Roanoke: A Post-Mortem on Undergraduate Research Game Design

Cathlena Martin, University of Montevallo
with Stephen Gilbert, Jesus “Chuy” Guizar, Will Kirkpatrick, and Sara Perry

Undergraduate research can be time consuming for a faculty mentor, but the resulting student learning makes the effort worth it. This essay reflects on an undergraduate research (UR) project on game design completed over the span of one academic year with four undergraduate students—Stephen Gilbert, Jesus “Chuy” Guizar, Will Kirkpatrick, and Sara Perry—as they developed a game. It is written from a professor’s perspective in order to provide future educators with the pros and cons of working on game design for an undergraduate research project. This reflection essay is based on my personal pedagogical experiences while directing an undergraduate research project on game design and draws heavily from my students’ experiences.

I teach at Alabama’s public liberal arts university, the University of Montevallo, which offers a Game Studies and Design (GSD) minor. Each of these four students completed eighteen credit hours of GSD coursework, including the History of Games, Survey of Modern Games, Game Design Workshop 1 and II, and Mathematics of Games, and they collectively approached me to request an undergraduate research project. Previously, I had worked with one of them in an independent research project, presented on a panel at an academic conference with two of them, and been to several in-state and out-of-state gaming conventions with all of them. These students had also served as officers and leaders in our SGA campus gaming organization, the Montevallo Organization of Gaming (MOG). In short, they were a rock-star group of students who were excellent gamers and enjoyed a tight friendship. Even with amazing students and a clear goal for the course’s outcome, we encountered challenges for this UR project of designing a legacy game.

A legacy game is the newest genre in board gaming and refers to a game that can be successively played as a campaign and incorporates permanent decision making, typically via destruction of components and alteration of the rules, with potentially each individual game affecting the next play through. Legacy games typically run ten to fifteen successive games, cost anywhere from $60-$140, and can generally only be played through once without a recharge pack. The first legacy game, Risk Legacy, was published in 2011, with Pandemic Legacy, Seafall, Gloomhaven, and Charterstone following. My students identified the goal to design a legacy game that could be played in a shorter amount of time, was affordable, and could be played through more than once. Our course objectives for the fall semester were as follows: understand the (brief) history of legacy games, foster a team collaboration, and create a base legacy game prototype. Our course objectives for the spring semester built on the previous semester, and included: flesh out a legacy game prototype, professionally print a base legacy game using GameCrafter, and present the game to at least two conferences/conventions.

Regrettably, we failed to achieve our ambitious end goal of producing a legacy game. This was in large part due to me allowing the students too much agency and being too flexible with deadlines. I did not enforce a core design maxim of start small and produce a minimum viable product first. On the plus side, my students embraced the opportunities provided them and even exceeded my expectations for convention attendance and networking. In the process, they became experts on legacy games, researching and consuming the small number of samples available at the time in the genre. Even though it was a learning experience through trial and error, they eventually found a team model that worked for them. And, they successfully branded their group, forming Space Lion Games. Overall, they achieved what most professors hope to instill in their students—a desire to learn, grounded in a base set of skills, and knowledge that they can keep developing after they have left the classroom.

Iterative Process

Game design is an iterative process, but it has to start somewhere, usually with either theme or mechanics. My students started with theme and then developed selected mechanics based around their theme. And while every game needs prototyping, playtesting, revising, playtesting, revising, and playtesting (rinse and repeat), the iterations of my students’ UR game were not so much iterations as four new games. Their initial idea was to create a competitive legacy game. They drew primary inspiration from Seafall, and they wanted to “emulate that same feeling that [they] had from Seafall and put it into our game with a few modifications” (Gilbert). But this changed significantly over the four key versions of the game:

  • Version 1: Fantasy Monster Hunting, which changed after a couple playtests
  • Version 2: Dystopian Revolution, which lasted  about five iterations
  • Version 3: Martian Homesteading, which lasted about twelve iterations
  • Version 4: Investigating Roanoke, which is the “final” version and had over thirty iterations.

The consistent“driving criteria” for the game as identified by Gilbert was that it “had to be fun” (Gilbert). But that was about all that stayed the same. Perry details some of the major changes exclusive to the fall semester:

We’ve gone through the full range of design while working on this project. We started off wanting to design a competitive monster hunting game and had so many cool ideas for it, but we discovered major problems with it and switched to what we were already basically designing, which was a co-op game. The design for this game has gone full circle, starting with the firm assertion that we shouldn’t have any dice, moving into semi-dice based, a full blown dice builder, and then moving right back to no dice. Although it’s been an extremely long process that’s put us behind on the project schedule, I think that the process has been essential to getting to this point. We currently are moving to a no dice format, but the elements of meeple bag building that we’re implementing as a core mechanic wouldn’t exist without all of the dice mechanics we struggled through. (Perry)

Part of the reason for these drastic changes was a lack of direction: “A major mistake was that we didn’t know what we wanted and failed to figure that out for a long time. Our lack of focus cost us tons of hours of progress when we would start over or revert to older versions of the game. Because of this, we never got the momentum we needed to get a game over the hump of ‘playable’ into ‘good’” (Kirkpatrick). When they struggled, they erased instead of modifying. I should have helped them better dissect their initial game to understand what elements to keep. In Gilbert’s words,“We don’t want to put out a mediocre game so we scrapped it and started from scratch” (Gilbert).

