On my podcast, I do a series called "Lessons Learned" where I go through every set I've led or co-led the design for and talk about what I learned from the experience. Earlier this year, I decided to start doing a "Lessons Learned" series through Making Magic. Here are the previous four:

Last time, I got through original Ixalan, so Unstable is next in line.

Unstable

Lesson: "Success often takes time."

Unglued came out in August of 1998. It was in many ways the first supplemental set. Unfortunately, that meant we didn't understand what a supplemental set was yet, so we just treated it like a small premier set, which ended up being a problem. We printed way too much of it, and although initial sales were good (enough to encourage us to design Unglued 2), it ended up being regarded as a failure. Unglued 2 was cancelled, and I didn't think we'd ever make another one.

Six years later, we managed to make Unhinged. There were plenty of design problems with the set (see Part 1 of this series), but again we overprinted it as we didn't really understand how to make supplemental sets yet. The set was dubbed a failure, and this time, I really, really didn't think we'd get to make another one.

Flash forward seven or eight years. I discovered that there were two people in R&D besides myself who wanted a third Un- set, Mark Purvis and Mark Globus. We dubbed ourselves the Council of Marks. Purvis is one of the people who oversees the business of Magic, and he believed he had the data to make the pitch that the Un- sets could be successful if handled correctly from the business side (by adjusting the print run, for example). Globus understood the nature of how R&D greenlit sets. Here was his plan:

We had learned from our printers that there was a lot of new printing technology either currently available or on the cusp of being available. Globus pitched the idea of working on a set that stretched the boundaries of what the new printing technology could do. It wouldn't have a specific release date, and it was something we would work on to better understand printing capabilities, but it was a real product, so if we happened to make something useful with our testing, we could potentially sell it. That set would be the third Un- set. What better way to push boundaries than a product defined by its experimental nature.

This meant that we began working on the design without any idea when, or really if, it would release. As so much time had gone by, I had lots of opportunity to think about how I could make the Un- sets better. My biggest realization was that I needed to make use of modern design technology. I needed to make an Un- set like we'd make any set. This led me to the idea of making a factioned Un- set. Ravnica introduced the idea of a faction set and was so successful that it became something we did regularly.

For the creative, I decided to lean into a theme that we had talked about but had seemingly shied away from: steampunk. That would lead me to the idea of "mad inventors," which would then lead me to the decision to try and bring Contraptions (a mechanical concept based on a one-card joke from Future Sight) to life.

I used to joke that Unstable was "the little engine that could," because no matter what obstacles were thrown at it, the set kept chugging along, intent on getting up the "production hill." Other products displaced it. The technology we were designing around didn't happen (or more accurately, hadn't happened yet). Magic decided to make a premier steampunk set (what would become Kaladesh) and delayed the product three different times.

But we used that time to continue to improve on the set. It went through three different development leads. We made numerous changes to adapt our mechanics to the technology that existed. In the end, we put out something I was very proud of, and it was well received by the audience. The product would end up getting reprinted four times.

The lesson here is one of endurance. Not everything happens quickly. Sometimes the key to success is to have a vision and slowly work toward it, even if it takes great time. This is not only true of Un- sets but also mechanics, themes, and set ideas. An important part of Magic design is understanding the long game. Magic, as I like to say, is a hungry monster. If you believe in something, you need to figure out what about it shines and then spend the time finding the right home for it. So many of my successes started with great doubt from the rest of R&D. The key is realizing that good ideas can win out, but they require a lot of nurturing and refining.

Dominaria

Lesson: "Great ideas take iteration."

This set was a daunting task. Early Magic had spent most of its time on the plane of Dominaria. Over 40 expansions took place there (or partially there). We eventually embraced traveling our Multiverse and made a lot of different planes. We found that it worked best if each plane had a clear theme. Themes help to define flavor and mechanical identity. However, the plane of Dominaria has a bit of everything. Its definition was almost a lack of definition. This was the challenge presented to the team when we started exploratory design for Dominaria.

The challenge was simple. We wanted Dominaria to feel like Dominaria but have a clear theme that we could craft the design around. We started by making lists of everything associated with the plane. We listed every theme, every mechanic, every location, every character. The plane had amazing depth. One of the advantages of having 40-plus sets occur there is that we'd already done a lot of world building. Also, a lot of Magic's story takes place there.

That was the point we kept coming back to. So much had happened on Dominaria. They had an ice age, a Phyrexian invasion, and were the center of a near collapse of the Multiverse. A lot had happened there. And not just in the backstory. The players had been there each and every time something had happened. The stories weren't just the past of the characters, they were the past of the players.

