As I'm very active on social media, I interact a lot with the players. A very common interaction is players telling me things they'd like to see us do. Now, I love hearing all the suggestions, and many of them are things we can and will act upon, but I realized the other day there are several requests I often get that I don't think we're going to do, at least not for the reasons they're being suggested. So, I thought I would take today's column to talk about some of the requests I get and why I feel they're poor reasons to design a Magic product (or a game in general, for those who like to apply my lessons more broadly). I chose to talk about five as that's what fit in my word count.

An aside before I begin:

One of the assumptions I'm making today is that we're talking about professional game design (aka you're planning to try selling the game). If game design is just a hobby for you and you're not working toward having an actual career in game design, feel free to be motivated by whatever you wish. A bunch of the things I'm going to tell you not to do are bad business but could be a fun inspiration if you're more interested in the process (having an interesting challenge) than the outcome (making the best game you can).

Why Design Games?

Before I get into why not to do things, I want to stress what I believe the goal of game design is. You're trying to create the optimal experience for your target audience. What exactly that experience is and who you're aiming it toward has a lot of latitude. Many games are designed to allow players to enjoy themselves, but there are others made to force people to think about something or to feel something. And the possible audience can be any subset of humanity (or animals if you're into animal game design). There's a lot of room for flexibility, but there needs to be a focus, and that focus should be creating the optimal experience for your target audience. That said, let's talk about some of the requests I get that I believe are poor motivations for game design.

Request #1: "Prove that you can do it right this time."

This request happens when someone wants us to do something that wasn't very popular the first time. "Yes, Thing X did badly, but I believe with current R&D technology, you guys could hit it out of the ballpark." This request plays into the idea that redemption is a good motivator for game design. It's not, for a number of reasons.

First, game design is hard. You should start every design with a positive motivation. "Here's a cool thing that I believe players would love that I want to figure out how to build a set around." Or, "This is an awesome world that players will adore as soon as they see it. How do I mechanically bring it to life?" Trying to redo something that failed starts you with a deficit. Instead of building off positivity, you're trying to figure out how to compensate for negativity. It's just a bad place to start a design; build from a foundation of positivity.

Second, there's no reward for doing the harder task. Let's say I make Game A based on a cool idea and make Game B based on a failed idea that I want to redeem. Let's say both are successful. I don't get any greater reward for designing Game B than designing Game A, yet I'm taking a much greater risk with Game B. But, wouldn't Game B provide more internal satisfaction as you would know it was harder? No. You're going to work hard on any game you make, and you'll enjoy its success just as much. It's not going to sell better, you're not going to feel better—you're just unnecessarily choosing a harder path.

Third, it's bad business. There's a reason no one remakes movie flops. Selling anything is hard. The last thing you want is to build in a reason for people to write it off before they've learned anything about it.

None of this, by the way, is saying that you can't redeem things. The point is that it can't be the starting point for your design. Let me use devotion as an example. In Eventide, we introduced a mechanic called chroma that had scaling effects based on colored mana symbols on your cards. It didn't go over well. The mechanic tested very poorly in market research. That saddened me because in my heart I believed it was a good mechanic. Years later, we were designing Theros and were looking at mechanics that would synergize with the set. I was also looking for a way to capture the adoration the people of the world had for their gods as I felt it was an important aspect of Greek mythology, our source material for the set. One of the designers (Zac Hill) suggested chroma.

The reason I put it in the set was not to redeem chroma, but because it was the perfect solution for our problem. To make it fit, we needed to tweak it, but because it had been a flop, we felt free to make changes, including renaming it. I believe if I'd started a design with the goal of building it around chroma, I would have been in a much worse place. What this means is redemption works best when it naturally fits into an existing design rather than being the impetus for the design.

Request #2: "That sounds novel."

This request is an offshoot of the last one. The big difference is this idea is not based on correcting a past mistake, but rather doing something that inherently sounds daunting. I'll give an example. On Twitter a month ago (days ago as I wrote this), a Magic content creator (Maria Bartholdi of the Good Luck High Five podcast) requested having an expansion set in the Blind Eternities. I responded that would be difficult as the Blind Eternities is not a place, but rather the connective tissue of the Multiverse that exists between various planes. I said that setting an expansion there would be like having it set only in doorways. Upon hearing this, numerous players insisted that we now do it because it sounded fascinating.

