The Man Behind The Mechanics: An Interview With Michael Elliott!

Sure, you know Mark Rosewater designs cards… But you only know that because he does all the PR stuff. Meanwhile, behind the scenes, Michael Elliott was the man responsible for shadow, cycling, slivers, licids, growing enchantments, the Rancor bouncing enchantments, echo, rebels and mercenaries, spellshapers, the oaths, fading, the avatars, gating, the en-Kors, the Laccoliths, the Flowstones, madness, the incarnations, amplify, and more…

So we sat him down for a good long interview to ask him why it looks like Wizards is predesigning our decks for us, how he would redesign Magic if he could do it from the ground up, his rivalry with MaRo, and what the most unpublishable mechanic he ever created was!

Q: How did you start out at Wizards?

A: I used to be a tournament bridge player; I played for six years before I started at Wizards of the Coast. It was fun travelling around, and playing against people all over the world, and that’s actually what hooked me into Magic. We used to go to play games after we got done playing bridge, and one of my friends introduced me to this new game. He’s like,”It’s really cool, here are the rules: There’s a big pile in the center, which is the land, and then you have five other piles of different colored cards, which have marks on the top to tell you what color they are, and during your turn you choose which pile you draw from.”

And it was a multiplayer game, and you started with twenty life – but he liked to play it so that you dealt out Nuclear War population cards and that was your life total.

So I really loved the art and the style of the cards, so I went down to several of the local game stores and I bought a bunch; I stood in line for Legends. And I picked up a starter pack with the rulebook, and I thought,”Wow, this isn’t anything like the game!” I played at the game store and learned the real rules… Although I’d won three local tournaments before I realized that you could block with more than one creatures.

Anyway, I went to local tournaments, and I was at a tournament in Phoenix called CopperCon, and there were a couple of guys who were asking people about Magic; they asked me what I thought about the game. I told them everything that I thought was wrong with it, and they said,”Wow, that’s really interesting. Would you be interested in flying out to Wizards of the Coast?”

Most of the people in R&D have varied backgrounds. I believe that Mark Rosewater used to write for some show – some sci-fi series or something – and Bill Rose had a background in accounting. I actually used to work in pathology.

As it turns out, I was under contract at the hospital until the end of the year, so I ended up flying out and started on January 1st, 1996. I started out as a developer and then I got promoted, and they did a reorg at one point and I got promoted to senior design. Early on, everyone was considered a developer, and then they broke up design and development, and Mark and I got stuck in design.

Q: So what have you designed?

A: Since I’ve been in Wizards for eight years, I’ve been in design or development on every single set since I got there, with the exception of Portal: Three Kingdoms. The first Magic expansion I’m not involved with is Darksteel, which will be breaking an eight-year streak.

I was lead designer on Onslaught, Legions, Urza’s Saga, Urza’s Legacy, Stronghold, Exodus, Planeshift, Mercadian Masques, and Nemesis. I’ve been a designer on sixty-three games and expansions, many of them not Magic – games like Pokemon, Star Wars, and the upcoming Simpsons game.

Mark Rosewater and I do a lot of the design stuff for Magic; Mark’s specialty is doing cool individual cards, and my specialty is doing mechanics.

Q: What mechanics are yours?

A: Um…. (He starts ticking off mechanics on his fingers) Over the years, I’ve done shadow, cycling, slivers, licids, growing enchantments, the Rancor bouncing enchantments, echo, rebels and mercenaries, spellshapers, the oaths, fading, the avatars, gating, the en-Kors, the Laccoliths, the Flowstones, madness, the incarnations, amplify, the thought devourer critters…

Not the free stuff. Not buyback. Those were not mine.

Q: So how do you deal with the fact that despite the fact that you’ve created a tremendous amount of cards, Mark Rosewater the one who everyone knows? He writes a weekly column, and you don’t… But you do the same kinds of work. Does that ever get to you?

A: Well, everyone has their responsibilities at the company, and basically the fact that I don’t have to do the web content and I don’t have to travel frees me up to be on more projects… And I love doing design work. I get to do the other CCG systems; I hope that one day Simpsons or NeoPets or Duel Masters will hit big, and then everyone will know.

We do have a bit of a rivalry, though; we both always wanna be involved in the new set designs. But part of it is, the fact that he does the public stuff means that he’s almost never able to work on other projects. So I get them. (He smiles.)

Q: When design all of these other non-Magic games, how do you keep from just aping Magic’s central mechanics over and over again?