That idea led us to the theme of history. The line I used in vision design was "Dominaria is a plane whose present is shaped by its past." Now history was a theme. It was something we could rally around. Normally, when I leave exploratory design, I have a good idea where a set is heading. When Dominaria exploratory design ended, all I knew was the theme of history. How does one mechanically capture that? We hadn't figured that out yet.

The lesson of this design is the importance of iteration. We began vision design not knowing how we were mechanically executing on our theme. We got to solutions, not because we had a sudden epiphany of how to do it, but because we took ideas and worked on them. I'm going to use two as an example today.

First is Sagas. We got to Sagas because I felt it was important to represent the idea of stories. What is history if not a story about the past? We started with the ask, "How can we mechanically represent a story?"

We tried a lot of different things. What we latched onto was a mechanic we'd abandoned while trying to design planeswalkers. We then spent a lot of time exploring all the different ways we could execute on it. Richard Garfield, for example, mocked up a bunch of cards that looked like a gameboard with a track running through it. He used icons to represent different effects that would happen over time. We incorporated another idea that we originally had done as a separate group of cards representing the art through which the people of Dominaria told their history. (We liked the idea that different people told of the same event differently.) Sagas would be further iterated upon in set design. What we ended up with was a game component that was so compelling that it quickly became deciduous and has been used in a whole bunch of sets.

The other game component was historic. Because of sets around us, we were avoiding using the graveyard, which was the obvious place to look for a mechanical means to capture the idea of history. So we made a list of the other game elements that had a sense of history. The big three were artifacts, legendaries, and Sagas.

It was Aaron Forsythe who suggested we make a mechanic that cared about all three. That idea took much iteration to get to the right place where it was playing well, but even then, we still had a big obstacle. It didn't feel enough like history to people. It just required a lot of work to creatively get to the idea.

Bill Rose, the VP of R&D, asked me to remove it from the set. I argued that it was key to our theme, and he gave me a month to improve it. I worked closely with the Creative team and iterated on designs, complete with finished art, because I realized this was more of a perception problem than a gameplay one.

In the end, I came up with the concept of batching, that is, using a flavorful word in the rules text to combine multiple gameplay elements and then defining that rule in the rules text. While this change might seem simple in retrospect, it took a lot of time and iteration to get there.

Looking back on Dominaria, I see a lot of struggle, but I also see great discoveries. Having to find the best version of our ideas led to new tools that upped the overall technology of our game design. For example, The Lord of the Rings: Tales of Middle-earth™, one of our latest sets, takes advantage of both of these tools, and you will see them get used a lot in Magic's future.

Guilds of Ravnica and Ravnica Allegiance

Lesson: "Synergy dictates structure"

I'm going to talk about Guilds of Ravnica and Ravnica Allegiance together because they had a shared vision design. This was our third trip to the plane of Ravnica, and it happened as we were transitioning from the "two and two block" model (each block had one large set and one small set on the same plane) to our "three and one" model (three large sets, each drafted by itself, and one large core set). To simplify things, we chose to do the vision design for both sets together.

The very first thing I asked about was shaking up the Ravnica set structure. Ravnica: City of Guilds had introduced the "guild model" where different guilds were broken up between various sets (4–3–3 based on having one large set and two small sets in the original block). At the time, it was a radical design choice. There were numerous arguments over whether we could even make sets using the structure. But in the end, we did, and it was a huge success.

Return to Ravnica block shifted from 4–3–3 to 5–5–10 (with both the first two sets being large and the third set being medium) but other than that kept the basics to the structure. Each guild had its own named mechanic, would show up primarily in one set, and would be designed to synergistically work with the guilds of the same colors. Original Ravnica had also done a good job defining the flavor of the guilds, which we tweaked slightly, but mostly kept the same.

When we started Guilds of Ravnica and Ravnica Allegiance, I asked if I could shake things up a bit. It was our third visit, and design technology had changed a lot since original Ravnica. I was hoping to reinvent what it meant to be a guild set.

I was told no. We were ramping up the story for a big climax in War of the Spark, which we wanted to be in Ravnica. We were worried that it would upset players that Ravnica wasn't guild focused, so we decided to visit the plane first and complete with two traditional guild sets to satiate the players' desire for guild gameplay. The set being the same was seen as a feature.