What sounds cool and what will lead to a good Magic set are not the same thing. As an example, let's take the idea of the Blind Eternities. For starters, we need locations. At bare minimum, we need a plains, an island, a swamp, a mountain, and a forest. We can be flexible and metaphoric, but we still need concrete places. The Blind Eternities isn't a place, so this would be hard to do. The set needs creatures. Over half of the file needs to be beings native to where the expansion takes place. Other than maybe the Eldrazi, there aren't really creatures from the Blind Eternities. And even if we made some up, where are we picturing them existing?

The set needs resonance. There must be things the players recognize so that there's some familiarity. What exactly is that in a non-place connecting planes in a Multiverse? The very concept itself would be hard for many to wrap their heads around. The set needs magical spells. That means a need for creatures using magic, and most of it needs to be associated with a color of mana (or at least use colored mana in their mana cost). The set needs a mechanical hook. This usually comes from the resonance of the world or from a theme of the environment that thematically connects to a mechanic. The Blind Eternities has none of that.

The point here is that Magic sets have a lot of requirements to make the game work, and we can't just ignore them because the concept sounds novel. Things that might make a great story or a cool movie don't necessarily make a good Magic set. Part of designing a system is being aware of the basic needs of that system and making sure that your idea can work within those requirements. So, when someone says, "Make this cool world, but don't do anything you normally do," it's basically like me saying to a carpenter "Build me a house, but don't use any of your tools." I'm not sure it's even possible, but if it is, I'm not going to end up with the best finished product.

Once again, this doesn't mean that we can't take the essence of what interests people and eventually find a home for it. There are players interested in the Blind Eternities. Perhaps one day, we'll find a cool way to reflect it in gameplay. While I doubt it will be a setting, perhaps the concept could be tied to a mechanic or a certain kind of card design. It is not, though, a good jumping-off point for a set.

Request #3: "You haven't done it yet."

This request can come from a bunch of different places. Sometimes they've seen a component in another game, sometimes it's something they've thought up and sounds cool to them, sometimes it's an extrapolation of something else we've done. The request basically boils down to "I enjoy it when Magic does cool and different things. This is different. You should do this."

Magic should do cool and different things, and I like to think it does with some regularity, but doing something different for the sake of being different is a bad motivation. Let me explain why with a metaphor. Every time we a create a new element for the game, we're adding a new tool to our toolbox. How are new tools made? Usually, they come about because they allow you solve a problem you previously couldn't solve. You don't design a tool first and then figure out what you can do with it; that leads to a lot of very inefficient tools.

When we add a new component to Magic, we want to be making something that will advance what we can do with the game. I assume you've heard the expression "If all you have is a hammer, everything looks like a nail." The idea being that your vantage point is shaped by the lens you look through. Starting a design with a tool you don't understand can shape your design in awkward ways as you struggle to justify its use. The better thing to do with new ideas is explore them in what we call "blue sky design," where we examine ideas outside of the context of a set, and then put the knowledge aside and wait for the right need to come along.

A good example from Magic is double-faced cards. Wizards also makes a product called Duel Masters (it's a trading card game made exclusively for Japan). In Duel Masters, they made double-faced cards. R&D makes it a habit to be aware of the other games we make, so we realized that double-faced cards were an option on the table. Our first thought wasn't "How do we get this technology into Magic?" but rather to treat it as an awareness of a potential solution to a future problem.

That problem didn't come up until we were designing Innistrad. I'd assigned my team the task of solving the werewolf problem: how do we make compelling cards that were a Human some of the time and a Werewolf the rest? We had many different proposals, but one of them was from a team member (Tom LaPille) who had worked on a Duel Masters design that had used the double-faced cards. I was skeptical at first, because the double-faced cards had a huge number of logistical issues, but when it proved to be the most efficient tool to solve our problem, I was motivated to figure out how to combat those issues.

Double-faced cards have gone on to be a valuable tool in Magic's arsenal, but a lot of this was because the initial designs were organic enough that they allowed me to win the fight to get them made in the first place and ensured the audience had a positive interaction with them, thus allowing us to further extrapolate their use in other sets. Had I chosen an awkward interaction there's a good chance I would have had to remove them due to increased internal pressure, or had they managed to make it to the public, I might have burned equity (aka soured the public on them, making their reuse very difficult) on a mechanic that had a lot of potential.

