This week we are starting what we plan to be a yearly tradition. The last week of each year we will present "Best Of" Week where we reprint what we believe are the best items of the year. This will include everything from columns to Magic Arcana to Ask Wizards.

For the columns, Aaron asked each columnist to pick their favorite column. For “Making Magic” I chose “When Cards Go Bad,” my explanation of why R&D creates “bad” cards. I chose this column for two reasons. First, it generated the most reaction and two, its actually my favorite piece I did in 2002. Without any further ado, here it is.

—Mark

This article first appeared on January 28, 2002.

Inthe first month of MagicTheGathering.com, I received numerous emails from many of you readers. Most of them said quite nice things while a few others were filled with, well, let’s just call it constructive criticism. I wanted to start this week’s column by stressing how much I appreciate all your mail, even the ones that call me names. One of the most important aspects of MagicTheGathering.com was to give all of you a chance to speak with the people who make Magic. As one of those people, I love hearing all your feedback.

I do feel obligated to stress that while my busy schedule (and large inbox) keeps me from replying to every post, I do in fact read all the email sent to me. So please, if you would like to have my ear, feel free to write to me at makingmagic@wizards.com. Every Magic player can have a voice. But I cannot hear what you do not say.

This week, I thought I would address one of the letters sent to me. The letter was a response to what has come to be my most controversial comment thus far on MagictheGathering.com. The letter was entitled “You Have Offended Me, Sir”, and it is printed below in its entirety:

Dear Mr. Rosewater,

Earlier, there was a question you answered on MagicTheGathering.com that intrigued me. I'll quote the article first before I go on:

January 4, 2002
Q: "Why does R&D print ridiculously bad cards in sets, particularly as rares?" --Elliot Fertik, Philadelphia, PA

A: From Mark Rosewater, Magic senior designer:
"This is a very complex question that I'm sure I'll discuss in greater detail in a future column. But the short answer is that weak cards are a fundamental part of the game. Richard Garfield has described Magic as a 'game of exploration.' Much of the fun of the game comes from players examining each new set to see what they can discover. Many players take great enjoyment in finding use of cards that others dismiss. R&D cannot make bad cards that are secretly good without also making bad cards that are actually bad.

"The history of Magic is filled with 'sucky' cards that later show up in high-profile decks (High Tide, Despotic Scepter, Lion's Eye Diamond, etc.). The reason that a large percentage of these cards are rare is that we tend to avoid making cards with narrow functions common or uncommon because they have no role in limited. One way to get a good appreciation of how cards rise in value as players find uses for them is to go back and look at a review of an old set."

You said you would discuss this in greater detail in a future column. I am requesting that you write this column as soon as possible, please. You see, I have played Magic: The Gathering since 1994. I have enjoyed it for the most part, and have few regrets. The regrets I do have, however, come mainly from things like opening up a pack and popping a Planar Despair, or an Okk, etc., as my rare. As a serious player, I have just wasted $3.50 of my HARD EARNED money on a packet of worthless cards. I find it personally offensive that you are trying to tell me that "weak cards are a fundamental part of the game."

I could forgive this if you were talking about cards like, for instance, Teferi's Response. That card, while weak, does have its' uses in some sideboards. I have never in all my years seen anyone play a Lion's Eye Diamond, for ANY reason (or an Okk for that matter). Can you think up a deck that could win a PTQ with the 'broken power' of Okk? I am looking forward to this article. I would, however, like to ask you for a favor. If you really think that cards like Lion's Eye Diamond, Planar Despair, etc., are fundamental to the game, would you trade me some dual-lands (Taiga, Tundra, etc.) for all of my "weak cards?" I don't mean that you should hand me three or four duals for my stack of hundreds of "weak" rares. I mean you should trade me on a one-for-one basis. Your Tundra for my Okk. Your Morphling for my Planar Despair. Your Black Lotus for my Lion's Eye Diamond. That way, I won't feel like you and your company have just called me a chump after EIGHT years of being a loyal customer, spending THOUSANDS of dollars on your product. Sure, there should be some weak cards, but at the very least, those cards could be playable.

I challenge you to respond to my letter publicly. Maybe even in your article. You have my permission to print it, in whole, or in part. You will receive my full cooperation. Heck, if you want, I'll even go one-on-one with you in a debate. That's how important this is to me. I guess you could say that's also how offended I am.

