ConspiracyWelcometo Planeswalker Week! Yeah, yeah, I know the banner says Kithkin Week, but we all know Kithkin Week was two weeks ago, right? Okay, here's what happened. (My theory anyway.) I think I got Punk'd. My theory is that editor Kelly Digges sent out an email with Kithkin Week listed before Planeswalker Week. Then he sent out another email to everyone except me telling them that it's been changed and Kithkin Week will now come after Planeswalker Week. Wouldn't it be a hoot watching me write the wrong column during the wrong week?

I checked my email and somehow Kelly found a way to retroactively change the email to say that Planeswalker Week went first. And he got to all the writers because whenever I corner one of them they just look at me funny and say, "Uh Mark, it didn't change."

What Kelly didn't realize is that I'm kind of a conspiracy buff. (Oswald acted alone? Ha! He was employed at the book depository for weeks before Kennedy's route was changed to drive by it.) I can sense when things don't add up. Two weeks ago my computer was "acting funny" and the tech guys had to swap it. Coincidence? I think not. Kelly was flying back from Pro Tour–Valencia when my article came in and thus couldn't "catch" the mistake. Very convenient. My three-year-old son woke me up at 4:30 a.m. the day the article was due, making me extra tired. I haven't figured out yet how they looped Adam into this scheme. He seldom seems to do anything that I ask him to. But I'm on to them. I'm watching.

Anyway, as far as I'm concerned it's Planeswalker Week so I want to write the column that I've been hinting at for months. How exactly were planeswalkers designed? I'm glad you asked. It's a very interesting story and I get to talk about it right now.

Da Planeswalkers! Da Planeswalkers!

When I think back through the chain of events that led to planeswalkers I believe it all began when I selected my design team for Future Sight. You see, I knew the set was going to be a bit more off the wall than usual so I was interested in filling the team with a number of people that had never designed Magic before. The idea behind this crazy scheme was that people that didn't know the limitations of normal Magic design might not be hindered by them when I asked them to design some cards outside the normal boundaries.

As I was the design lead for Future Sight, I got to pick my team. (I try as Head Designer to allow the lead designer to pick his or her own team, with a little guidance from me, of course.) When I was looking through the list of potential designers I was intrigued by the people that showed out-of-the-box thinking. In addition, I liked the idea that the team would have designers that approached the design from very different perspectives. With all these goals in mind, one of the very first people I put on my wish list (yes, you actually have to ask people if they want to be on the team—no, we don't often get turned down) was Matt Cavotta.

When I had hired Matt I brought up the idea that perhaps one day I'd put him on a design team. He was very eager for the opportunity. So why was putting Matt on the Future Sight design team so important to the creation of planeswalkers? As I explained in my column on creativity (Connect the Dots), I believe creativity is the ability to join together things that others have not yet thought to connect. Matt was about to take an issue from the creative team and merge it with one from the design team.

The creative team issue was this: One of the purposes of the Future Sight storyline was to refocus (and, to a lesser extent, to repurpose) the role of the planeswalker. At its core, the Magic IP (a.k.a. intellectual property—basically all the creative elements that make up Magic's environment, characters, items and stories) is about two things: the color wheel and wizards fighting with magic. Yes, there are many more aspects of it, but these are the essential items that make it unique from every other fantasy IP out there.

Story_CircleAs such, the creative team has been taking great efforts to help increase the ability for stories about Magic to revolve around those two basic ideas. The color wheel is pretty easy as it is deeply interwoven into just about everything. Wizards fighting with magic, on the other hand, while always there was seldom center stage. This problem was at its worst in the game (not in a meta sense, because obviously big picture the game is two wizards fighting; I'm talking here about the individual cards) because the focus on the cards made the true standouts anything but planeswalkers. Why? Because everything but planeswalkers actually appeared on the cards. Sure planeswalkers appeared in names, flavor text and occasionally art, but they were never allowed to have their own card. As Magic, the trading card game, is fundamentally about cards, it's hard to be the focus when you're excluded from being a card.