A: Sometimes we do. There are multiple cost models you can use, but you don’t always have to use the multiple-mana style. Several games – NeoPets, Simpsons, and to some extent Star Wars – are classic mana models, and if you start with a classic mana model, you’re going to end up with a game that’s similar to Magic, but if you go away from that, you will wind up with a different strategy.

I’ve worked on X-Men, Netrunner, Star Wars, Harry Potter, Simpsons, and NeoPets. And out of those, NeoPets is probably the most different from Magic, with Star Wars being a fairly close second. The Simpsons, though, is the first multiplayer game we’ve designed since Jihad. It’s designed so each person can buy a deck and actually play. It’s a very beer and pretzels game.

Q: So if you had the opportunity to redesign Magic from the ground up, what would you change?

A: That’s an interesting question, since we talk about that a lot internally. If we were to launch the game now, it would be similar to Duel Masters, where every card counts as both a spell and a land; you can either cast it, in which case you face it up one way – or you can turn it upside-down and use it as a land. That way, it’s impossible to get mana-screwed.

There are a few things in Magic that are somewhat kludgy in terms of game play, and the whole mana curve system in particular is very cumbersome. To some extent, it’s a strategy and a deckbuilding thing, but it’s also a barrier to entry, since you have to figure out this complex mathematical formula before you can start building decks. If you’re a high-level gamer, that’s good – but if you’re a low-level gamer, it can lead to a bad play experience where you get manascrewed.

Also, some of the early abilities were also way too complicated for their effect, like banding and other stuff that didn’t fit on the cards. I’m a big fan of”the game system is simple, and the cards tell you what they do”; you shouldn’t have to look up how rampage works. Even after all these years, banding is still our top most frequently-asked rules question for Customer Service.

But note that if I were starting out with Magic today, one of the things I wouldn’t fix is the variation in the power curve. I think in most CCGs, you need to explore the power curves first and rein back later; you want the Black Lotuses and the Moxes and the Time Walks. It’s too easy to find the variations in the power level when they’re all even; you can do a more sophisticated balancing later on.

Q: Do you think that Magic should have more colors?

A: Not really. I think the best games are five- or six-color game systems. If you have four, you have a clear enemy and two enemies or allies – but with a five- or six-color system, you have the”allies and enemies” syndrome, which creates a much better play balance. With three, you have two enemies; with four, you have two enemies and friends or vice versa. Five is a good number, because you can have two enemies and two friends. With six, you have three enemies and two friends, and one is a very strong enemy.

I just did a game design for NeoPets. The game is geared to ten- to thirteen-year-olds, but there’s enough depth and combo stuff to keep it interesting. In fact, as an anecdote, Paul Barclay was forced to playtest it for a couple of days, and he ended up spending the entire weekend building NeoPets combo decks. And he normally hates all of other CCGs except for Magic, so that was an interesting stamp of approval. I brought that up because NeoPets is a six-color system, where each color has a strong enemy, two weak enemies, and two allies.

The other problem is any time you go above six, the more likely it becomes that additional colors aren’t meaningful for gameplay, and you’ll have situations where you don’t get the right colors for your decks.

Q: What’s your main goal for each expansion?

A: To do something different that we haven’t done already; something fresh. I’m always trying to find mechanics or playstyles that go in a different direction than the last block did, so you feel that things are shaken up.

Q: What’s the most success you’ve had in terms of shaking things up?

A: I think one of the most game-shifting mechanics that I’ve done was the madness mechanic. Torment shook up the environment more than I actually would have liked, but I think it had the most impact.

Q: What would you have different with madness?

A: I think we underestimated how easy it was to discard stuff, so I would have adjusted the cost on things, had a few more effects that would have worked well against madness to tone the power level down a bit. It reminded me of the free mechanic of Saga, where the mechanic just went out at too high a power level. (Note once more that I was not responsible for the free mechanic.)

Q: There’s a perception out there that Wizards is predesigning our decks for us. Do you predesign decks as a part of the development process, or is that the result of some other developmental technique?

A: Part of that is that one of the way I design cards is that I’ll be playing a deck and I’ll get into a situation where I’ll say,”I’m losing. How can I win this game?” That’ll make me think of something. Those cards were developed to fit into decks that already exist, so they’re easy to put in.

The other times, I’ll think of a cool effect and it won’t get played, so I try and think of a couple of cards that would help that effect show up. Otherwise, you have a cool rare with no support for it, and to some extent that destroys the spark of what you’re talking about… So you end up with two or three cards that work well together.

But no, we don’t go through and build the deck and say,”This is what will dominate for the next two years.” Randy and his group will develop a lot of decks that are similar to what the tournament players play, but there are often variations that the pros come up with that weren’t decks that we had tested. And some of those people claim that we built it for them, and we had no idea that deck was out there!