I want to make something else known though. The game is fun, and I do enjoy it. It has brought me many a smile. And, for the most part, I hold the opinion that you and the R&D staff do a good job. I have a suggestion for you on how to solve your "weak cards" problem. Get together with the top players. Grab some of the second-day players from a Pro Tour and sit them down with a cardlist of a new set and say, "What is good, what is bad, what is broken, and what is absolute junk?" I can GUARANTEE that most, if not all, of these players will not only sign the NDA (non-disclosure agreement), but will willingly give you their free time to look over a cardlist (or whatever) to help make their game better.

Thanks for your time, I look forward to your response.

Nathan Woodall
Kenner, LA
DCI#10704056

My response:

Nathan,


Mr. Woodall is not a fan of these cards

First let me applaud you for an excellently written letter. You do a very good job of expressing your feelings and I hope I can answer your letter with as much thought and thoroughness. By the way, I have a lot to say so be warned that this column will be a little longer than normal. If you want the short version just jump to the end here.

Second, I understand my comments upset you. And you were not alone. I received more mail on this topic than any other. Please be aware that my job at Wizards (as well as all of R&D) is to create a good game and make you, our consumer, happy. So, when I hear you’re upset, I want to either fix the problem or properly explain why the problem needs to exist.

“Bad” cards fall into the latter category. When I said, "Weak cards are a fundamental part of the game," what I was trying to say was that due to the nature of trading card games, it’s impossible not to have “bad” cards. They exist because they have to exist. R&D has no control over this. We never have. Alpha didn’t have “bad” cards because Richard Garfield didn’t know any better. It's just the nature of trading card games.

I apologize that it's taken eight years for us to have a forum to properly explain this to you, but its never been our intention to imply that "bad" cards are anything but a necessary evil. No matter what I say, I’m sure some many readers will walk away hearing nothing but "R&D makes bad cards on purpose." I’m spending my entire column today discussing this issue because I want you to understand the reasons behind this decision. “Bad” cards exist because they have to exist. Why? Well, here is my best attempt to explain the reasoning behind the existence of “bad” cards:

1) All The Cards Cannot Be Good

This first point is the most important. Card power is relative. Ancestral Recall, as an example, is only a good card until we create a card that allows you to draw four cards for . The thing that defines the power level of any one card is the other cards that exist with it in the same environment.

One way to look at this phenomenon is to look at Volcanic Hammer. When Volcanic Hammer was reprinted in Seventh Edition (it first appeared in Portal), many players complained. Why did Wizards put such a “bad card” in the basic set? In the current Standard environment though, Volcanic Hammer is seeing play. How can a “bad card” be good enough to play? The answer rests in the card Lightning Bolt. Lightning Bolt is strictly better than Volcanic Hammer. It has the same effect but its one mana cheaper and is an instant rather than a sorcery. When players first saw Volcanic Hammer, they compared it to Lightning Bolt and, in comparison, Volcanic Hammer seemed pretty damn sucky. But, when Lightning Bolt is removed from the picture, such as in the current Standard environment, Volcanic Hammer looks a lot better.

As an experiment, let's say we got together a collection of the top three hundred pro players and had them select the 1500 most powerful cards in Magic’s history. I chose 1500 as that is roughly the size of a full Standard environment. We then ran a Pro Tour for these three hundred players where the format was decks built using only those 1500 cards and basic land. After the tournament, we count how many of each card was used. Any card used in any deck or sideboard (even if there’s only one in the entire tournament) is counted.

Experience (as in: years of looking at outcomes of premier events like Pro Tours, Grand Prix and Nationals) tells us that only 300-400 unique cards would see play. Why? Because even among the best cards, some cards are just better than others. Mahamoti Djinn is a solid creature, but it's no Morphling. Regrowth is an excellent spell but it's not Yawgmoth's Will. In this environment, some of the “good cards” become “bad cards." The phenomenon always holds true. No matter what 1500 cards you pick, the cards will rank in a power order. When a player goes to build a deck (assuming his goal is to build the most competitive deck), he will choose cards at the top of the list before cards at the bottom.

If we can have 300-400 good cards, doesn’t that mean we can make a large expansion where all the cards see play? Yes, in theory, we could design a 330 card set where every card sees play. But what about the next set? Would anyone buy the next small expansion if none of the cards were tournament worthy? Of course not. The only way to then make the next set have tournament-worthy cards is to increase the power level. The new more powerful cards would then displace some of the cards from the first set. Unfortunately, this solution would ultimately destroy the game as the power level would keep increasing until it spun madly out of control.