Meanwhile in design of Future Sight, I was making the team focus on the "timeshifted" sheet. Yes, there was plenty of work to do on the nontimeshifted side of the set, but I knew that making cards "from the future"—that is, cards that hinted at where the game might be headed—was going to take more time. While I normally don't like to start design with thinking "out of the box" (as I've said many times in this column: "Don't look outside the box until you're finished looking inside"), this set forced us to move to places that were not normal for Magic design.

Make planeswalkers matter more. Figure out where the game might be headed. This metaphorical peanut butter and milk chocolate was swimming around in Matt's head until the fateful day that they collided. I remember Matt coming up to my desk. "It's so obvious," he said. "We need to make planeswalkers into cards."

Shortly before this happened, I was informed that due to a scheduling snafu I was losing a chunk of time from the design. As Future Sight was a small set (and an extra difficult one at that), this was a significant decrease percentage-wise. I was very focused on getting the design done in time for its handoff to development. As such, I assume here's what I heard: "It's so obvious," says Matt. "Let's do something that at its best would take significantly more time than a small set would have, yet do it for a design that's already extra-complicated and behind schedule."

I said we'd look into it. (By the way, I wasn't blowing Matt off; I really had every intention of checking it out. I just was skeptical we'd get it done for Future Sight.) Matt then brought up the idea to the team. It was received lukewarmly. Once again, not because it wasn't an interesting idea but rather the idea of creating an entire new card type is daunting at best. After a few weeks of no real response, Matt decided it was time to put down his thoughts and write a letter to the team. What follows is the actual letter. I should note that as with many real documents, I am unable to show you the letter in its entirety. While I know it is frustrating not to see everything, feedback from my readers has taught me that people would rather see parts of official documents than never seeing them at all. What I will do is show you all the relevant parts and then paraphrase any section that I am unable to show. I'm also going to jump in to point out things.

Hey guys,

A few weeks back I brought up the idea of a Planeswalker card type for the Pop bonus sheet. I don't think I did a great job explaining why it could be a very important part of growing the Magic brand and unifying design and creative. I want to sell you guys on the idea not only because we in the creative department feel very strongly about its importance and potential coolness, but also because I will need all of you to help pound out the kinks and make it a workable idea that will not be ignored just because it's not functionally plausible. If any of you guys end up thinking it's as cool as we do, I am sure a compelling but realistic design can be found.

The next section of the letter explains in detail many of the things that the creative team was working on and explains how important the planeswalkers were to that work.

Here are some of the rough ideas we have kicked around recently. The important thing about these cards is that they are NOT creatures that you can order around. They are their own bosses, and the closer they can get to feeling like a pseudo-player and not a "controlled permanent" the better. I hope some of you have made it this far, because this stuff needs the fine-tooth comb. Also, any completely divergent takes on how this could be done are absolutely welcome.

Potential Planeswalker Card Type Rules

1) Whenever an attacking creature deals damage to a player, that creature's controller may have it deal its damage to a Planeswalker instead.

2) Planeswalkers may be targeted as if they were players.

3) When a Planeswalker takes damage put [Damage] counters on it equal to the number of damage it has taken. If a Planeswalker has a number of [Damage] counters equal to or greater than its [Walk number] remove it from the game.

4) Planeswalker cards are not "under a player's control"

5) Planeswalker cards are legal in all formats (a long shot, but our best way of indicating the ability to travel between blocks/planes.)

Yes, planeswalkers started with only five principles. (Well, according to Matt.) #1, as you know, managed to make it all the way through the process. The big change was letting the creatures attack the planeswalker directly. #2 was talked about and dumped because it just didn't work within the rules. There are just too many cards that target players that made no sense with planeswalkers. It did, though, lead us to the damage redirection aspect of the cards. (I'll talk about this more below.) While the specifics changed a little, the basis of #3 also made it through the process. Four and five, not so much. While we spent some time trying to make #2 work, #4 and #5 were scratched before the first planeswalker design team (which I'll talk about soon) officially met.

