), we only do so when there is a special reason and even then very sporadically.

A good example of a card from this category is Shadowmage Infiltrator from Odyssey. Odyssey has eleven multi-colored cards. There are two cycles (one of each allied color combination), one at uncommon and one at rare, and one five-colored card (Atogatog). The uncommon cycle was an Atog cycle (Phantatog, Psychatog, Sarcatog, Lithatog, Thaumatog). Shadowmage Infiltrator, not an atog, had no choice but to be in the rare cycle, even if it "feels" like an uncommon.

Not every card in a cycle is going to be as good as the others. So, for every Oath of Druids (from Exodus) in a cycle, you will occasionally also get an Oath of Mages.

10) Spread Good Cards Among All Rarities


If these cards were rare, there would be more "good rares"... and fewer happy players.

This category exists because of a key belief of R&D: “good” cards should be spread among all rarities. If all “good” cards were common, Wizards would be unable to make the kind of profit necessary to keep printing the game. But if all the “good” cards were rare, we would make a huge barrier to entry for newer players. R&D decided to make a compromise. Have some “good” cards in common and uncommon to make the game accessible and have some “good” cards in rare to reward the players who purchase more cards.

One of the points that you bring up in your letter (a point I should stress which I hear often) is that you feel R&D should allocate more of the good cards as rares. This, you feel, would make buying a booster pack more economical. I’ll be honest… This point confuses me. For the sake of example, let's assume that every common card is worth 25¢, every uncommon is worth 50¢, and every rare is worth $1 on the secondary market. I understand that this is a gross oversimplification, but I think it will illustrate my point. Let’s assume Set X has sixty “good” cards. Once again, I am just pulling a number out of the air.

Under R&D’s current system, the cards are spread amongst the rarities. Again for the sake of simplification let’s assume that it breaks 20/20/20 (common/uncommon/rare). I imagine that you want a few “good” common and uncommon cards, so I’ll assume you want a model like 10/10/40. If you wanted one copy of each “good” card, R&D’s model would cost you $35. Your model, on the other hand, would cost $48. Assuming you wanted four of each card to maximize deck-building options, the R&D model would cost $140 while your model would cost $192. Be aware that my example doesn’t spike up the cost of the top-end cards as happens in the real world. In short, R&D’s current system makes getting the “good” cards more economical, not less.

A Final Note

As you can see, there are many issues that go into choosing which cards are rare. As I explained in my “bad” card article, there is a finite number of “good” cards that R&D can create without pushing up the overall power level of the game. While R&D could move up “good” common and uncommons to rare, this will only make the “good” cards more expensive to acquire. And this will further upset the players that dislike having non-rare-like cards as rares.

This is a very complex topic with no easy answers. The problems you’ve voiced exist not because R&D is unaware of them but because there is no simple way to correct them. In short, “bad” rares exist because spreading “good” cards across all the rarities requires us to spread “bad” cards across the rarities as well. In addition, certain types of cards (whether “good” or “bad”) have certain qualities that force them to be rare.

I know this isn’t the cleanest of answers but then this is one of the messiest of issues. I guess the short answer is we’re doing the best we can. I hope this column will give you a better understanding of why.

A Recap

In short, cards become rare for the following reasons:

  1. They’re too complex to be common or uncommon
  2. They have rules complications we don’t wish to put in common or uncommon
  3. They’re too wordy and require microtext (a smaller font) which requires them to be rare
  4. They’re big creatures or big spells that need to be rare to keep their specialness
  5. They’re cool, unique creatures or spells that need to be rare to keep their specialness
  6. They’re narrow cards created for constructed (and not limited)
  7. They’re cards that prove disruptive to sealed or draft and are made rare to minimize their appearance in limited formats
  8. They’re cards that could be uncommon or rare but there’s no room left in uncommon
  9. They’re part of a rare cycle
  10. We need to make the card rare to keep a balance of “good” cards throughout the three rarities

As with my “bad” card article, I am interested to hear all of your feedback. If you want to voice your opinion, click here. I will be reading the thread and I will try to jump in when I have something new to add or if there are questions people want answered.

Join me next week when I explore an expansion of legendary reknown.

Until then, may you not draw you expensive spell until you have the mana to play it.

Mark Rosewater

Mark may be reached at makingmagic@wizards.com.