Q: Like what?

A: Some of the madness decks; the Upheaval deck is another one. People say,”Oh, you knew about the Upheaval + Psychatog combo.” And we did. But when we played it, we touched on it a little bit, but we underestimated how good Upheaval was. In our test environment, the decks were a lot faster, and you weren’t able to ramp to six or eight mana.

Q: So why did that change?

A: Part of that is that at the end of development, one or two of the cards will change, which throws out all of the testing. One of the cards that knocked Upheaval down got changed from two mana to three, and suddenly you have a slower game where some of these decks that weren’t viable are now the dominant decks in Standard.

Q: Did you see, say, Astral Slide decks coming?

A: We didn’t fully see how it interacted with Exalted Angel, which was kind of an echoish effect – but even so, it still isn’t what we would consider a dominant deck. While it’s probably higher than we anticipated, it’s not out of control or anything.

Q: So given these things that you’ve miscalculated, is the four-month cycle of development still viable?

A: Right now, Magic is such an established brand that we’re a year out on development. If something major comes up where there’s a big shift, we can easily adapt.

Q: Like when? When have you adapted?

A: One of the classics is Tsabo’s Web – that was a response to Rishadan Port going insane. That was another one where we undervalued the speed of the environment. At that point we thought the environment was slowing, and that Port wouldn’t have an impact on the speed game that it did, and we underestimated the impact that it had on multicolored decks – so there’ s a case where we felt we had to put in a magic bullet.

Tsabo’s Decree, the anti-rebel card, was also kind of a magic bullet. Most of the time, we camouflage it because they’re useful cards that happen to be good against a particular thing. We would try and avoid doing a card that was just like,”Destroy all goblins.”

Q: So how do you feel when you design a card that gets banned – or that everyone says should be banned because it’s overly powerful?

A: I take it very personally when I do a card that actually ends up getting banned. You have to push the bubble, so it’s bound to happen, but Urza’s Saga was an example of where we had a very short development time; Rosewater and William were both on other projects, so most of the R&D”team” was just me and William playtesting. We didn’t have the full team dissecting cards like we do nowadays. Our developers have improved lately in the past three or four years, so now Mark and I don’t have to sit on development teams because, well, we’re just not that good at it.

Q: So how many cards get booted out of any given expansion?

A: In a typical small expansion, 143 cards, probably five or six cards get booted out because the effect isn’t able to be costed at a point where it’s interesting and not broken. There are a number of cards where at one mana, the card is way too good, and at two mana the card is not playable. So you wind up with a bunch of cards where the ideal cost is like 1.4 mana. And so those cards often end up getting killed, because we don’t want to put out a card at a really bad cost, because that defeats the coolness of the card.

Q: All right, then – what’s the worst mechanic you’ve ever created?

A: There was one earlier that was anti-mana; you could cast things that would tap lands. (Yoiks – The Ferrett) So I feel safe discussing that one, because we will likely never get that one. But there were cards where if you had three black lands tapped, you could play it for free.

There are a couple of other ones that have been moved off, and often things get resurrected. For example, the incarnation mechanic originally started as spells that you could play, and after the spell went off there was a residual effect that lasted in the graveyard. And nobody liked that one because you had to remember all of that stuff going on; I pitched that six years ago. And then we said,”Let’s do a graveyard block,” so we brought back a variation on it – at a much lower level than we had pitched it originally, but then again we only did seven cards as opposed to the fifteen.

There are a couple that are currently broken that I will figure out how to put into the game. I promise you.

Q: Do you design for complexity? After all, Magic’s a very hard game to learn; do you take that into account when designing?

A: The way it effects things is that in a certain block, there are certain common card mechanics, and certain rare mechanics. Generally, the uncommon and rare mechanics are similar, so they’ll push each other out of the set, and you’ll end up doing a common mechanic because you have to put it on fifteen to twenty cards – and you don’t want to wind up with common Licids.

You wouldn’t see banding on Benalish Hero these days. It’d be a rare. You don’t want the players who aren’t hardcore tourney players picking up the card and saying,”What does this do?” and then having to look it up.

Q: So what’s the best mechanic you’ve done?

A: Slivers. They’re just cool, and it’s something that casual players love. The inspiration was Plague Rats; I was staring at an old Rats card and said,”You know, it’d be cool if instead of adding just power and toughness, you could add all sorts of other abilities.”

Q: And your worst mechanic?