What follows is some ideas of Matt's of what planeswalkers could potentially look like. Remember, for all of these, Matt was looking for a card that would basically act on its own thus giving the illusion of independence.

Ydarb The Twisted ("programmed" planeswalker)
Legendary Planeswalker - Goblin

Act 1: CARDNAME deals 4 damage to each of the players with the lowest life total.

(Acts are abilities that would be played at the end of each player's turn, in order, one per turn. After the last one is played, this card is removed from the game.)

Act 2: CARDNAME deals 4 damage to each of the players with the lowest life total.

Act 3: The players with highest and lowest life totals switch their life totals.

This version has planeswalkers walking in doing their thing and leaving. Note that Matt had some of the effects based on unknown information thus allowing some variance in what happens.

Izzob The Mad (watered down planeswalker)
Legendary Planeswalker- Demon

Cardname is black

At the beginning of your upkeep, if any player has four or more creatures in play put CARDNAME into play from your hand. At the end of your upkeep, destroy all creatures with the lowest power.

At the beginning of your upkeep, if there are no creatures in play remove CARDNAME from the game.

This version allows the planeswalker to come and go multiple times. Also, note that it is not summoned (aka no mana cost). It just comes when the conditions are right.

Attovac The Foul (Independent Planeswalker)
Legendary Planeswalker - Elf

(This "walk" number indicates the amount of damage this card can take before it is removed from the game. Whether this damage sticks around turn to turn is up for debate.)

Appear (Appear condition is met and CARDNAME is in your sideboard, you may put CARDNAME into play.)

At the end of any turn, CARDNAME deals 3 damage to each player.

If the life total of all players is less than 17, remove CARDNAME from play.

I went back and double checked to make sure I copied this correctly (and I did). I'm not quite sure what Matt meant here. I think he was messing with a condition that he called appear that would define when the planeswalker shows up. Once again, it has no mana cost and thus cannot be played normally. I do believe there is some text missing above as it doesn't make sense as is.

The reason I felt it was important to show you this letter is to get a sense of where planeswalkers started. Also as a Magic historian, I enjoy tracing back things to their earliest written version. The letter you see here is the first time anyone wrote down their ideas about planeswalkers. While a lot changes over the next year, it's interesting how many basic concepts that made it all the way through showed up in Matt's first version.

Up in the Sky, a Bird, a Planeswalker

This letter did a good job of kicking me, the Future Sight design lead and Magic Head Designer, in the tuchus. I realized that Matt was correct. This was the right thing to do. Creating a planeswalker card type was strategically a very important move. It was going to be a creative challenge but an exciting one. I also knew, though, that we would never add a card type to Magic if we didn't do it right. No matter how important the general goal was, R&D is very cautious about the game. Nothing gets added that can't hold its own weight. If we couldn't find a way to do planeswalkers right, there would be no planeswalkers.

Once I realized this I knew what had to be done. The Future Sight design team was not the right place for planeswalker design. No, if we were going to do it right, it needed its very own team. I put together the team as follows. First I added Matt. He was by far the most passionate person about the idea plus he was on the creative team so he could make sure we were fitting with what they needed. Next, I added myself as this was a daunting design challenge and I felt like the Head Designer needed to be directly involved. Third, I added Mark Gottlieb. We were treading into uncharted waters. Having the Rules Manager in the meetings seemed prudent. Finally, I added Brandon Bozzi, another creative team member. Brandon had been on a number of Magic design teams and he seemed like a perfect fit.

Matt CavottaMark RosewaterMark GottliebBrandon Bozzi

At the first meeting, I laid down what I thought the design parameters were. (We still weren't up to nineteen yet.)

#1 – This was a new card type. This meant it had to act like one. If we could recreate what these cards were doing as an existing card type then they failed.

#2 – The card had to capture the flavor of being a planeswalker. The impetus to doing this was to build up the planeswalkers. If the cards couldn't do that then they failed.