The lesson here is not to avoid the unknown, but to understand how to harness it properly to ensure you're making the best possible use of it. In general, that comes from finding a home for its first use, not forcing one.

Request #4: "Do that again, but different."

This request is a bit different than the first three. Instead of wanting us to do things we've never done before, this request wants us to redo something we've already done and apply it to a new thing. For example, "Make another Ravnica, but with different guilds." The impetus behind this request is that the player likes what we've done so much that they just want us to do it again, with a tiny twist.

Here's the thing. We do use successful designs as a springboard for related ideas, but one of our goals when we do that is to create a unique identity for the new thing. For example, I believe we'll do two-color factions again, but when we do that, our goal is going to be to make it feel different from the guilds of Ravnica. Remember, Ravnica block itself was us reinventing the multicolor theme of Invasion block. Part of making a game that constantly creates new content is positioning what we do in a way that makes things feel fresh and new even when they're influenced by the past.

The lesson here is that the past can be a good tool for new ideas, but we have to understand that each idea needs to be built from the ground up maximizing what makes it work best. Also, because we live with our past, we have to be conscious of how new ideas can find their own expression. This means it's good to listen to players when they express a desire for a new version of an old thing, but we want to understand which component of the old thing was the essence they'd like to see extrapolated into a new product.

Request #5: "Break your rules."

One of the things I've been very proactive about as head designer is sharing with all of you the thoughts and structure behind how we design Magic. I strongly believe that an informed player who understands why we do what we do has the ability to further appreciate the game. A side effect of this education is the last request I'm going to talk about today. This request comes from players who understand our systems and request that we break them. "This color doesn't normally do Thing X. Have it do Thing X." "This creature type doesn't normally belong to this faction. Make one in that faction." "This mechanic normally only appears on sorceries. Put it on an instant." The impetus behind this request is easy to understand. It's exciting to do the thing you normally can't do. However, breaking rules just to break them leads to various problems.

1) You undermine the system – The color pie, for example, isn't just for flavor. It's the basic foundation of the game, allowing five different colors to exist. Breaking the color pie weakens the overall system, and if done enough, can permanently harm the game. There's a reason people don't just go around breaking laws. Systems, especially in games, matter and must be treated with respect.

2) You confuse the messaging – The reason different factions have different creature types is to give them identity. If you start making exceptions, you muddy that message. Remember, zero means you don't do something. One means you do. Even making just one exception will change players' perceptions. And with a trading card game, you don't even control which piece players see first.

3) You can worsen gameplay – If a certain mechanic is only on sorceries, there's a reason for that. We spend years working on the sets, and we're very methodical in our choices. What might not seem like a big deal on the surface can have all sorts of ramifications when actually applied; the rules don't work, it leads to unintuitive play, the execution on digital is problematic, it has a negative impact on tournament play, the rules text won't fit on the card, or it combines strangely with another component of the set. Magic is a complex game with a lot of interactions and uses. Changing one thing can often cause problems.

This isn't to say rules can never be broken, but they have to be done carefully and with thought put into what effects are being added to the game.

Final Thoughts

There's a thread running through today's column. You can redeem your failures. You can do things that are intriguing. You can explore ideas you've never approached before. You can repeat your successes. And, you can break rules that have never been broken. But none of those are where you want to start your designs. These are things you want to examine separately and then keep in limbo until their rightful place shows up. They should be used because there's a way to organically fold them in to your game, not because you're out to prove something or because it's a challenge.

Good design is about understanding the needs of your system and making sure your jumping-off point is setting you up for success. Yes, you want to leave yourself open to the possibility of the unknown, but that shouldn't be what leads you. I understand it's exciting from the outside to want to zig where the game normally zags, but as a game designer, you must temper that chaos with structure. You have to make sure any new elements are reinforcing what you're doing, not driving you to do things you shouldn't.

And that brings us to the wrap-up of today's column. I hope you enjoyed it, and as always, I'm eager to hear your thoughts. You can email me or contact me through any of my social media accounts (Twitter, Tumblr, Instagram, and Google+).

Join me next week when I hear a who.

Until then, may you continue to be a source of endless requests.

• Episode 612 Lorwyn Cards, Part 2
• Episode 611 Lorwyn Cards, Part 1
• Episode 610 Bends and Breaks