R&D long ago figured out the solution to this problem. Take the 300-400 good cards and spread them out over the seven sets legal (the two blocks and the base set) in any one Standard environment. But that leaves 1100+ cards that will never see Standard play. What do we do with them?

2) Different Cards Appeal to Different Players

The solution to the aforementioned problem leads to the second reason “bad” cards exist. Different cards have different functions and appeal to different players. When last we left R&D they had 1100 cards unplayable in Standard. To make those cards matter, R&D looks to other formats. A good chunk of cards are designed to be used in Limited (both sealed and draft). Other cards are made with Block Constructed in mind. Still others are designed with older formats (Extended, Type 1, etc.).

Next, R&D looks to other types of play. We make some cards for the multi-player crowd. We make cards for the flavor crowd. We make cards for the silly crowd. We make the big creatures and spells for "Timmy." We make the combo cards for "Johnny." We take each different group of Magic players and throw some cards their way.

The problem is players tend to define “bad cards” as cards that they personally see no reason to play. But certain cards aren’t meant for them in the first place. A good example would be Goblin Game from Planeshift. This card was designed as a fun, kooky card for social players who enjoy things like Unglued. (As a quick aside I should point out that despite rumors Goblin Game was not from Unglued II.) It upset a number of serious players. Because, to them, it was a waste of cardboard.

This is also the category Okk falls in. Okk is cool because it’s a 4/4 goblin. If that quality of Okk doesn’t appeal to you then you’re not the audience for Okk – it was certainly not made to win PTQ's with.

One of Magic’s strengths is that it is many things to many people. Its adaptability allows each gamer to shift the game to his or her liking. The downside to this cool aspect of the game is that players have to realize that they will open up cards designed for other types of players.

3) Diversity of Card Powers is Key to Discovery

The next reason “bad” cards exist goes to the heart of what makes a trading card game tick. Trading card games, and Magic in particular, are very much about discovery. When you play Uno, for example, you don’t have to know that “Draw Four” is better than a blue 6. All the cards are shuffled together and you play what you get. But in Magic, you pick and choose which cards you use. That makes the ability to differentiate between cards very important. As you grow as a player, you get better at determining a card’s potential. This ongoing challenge is an important part of what keeps Magic fresh.

The best way to examine this quality is to think back to your own Magic history. Can you remember key times where you finally “got” some concept? When all of a sudden things just clicked and you realized why a card or a series of cards were better or worse than you originally thought? That is part of the thrill of playing Magic and R&D purposefully slopes the cards to allow a constant sense of discovery.

Here’s the problem: Imagine cards' relative "difficulty" as a slope. Any card below your comprehension level on the slope is either obviously playable or obviously “bad,” and the rest require some thought or game play to categorize. But we have to design Magic for all players. That means the more advanced a player you are, the more cards you label as “bad cards.” But the lower-level cards are crucial to allowing the beginning player the same sense of discovery and exploration. You may think that the “lucky charms” (Crystal Rod, Iron Star, Ivory Cup, Throne of Bone, and Wooden Sphere) are bad, but our testing shows that most beginners are drawn to them and only learn over time that they are not as good as they seem (usually because a more advanced player tells them). That is why we keep including them in the basic set.


These cards fall on an important part of the learning "slope," and consequently keep getting printed

There are two responses I expect to my reasoning, so let me preemptively answer them. First, "Magic is an advanced game. R&D’s first level is too low. Magic players are a smart crowd. They can figure it out." My answer to that is that we’ve spent a great deal of time and money to understand our player base. We make cards for the first level of difficulty because many players exist at that level. Remember that the average Magic player is 13 years old. The future health of the game rests on there being a good entry point for beginners. If new players stop joining, there will be no Magic for the advanced players to play.

Second, "Your ideas are outdated. The Internet has changed everything. Information flows freely and card powers are deduced much quicker than they used to." My answer to that is yes, the Internet has changed things. But that doesn’t alter the need for discovery. First, a lot of Magic players (in fact, a majority) don’t read about Magic on the Internet. Still others enjoy the discovery process and go out of their way not to read articles on card analysis. The discovery is a fun part of the game. Because some choose to take short-cuts does not mean that R&D should deprive others of the journey.

4) Power Levels Are Relative

One of the things R&D does to throw monkey wrenches into the above slope of discovery is to purposefully design cards that are hard to instantly analyze. A lot of these types of cards have a very narrow function that are either “good” or “bad” depending on whether there exists a deck that can use them. A good example would be a card from your letter, Lion's Eye Diamond.