#3 – We wanted the planeswalkers to be good. This meant we had to create a design that development would be able to push. This means quite a number of things, but the most important was that answers to planeswalkers needed to exist without players specifically putting in cards that dealt with them. Planeswalkers couldn't be like artifacts or enchantments that depended upon then opponent having cards that singled the type out by name. Also, there are certain areas of the game that R&D has found problematic (fast mana as an example). Planeswalker design would have to keep from stepping on any of these design landmines.

#4 – The planeswalker design had to be simple enough for people to get. Yes, it got some complexity, but not so much that people didn't try to play with the cards because they just didn't "get it". This is a delicate balance but an important one. Note that the closer we tied the mechanics to flavor the more natural and thus simpler it would feel.

#5 – The design had to be elegant. When players saw it, it had to feel right. This is one of those designer things where I can just say that we know it when we see it.

While I listed out five things, what I'm talking about above is actually much more than just five. (Who knows, maybe more than nineteen.) Being flavorful and being simple and being elegant are very complex concepts when you actually get down to what they require from a design standpoint. Suffice to say that the team had its work cut out for it.

Snakes on a Planeswalker

I started the design knowing we had a daunting task before us. Magic had never printed a new card type before. (Yes, yes, we had mana sources and tribal—I mean a new card type that did something the game had truly never done before.) This isn't to say that we'd never designed one before. In fact, as you will soon see, one of the other new card type designs we did got merged into planeswalkers. But we had never made it far enough in the process with one to ever print it.

Because the design space was so wide open, I decided to follow my instincts and set a few limitations. First, I declared that planeswalkers were going to be a permanent type. While instants and sorceries are nice, they've never been as iconic as the cards that sit in play. If we wanted to make planeswalkers matter, I felt strongly that it meant they sat in play. Also, permanents are things. Planeswalkers are things. It just felt right to insist on planeswalkers being permanents.

Second, I stated that we were going to give planeswalkers mana costs. The reason for this was a very practical one. If we were going to do planeswalkers, it meant designing a lot of them over the long haul. Mana costs are a clean and simple way to make sure only the proper color or colors can play the card. If the planeswalkers didn't have a mana cost, we'd have to work in other ways to make sure they stayed "in color." While these other ways exist, they aren't plentiful and thus only mana costs would serve us for the long haul.

PlainsThird, I told the team that we were going to make use of the design of another new card type that never got made. During Ravnica design, Richard Garfield had created a new card type he called structures. To reflect the city feel of the set, Richard had come up with the idea of a card type that had enchantment-y / artifact-y type effects but could only be destroyed if it took enough damage from creatures. To allow this, the card type let creatures attack it. The flavor was that there were mystical buildings in the city that had the power to create magical effects, but unlike an enchantment, the building could be destroyed by creatures who could tear it down. We liked the design but Ravnica was so chock full that we never needed to use it. (Also, while cool, it didn't neatly fit into what the block was doing.)

Along come planeswalkers with the need to find a universal answer. Add onto this our desire to have planeswalkers feel like a player analogue, and structures seemed to be a perfect fit. While talking about this, Matt piped in with his idea of planeswalkers being able to be targeted by any spell that could target a player. Gottlieb liked the general idea but knew that the game couldn't handle treating cards like players. There were just too many cards that wouldn't make sense if they targeted a planeswalker. It was at this point that Gottlieb first proposed the idea of allowing players to redirect damage dealt to their opponent to their opponent's planeswalker. This way a direct damage spell could essentially (but no, not technically) target a planeswalker.

As this was coming together, we defined the idea of loyalty. Having planeswalkers have life didn't make sense as it was important that they were easier to destroy than you—otherwise the players would have no reason to change their focus to attacking the planeswalker. Planeswalkers always having less life than you didn't feel right. We then struck upon the idea of loyalty. Damage doesn't kill the planeswalker, it just encourages him or her to stop helping you. On the flip side, there are certain things each planeswalker likes to do. Allowing him or her to do these things makes him or her happier and thus more willing to stick around.

Okay, we knew how you got the planeswalker into play. We pretty much knew what was going to force it to leave play. All that remained was to figure out what they did while they were in play. That's all.

