Mirage was the second Magic set that came out after I started playing, so I'm not exactly full of development stories. Luckily, we still keep some of the old guard hidden around the office. I went over to Mike Elliott's cage and bugged him until he agreed to pinch-hit for me and write an article for Mirage week. Besides, this way there's one more article of old anecdotes for me to read, too.
So much has changed in Magic set development over the past few years that it is amusing to look back on the old days. For Mirage, developed in the winter of 1996, the development team consisted of Bill Rose, Mark Rosewater, Charlie Catino, William Jockusch, and myself. Interestingly, none of this group works on development on a regular basis anymore. Mark works mostly on design and support content. Bill is the head of R&D and doesn't sit on many teams. Charlie heads up shared services and also does not sit on many teams. William is no longer with the company. And I work mostly on design and sit in on a development team about once every fifteen months. Most of the current developers would laugh at the techniques we used back then for set development.
While Mark, William, and I had done some playtesting on the previous set, Alliances, that set was mostly done by the time we all came to work for Wizards. Many people who have played Magic have dabbled in creating cards, and Mark and I, who at that point had never been given the opportunity to be on a design team, both had a huge library of card ideas we were chomping at the bit to inflict upon the world. Our "eagerness" created an interesting dynamic on the team. When we were doing card passes, comments would come up on individual cards about how exciting a card was, and if a card was criticized, even slightly, either Mark or I would chime in with, “Yes, we should replace that!” We would then pitch idea after idea in the meetings until Bill picked something he liked. On a typical set these days, between 20 and 40 percent of the cards submitted by the design team get cut by development and replaced during the development process. Mirage had a much lower percentage than that, so opportunities to stick in new cards were scarce. We had to pick our spots. For example, it was felt there was no really huge "wow" card in the set. Everyone was very happy with the recursive damage card (Hammer of Bogardan), but we wanted a really cool card that casual players would love. The precedent we were hoping to mimic was Jester's Cap. Jester's Cap was hugely popular during the Ice Age days, as it let you do something you could not do before. The team, with Mark leading the charge, wanted a similar type card for Mirage. Mark and I brainstormed dozens of goofy game altering effects before we finally came up with the card that went out as Grinning Totem. And when the set came out, we were surprised at how well the card was received. This really highlights how different the environment was back then, since nowadays Grinning Totem is regarded as just another overpriced gimmick card and we break the game rules much more often today than we were willing to do back in 1996.
Mirage was one of the first sets that was developed for Sealed Deck play. While a primitive pointing system was used prior to Mirage, the development team did a lot of work refining the scale and grading system that is still used today, although a few modifications have been made along the way. Many cards were adjusted up or down because this color or that was pointing too high or too low. Our fine-tuning was not to the level it is today, and I will be the first to admit that the colors for sealed did not end up as precisely balanced as they looked on paper once the players got their hands on them, but the set certainly played a lot better than Ice Age for sealed, and you didn't have a bunch of hugely complicated commons floating around like Ice Age had.
The keyworded mechanics of the set, phasing and flanking, pretty much stayed intact through the development process. The biggest argument over mechanics was whether or not phasing would trigger comes-into-play and leaves-play effects every time a creature left or came back. The final solution was not the best implementation of the mechanic, and we have since done pseudo-phasing effects that just trigger the way people think they should. If we did this today, they would probably all trigger both ways and just be costed for that effect.
Color hosers were all the rage at the time also. Today there is more of a feeling that color hosers are very narrow and create an unwanted "silver bullet" effect. For example, my fast green beatdown deck wins a good deal of the time, except when my opponent plays Perish, which crushes it like a bug. At the time, there was a mix of developers who wanted really strong color hosers and those who did not like the huge swings created by these effects. The poster child for this discussion was Forsaken Wastes. Many arguments were made that it was not a good enough hoser, since white could just Disenchant it. So arguments were made to give the card protection from white, but it was felt that change violated white's strength of destroying enchantments. After several heated discussions, the compromise was made to allow white to Disenchant it, but the white player would lose life doing so.
We didn't have the color pie back then either, which is too bad because, hey, who doesn't like pie? We had more of a "color stew." There were no real restrictions on what colors could or couldn't do. A number of people felt that many mistakes had been made in the past with cards like Anarchy and Swords to Plowshares, but any discussion on a particular ability was still a card-by-card battle, with the person willing to argue the longest (usually until one or two of the other team members had fallen asleep or left) won the argument. There were a number of heated color wheel arguments. For instance, Illumination and Withering Boon both originally countered any spell, with Illumination (WW) giving your opponent life and Withering Boon (BB) costing you life. Bill, who has since developed a loathing for counterspells, was actually a strong proponent through most of the development for bleeding the full counterspell ability to white and black, blue's allies. In the end it was determined that this was an ability we didn't want to really give to colors outside of blue, so the two spells ended up being limited counter versions. (Editor's note: This theme was revisited in Invasion—properly costed—as Absorb and Undermine.)
Another heated flavor debate involved the Dragon cycle. There was a building consensus at the time that evasion abilities such as flying should be focused more into specific colors and that some colors should be penalized heavily to get flying. William made a very complicated chart in which he calculated how much power and toughness you got for each of the various mana costs in each color, and then we assigned mana cost values to abilities. For example, blue paid .1 mana per point of power or toughness for flying. When you look back at the chart today, it looks hideous, but at the time it seemed to make sense. No one wanted the green Dragon to have flying, but it was a Dragon and no one wanted to break the Dragon cycle either. Under the green-pays-through-the-nose (and other orifices too) costing system, the flying green Dragon would have been about a 2/2 flyer for 4GG, with trample of course, since green got that for nearly nothing. The compromise was to make a hopping Dragon, which could still maintain the gist of the cycle and just pay a bit for the flying ability.
We have certainly come a long way in our development skills since the old days, but given the short period of time we had for developing the set, I am fairly happy with where the set ended up.
Send questions and comments to firstname.lastname@example.org.