At first glance, the card appears to really suck. But let’s jump in our time machine and travel back to Pro Tour - Rome in November of 1998. It was an Extended tournament at which all of Urza’s Saga was legal to play. It marked the highest power level to ever exist in Extended and possibly the highest of any Pro Tour. A good portion of the decks were able to win on turn one or two. The deck that many of the top pros thought was the best deck of the tournament was played by Brian Hacker (who due to a play mistake missed the top 8). His deck included four copies of Lion’s Eye Diamond that were key to the deck.

The point is that a card’s utility can change wildly based on the current metagame. One day a card sucks, the next it’s a fundamental card in the best deck at the highest power-level Pro Tour ever held.

5) Diversity of Power Rewards the More Skilled Player

Mostly up to now I’ve explained why “bad cards” have to exist. I also want to point out that “bad cards” have some good effects on the game. I think the best reason to have a diversity of card power is it increases the skill in the game. A lesser player, for instance, is more likely to draft a sub-optimal card. A more novice player will put questionable spells into his decks. Both occurrences increase the chances that the better player will win.

As an example, let's assume R&D made a set where every card was exactly the same power level (actually impossible, but for the sake of this argument, let’s assume we could do it). Player A and Player B are both drafting in the same eight-person Booster Draft. Player A is a professional player; Player B has only been playing for four months. In this scenario, Player A has much less of an advantage than normal. No matter what cards Player B drafts, they’re still good. Player B’s deck might not have synergy and the mana base will probably be a little off, but he’s going to be drawing nothing but solid cards.

Magic has randomness built into its design. The decks are shuffled, after all. Diversity of power helps tilt the odds back towards skill.

6) People Like Finding “Hidden Gems”

There is a second benefit of “bad cards.” One of the joys of Magic is discovering the card that everyone else has dismissed. In order to allow rogue deck builders to do this, R&D has to create some “good” cards that seem “bad." To quote myself from my “Ask Wizards” answer: (mostly because I don’t think I can say it any better) “R&D cannot make good cards that seem bad without making bad cards that are actually bad.”

A good recent example would be Pure Reflection (from Invasion). This often-dismissed card ended up as a key sideboard card in Zvi Mowshowitz’s winning PT Tokyo deck.


Henry Stern, William Jockusch, and Worth Wollpert (among others) spend countless hours trying to create balanced environments

7) R&D is Only Human

Let's say tomorrow I email you the list of Torment, Judgment, and the three sets from next year’s block. Using those sets, along with 7th Edition and Odyssey, I ask you, and any friends you want to include, to discover the metagame for next year. Each day, I’ll send you another email telling you how we’ve changed cards from the lists I just gave you. Sound hard? This is what R&D does every day.

And were not bad at it. But hey, it’s a hard job and we’re not going to be perfect. Millions of Magic players will come up with ideas that R&D and its playtesters will just miss. And the problem only compounds itself as many advances are built upon other advances. One missed powerful combination can lead us astray. Some “bad” cards are cards we thought would be better and just aren’t. But as a trade-off some cards are better than we expect, so I like to think it all averages out in the end.

An example of a recent “boo-boo” by R&D would be the cycle of Shrines (Aven Shrine, Cephalid Shrine, Cabal Shrine, Dwarven Shrine, Nantuko Shrine). R&D thought the cards would see play in multi-player decks as the cards grow in power the more graveyards they have to reference. Obviously, we were wrong.

In Summation

That in a mutant-sized nutshell is why we make “bad cards." To recap (or to fill in for those unwilling to read the long version):

  1. By definition, some bad cards have to exist. (The most important reason.)
  2. Some cards are “bad” because they aren’t meant for you.
  3. Some cards are “bad” because they’re designed for a less advanced player.
  4. Some cards are “bad” because the right deck for them doesn’t exist yet.
  5. “Bad” cards reward the more skilled player.
  6. Some players enjoy discovering good “bad” cards.
  7. Some “bad” cards are simply R&D goofing up.

While this column is just about over, I know this issue is not; I expect it will spur much debate. So I invite you to join me on our bulletin boards. I will be reading all the posts and occasionally chiming in, so please feel free to let your opinions be known.

Join me next week when I explain why there are no folk like merfolk as we have our first theme week not about a mechanic.

Until then, may your opening hand always have the right color mana.

Mark Rosewater

Mark may be reached at makingmagic@wizards.com.