Planeswalking on Sunshine

To tackle this problem, we started with Matt's idea of the planeswalkers coming into play with a preset plan. The team thought it was cool that the planeswalkers did something different each turn and did so with a larger sense of purpose, with a plan. This, we felt, really made them feel like another player rather than just a creature or enchantment. The one thing I didn't like though was Matt's idea that they just left when they had done each thing once. I knew we could only write three or four things on the card (if that—while my column isn't really about how the layout happened, note that this was always a worry for us as we knew text space was going to be an issue), and I wanted planeswalkers to be able to last longer than that.

Colfenor's PlansMy suggestion was to loop the effects. That is, if a planeswalker had three abilities, after it did the third one it'd go back to doing the first. The team liked this idea but was worried what would cause the planeswalkers to go away. What if the opponent was unable to attack the planeswalker? The team felt like there needed to be a secondary force that restricted how long a planeswalker could stick around.

The solution to this problem was to add loyalty changes (usually down) to the abilities. This way without interference each planeswalker would have a natural duration. Yes, your opponent could speed that along but if they were unable to do anything the planeswalker would have a set time it naturally went on its way. It's also important to note that in this version, the loyalty loss wasn't a cost. This meant that if you had an ability that required a certain loss, the planeswalker would get the effect before leaving play due to it loyalty hitting 0. Remember this when I show you some designs in a moment.

With these parameters in mind, the team set out to design five planeswalkers, one in each color. Note that we didn't actually need five as the plan for Future Sight was to only include one to three, but we thought designing one per color was a good exercise. It turns out that we decided to do three on the Future Sight timeshifted sheet as it didn't seem right to only have one card that required as much to learn as a planeswalkers.

Once we designed the cards, we made up three decks. Each was monocolored, used four of its planeswalker, and was filled out with cards from Ninth Edition. Here are the planeswalker designs we used:

Planeswalker - Fendari
9 (loyalty)
1. Put two 1/1 green Saproling tokens into play. Lose 1 loyalty.
2. For each Saproling you have in play put a 1/1 green Saproling token into play. Lose 1 loyalty.
3. All Saprolings get +5/+5 until end of turn. Lose 3 loyalty.

Planeswalker - Bianco
6 (loyalty)
1. All players discard two cards.
2. All players sacrifice two creatures.
3. Return all creature cards in your graveyard to play. Lose 5 loyalty.

Planeswalker - Vittorio
8 (loyalty)
1. Target player draws two cards. Lose 1 loyalty.
2. Target player gains control of target artifact or creature with a converted mana cost equal to or less than the number of cards in that player's hand. Lose 2 loyalty.

Let me walk you through how exactly these cards work. I’ll use Fendari as my example. Before I do I’ll answer the question you are all asking. Why do all the planeswalkers have Italian names? All I know is Brandon named them all and he found it funny for some reason. That out of the way, here’s how it works. Fendari comes into play with his 9 loyalty (I don't think we were using counters at that point; we just said it was like a life total and had players mark it down). At the beginning of his next turn, he puts two 1/1 green Saproling tokens into play and loses 1 loyalty, dropping him down to 8 loyalty.

On his second turn, he doubles all Saprolings tokens (can you tell I designed this one? I do so love my doubling) and loses 1 loyalty, dropping him to 7 loyalty. Assuming no other Saprolings in play, there are now four 1/1 Saprolings. On his third turn, all his Saprolings get +5/+5 until end of turn, giving you four 6/6 creatures. As they don't have trample, the opponent has some chance of chump blocking to stay alive. Fendari loses 3 loyalty, dropping him to 4 loyalty.

On his fourth turn, we go back to the first effect. He makes two more Saproling tokens for a total of six (assuming, of course, none died when they attacked the round before). Fendari loses 1 loyalty dropping to 2. On the fifth round, he doubles all Saprolings, making a total of twelve if he's managed to keep them all alive. This cost another loyalty, dropping him to 1.