The lesson of this set is about a mechanic we made for the Orzhov. Because we weren't innovating on the structure, I thought it might be fun to push innovation a little more on the mechanics. My favorite mechanic we made was called debt. The way debt worked was that you would give debt counters to your opponent as a rider on certain spells. At the end of the turn, your opponent could pay 1 mana for each debt counter they wanted to get rid of and then they would lose 1 life if they had any remining debt counters, although always only 1 life regardless of the amount of debt.

The mechanic was very flavorful. Early on, you could often ignore debt, but as the game progressed, it would become more and more of an issue, sometimes even being the reason you lost. There were a lot of fun games where players were frantically trying to get rid of debt, and it would shape the whole endgame. I was so enamored of the mechanic. It was flavorful, it was novel, it was super Orzhov. So, what happened to it?

One of the most important lessons of Magic design is that the overall system and structure is more important than any one design or mechanic. Something that works well in a vacuum but doesn't integrate with the rest of the set is a problem. Debt was great when that was all you were focused on, but we were making a Ravnica guild set. That meant the mechanics had to synergize with the other mechanics in the same color. For a set to be dynamic, different elements must work together. For instance, you need cards to be fought over by different drafters. You need cards to click together with various other cards to create variance of play over time.

Debt was awesome, but it wasn't exactly a team player in a set where the factions had to intermingle and play nicely with one another. It was removed because it wasn't a functional tool in the grander scope of the design. Now, as I said above, Magic is a hungry monster, and not every set has the same needs, so it not working out in one set doesn't mean it can't find a home in another.

The key lesson here is recognizing what your synergy needs are so that you structure the set appropriately. Different sets have different needs, and it's important for the lead designer to understand what set they're making. It's a hard thing to do when you're in the middle of your design and you make something lovable, but it's an important part of making your set better and improving as a Magic designer.

War of the Spark

Lesson: "Embrace what the set is about."

I remember the day I first heard about War of the Spark. Doug Beyer was pitching his initial version of the Bolas Arc (a lot changed along the way), and it ended with a giant battle between Nicol Bolas and his army and essentially every Planeswalker we knew. I turned to Doug and said, "You do know I only get three planeswalker cards per set."

It was a three-year arc, so I had a lot of sets to work on before I got to War of the Spark, but I thought about it a lot. It was this daunting design challenge waiting for me on the horizon. How does one design a set for a Planeswalker war? For a while, I leaned in on the "war" part. What if we could find a mechanic that conveyed a sense of war? We tried something we called skirmish, where you had an outside game piece that you and your opponent fought over. Each time you dealt creature damage to your opponent, you advanced toward your end of the game piece. If you ever got there, you got a reward.

The lesson of this design happened about halfway through the design. Something was bugging me about where the design was heading. I couldn't quite put my finger on it, but I knew something was wrong. After days of agonizing over it, I finally figured it out. I was designing a Planeswalker war set, but the audience wanted a Planeswalker war set. I was focusing on the wrong thing. Why? Because I didn't think focusing on the other thing was possible and was trying to find a solution that felt more possible. I finally said to myself, "I'm fighting player intuition. A Planeswalker war needs a lot of planeswalker cards." I decided I needed to change my attitude. Instead of wondering if it was possible, I needed to focus on how to make it possible.

I walked into the next design meeting and said, "Here's what I want to do. Let's just do what the set wants us to do and figure out how to make that work." What the set wanted was a lot of planeswalkers. It was a Planeswalker war. How do we do that? That led us to explore as-fan, which made us realize that we had to do uncommon planeswalkers. Okay, how do we do that?

We explored how to simplify planeswalkers. I made use of a tool I'd been holding back on, static abilities for planeswalkers. We investigated how we could make hybrid planeswalkers to help with as-fan issues. Instead of saying, "Well, this can't be done," we explored what we had to do to make it possible. And it worked. War of the Spark had 36 planeswalkers, twelve times the average of a normal set. It became the focus of the marketing and was the key selling point of the set.

The lesson of this design is that you can't fight your theme. You must understand what your set is asking for and then figure out how to bring that theme to life. The best Magic designs deliver on the thing that players want most. Sometimes that's daunting, but not facing it will only lead you to the wrong design.

Live and Learn

That's all the time we have for today. I hope you enjoyed these insights into the design process. An important part of that process is getting feedback from the players, so if you have any thoughts on today's article or any of my lessons from it, including thoughts on the sets in question, feel free to email me or contact me through my social media accounts (Twitter, Tumblr, Instagram, and TikTok).

Join me next week when Wilds of Eldraine previews begin.