A: I don’t think the rebels worked out as well, because they had too much shuffling. It created a slow, defensive game, the deck type was too obvious, and it kind of cast a shadow on the whole block because the deck dominated the format. Some of the other stuff like the sleeping enchantments kind of got thrown by the wayside because of that.

Q: And your worst mechanic in terms of creating confusing rulings?

A: *laughs* Oh, madness! Probably followed closely by Licids. The Licids had all sorts of huge, complicated interactions – but they’re in the past, but we don’t have as many questions on it.

Speaking of rules headaches, Stifle, by the way, was a reaction to the Decrees. The whole other stuff, like the fact that it can stop you from losing with Phage the Untouchable, was all secondary to the fact that we were trying to get in a something to someone pulling off an uncounterable Armageddon… And I argued very heavily that we should get that in. The secondary effects happened to overshadow the original idea, which happens occasionally.

Q: What’s your favorite color?

A: My favorite color is white. Because I kind of like the weenie strategy with combat tricks, and you get more combat tricks in general in white than you do in red or green.

Q: Your fondness for white’s surprising, given how everyone’s been bitching about how weak white is lately. What have you learned from balancing the color pie?

A: (He moans in agony.) Don’t make really good card-drawing cards. *laughs*

Seriously, everyone hates counterspelling. The most-hated deck types for casual players are LD, discard, and then counterspelling – probably in that order. Tournament players will play whatever the top decks are, and their game pattern is a lot different from the casual players. But the casual players will often build the goofy deck to get their trick to get fifty squirrels in play – and when you have counterspells, that’s the thing that annoys them the most. So I always attempt to keep counterspells at a low level whenever possible. But since that’s not always under my control, to some extent my opinion doesn’t matter; I can price a block counterspell at 7UUU – but if the playtesters want it at 1U, it’s at 1U because of our organization structure. That doesn’t happen too often, though.

In the meanwhile, we try and make all five colors not too far apart so that all of them are viable in some constructed deck. If you get an environment where everyone’s playing mono-red, or when 40% of the decks in Onslaught Block are Goblins, I would consider that type of environment to be a slight mistake. But you have to keep the timeframe in mind: If there’s a four-month period where white and red are dominant, that’s fine – but if it’s set after set, then that’s a mistake. You want other people to keep thinking that these other colors are viable.

One of the rationales of the color pie so to try and divide it so that they’re even and that all five colors are viable… But you only have to miss on one or two cards to have your color wheel fly out of control, so it’s a very hard task for development. I’m somewhat glad that I don’t have to do as much control anymore, since that’s a hard and thankless task; if you do it right, nobody notices, and if you do it wrong you get yelled at.

Q: What about Amplify?

A: The playtesters were afraid it was too keyword-heavy, and I went on vacation for a couple of days, and came back to find out that they had cut amplify down to four cards. And I said,”Hey, this isn’t a tournament mechanic, the casual players are gonna love it.”

Q: What card has Rosewater designed that you wish you had?

A: (Thinks) Tough question. We’ve both done well over a thousand cards… (Thinks more) Wow… Probably Phage the Untouchable.

Q: What card do you think Rosewater envies? (This is my blatant attempt to get MaRo to post in our forums… The Ferrett)

A: Probably Living Death. I know he likes that effect and he’s done a couple of variations on it. He used to play a Reanimator deck.

Q: How do you know when you’ve done a good card?

A: If I’m excited to pitch it around to people. If I can’t wait to show it around to people, I know I’ve got a good card. If I call up Bill Rose at night and pitch it to him on the phone, and I know I have a good mechanic. I usually won’t bother Bill on a single card.

An interesting side note is that the Magic foil system was something I had also pitched from day one since I got to Wizards. And it took me that long to put it through.

Also, we’re discussing things that I was responsible for, putting the rarity symbols on the cards was also my idea. I originally pitched that I would like to have the numbers and a C, or U, or an R on each card so that people wouldn’t get beaten up on the trades. It always bugged me when I saw a kid trading a Craw Wurm for a Shivan Dragon, so I thought it would be cool to have a rarity symbol on the cards to give them a rough idea of the card’s worth – even though, as we all know, not all rarities are created equal.

They thought it was too obvious and it was throwing the card value in people’s faces. So I said, What if you did something with the expansion symbol? So that’s where the rarity system evolved out of. Later on, we decided that other brands would do symbols; they’re not nearly as obvious, but they’re a lot easier than shading the symbol.

Q: Do you like the job?

A: Oh yeah! It’s fun. I have no regrets. Sometimes I’m on a bunch of different projects, but it’ll get stressful, but most of the time it’s very fun. It’s very satisfying to walk around at tournaments and see stuff you worked on.