On his sixth turn, Fendari gives all Saprolings +5/+5 until end of turn, creating twelve 6/6 creatures. The act of doing this causes him to lose 3 loyalty, putting him at -2 loyalty—which is 0 or below, resulting in Fendari being put into the graveyard. Note as I said above that as the loyalty loss isn't a cost in this version, he gets off his last effect before leaving play. Hopefully twelve 6/6 Saprolings can get the job done.

All of the above assumes no damage by the opponent either with attacking creatures or damage-dealing spells. Note though that even 1 damage keeps the third ability from going off the second time. Another point of damage and Fendari doesn't get to double Saprolings twice. The idea behind this design was that each planeswalker got a plan which the opponent could mess with, mostly by speeding up their exit.

For the completionists, here are the two cards we made that didn’t get playtest card versions:

Planeswalker - Nursimi
6 (loyalty)
1. Tap all creatures controlled by target player and those creatures do not untap during their controller's next untap step. Lose 1 loyalty.
2. All creatures controlled by target player gain "whenever this creature deals damage, its controller gains that much life". Lose 1 loyalty
3. Put an X/X white Avatar token into play where X is equal to target
player's life total. Lose 2 loyalty.
4. Nursimi gains 3 loyalty.

Planeswalker - Tienti
8 (loyalty)
1. CARDNAME deals 2 damage to all creatures.
2. CARDNAME deals 3 damage to all players. Planeswalkers lose 3 loyalty.
3. Each player sacrifices four lands, then CARDNAME deals 4 damage to each creature and player. Planewalkers lose 4 loyalty.

You can see that we were experimenting with different ways to play around with this design. Nursimi, the white planeswalker, does not naturally leave play. If the opponent doesn't mess with it, the card goes up in loyalty over time. Nursimi, like Vittorio, plays around with having a number of effects other than three. It's also interesting to note that the white design goes almost full circle (see last week's Latest Developments, as Devin shows all the changes the white planeswalker went through) and touches upon a design idea from the very first design. Tienti, meanwhile, was playing around with having its loyalty loss be something global rather than personal. It also was a card that specifically dealt damage to other planeswalkers.

Back to the Future

Mindless_AutomatonWith our designs done, the team turned them into the Future Sight development team. (The designs took long enough that the set was already in development by the time we finished.) The blue, black, and green ones were put into the file. Tarmogoyf, for those that don't know this already, was the green rare taken out to make room for Fendari. Once in the file, they were added to the FFL (the Future Future League, R&D's playtest league that plays Magic six months to a year ahead of real life.) During playest, we learned some valuable things about the planeswalkers—the most important, many of the playtesters hated them. Mons Johnson (the man whom Alpha's Mons's Goblin Raiders was named after), in particular, made it well known that he didn't like the current design.

The issue was that many people hated the robot-ness of the cards. Because the player never had any input, the cards often did stupid or useless things. For example, when playing Fendari, if the opponent was able to get rid of the two Saprolings made on turn one (which is actually turn two, as nothing at this point in planeswalker design happened during turn one) then nothing happened during turns two and three (a.k.a. turns three and four).

The planeswalker design team explained why we had made the cards function as we did, but in the end enough people were unhappy. We had said since day one if we couldn't do them right then we weren't going to do them. Having over half the people who played them complaining about them was not doing them right. If we were going to do planeswalkers it was because we found an execution that we thought would be uniformly beloved (well, by the majority—there's always some small minority that hates everything no matter how much the majority enjoys it). No matter what drives an idea into the design process, only quality sees it through. The only reason we add things to the game is because we believe they are net positive. If something can't pass that metric it doesn't get added no matter how good the intentions behind it are.

Because of this reasoning, the planeswalker cards were pulled from Future Sight (welcome back, Tarmogoyf). We didn't have time to do them right and we were only going to do planeswalkers if we did them right.

The cards were sent back to me with a note from development: "fix these." And so begins the next stage of planeswalker design.

So how did we end up from the unbeloved robots to the cards of today? Join me next week and I'll tell you. (Come on, it said Part I in the title.)

Until then, may you not settle until you do it the right way.

Mark Rosewater