But the process was useful. After going through UR, Kirkpatrick noted that, “our game likely never would have been made if not for the numerous failings that came before it” (Kirkpatrick). The iterations were necessary. The students did not complete a prototype of a legacy game, but they did produce a base game—Roanoke. Roanoke is a survival based game about the colony of Roanoke with the following core mechanics: cooperative, hidden roles, and resource management. In terms of narrative, the players are delivering supplies to the colony of Roanoke in the late 1500s. After delivery and spending a night at the village, they awake to find all the residents, and the ship they came in on, gone.

During their iterative process, we encountered several challenges that could have been mitigated by me as the instructor. Having had these students in Game Design Workshop, where they produced a game every two weeks and polished a final game with smaller deadlines along with way, I wanted to give these students more freedom with this project. This resulted in a lack of an early minimum viable product, not enough constraints, and deadlines that were too flexible.

Challenges to Overcome

Unfulfilled Early Minimum Viable Product with Constraints

When the students pitched making a legacy game, I knew it would be a massive undertaking. But they were passionate about their project and I supported them in their mission. However, I did not always support them in the right ways. As Perry identified, “I think one of our biggest problems was trying to go too big, so we’re starting really small and building up from there” (Perry). Kirkpatrick echoed these sentiments: “There were lots of mistakes made throughout this design process. Chief among them, I believe, was that we tried to design a game that was too big for our design experience. We wanted a game that had a lot of parts to it, and we didn’t have the design chops to pull a fun game out of so much stuff. The next design I work on will be a small, simpler design” (Kirkpatrick). While he was self-deprecating and chalked the failure up to their design experience, it is just good design practice in general to start small and build tentatively no matter what level of designer you are. And I should have enforced that. However, every failure is a chance to learn. And, for the future, Kirkpatrick learned that he “would for sure start much smaller” (Kirkpatrick). As their professor, I should have reinforced the lessons they learned in their game design workshop course, limiting them and requiring an early minimum viable product that they had to build on instead of wholesale scrapping in favor of the next iteration. Additionally, I should have supported them with more constraints. As Kirkpatrick notes: “We had constraints, but they were too loose and didn’t help guide the design process as much as they should have. Another lesson learned: Constraints are not only helpful, but often necessary” (Kirkpatrick).

Lack of Firm Deadlines and Dedicated Meetings

The design issues could have been addressed more readily if I had firmly enforced deadlines and required more formal meetings. At the beginning of the semester, we had a schedule with tentative deadlines for playtestings and prototypes built into the syllabus, but our University does not treat UR as a class with a dedicated meeting time. Instead, it is an independent study course where the professor and students determine when and how often to meet. In the syllabus I had a generic statement common in our UR curriculum about instructor meetings: “You will meet with the instructor at least once every two weeks, though some meetings could be held weekly as needed. Specific meeting dates and times will be agreed upon between the instructor and student.” But our meetings ended up being haphazard. The students and I attend MOG where we get together every Tuesday and Thursday night from 7:30pm-midnight during the Fall and Spring semesters and play games. This means I saw them quite a bit. Because I knew they were all free and available during these times, we scheduled our meetings during MOG. But our meetings sometimes took a back seat to other gaming activities. For example, the students needed to play a lot of published legacy games for their research project, and legacy games are time intensive. I would get casual reports from them about their progress, and I knew they were working diligently on their game, but I should have enforced a different meeting time outside of MOG when there were no distractions. On the plus side, the students did have to learn to manage their own team time better. By the end of the spring semester, Perry commented, “we’re just getting to the point of being able to set our own deadlines, which is exciting” (Perry). As all four of them have now graduated, this will be a skill they need moving forward to be independent designers.


Even with these challenges, the students managed to succeed in several areas, making them better designers overall. They each learned lessons or skills they will apply to future games they design. They have multiple gaming conventions under their belts, along with the networking and experiences conventions bring. Space Lion Games now has a cohesive group of designers who, after quite a lot of trial and error, have learned to work as a team, and they will take their lessons from the UR project with them to future projects. As Kirkpatrick notes, “There are a thousand things I want to make sure we do going forward. Chief among them is continue making iterative design changes, to continue going to conventions, and to begin mathematically modeling our game more to stop prototypes that are so frustratingly broken that very little data can be taken from them” (Kirkpatrick).

Enhancement of Design Skills

Since the entire project revolved around designing a game, it gives me great pleasure to report that a success of the project was that they matured as designers. As lead designer for the game, Kirkpatrick took away the most overall design benefits, but I believe each of them benefited:

We learned the importance of combining theme and mechanisms into a package that feels cohesive to the player. We learned how game mechanisms interact with each other and how players react to different faults and successes in games. We learned about the importance of iterative design, about the difference between balance and perceived balance, and about how to efficiently produce changes and prototypes. We learned about the necessity of playing games, the importance of variety, and how simplicity often improves a design. There are too many lessons to fully go into. (Kirkpatrick).

And that just scratched the surface of successful lessons.


One of the key lessons was teamwork.We designed the UR project to mirror a company dynamic. Each member was to have their own research and agenda that supported the collective goal. They were each to function as their own “department” within the company, but the collective outcome for all four students was the same—produce a board game. Kirkpatrick, with the help of Gilbert, was to design the actual game, specifically focusing on the mechanics. Guizar’s task was to produce original artwork for the game. Perry’s part was to complete a market study of similar games to the one that they wanted to create. She was also to develop a PR and marketing scheme.

Yet even though we outlined Kirkpatrick as the lead designer, it took the team a while to learn what that means and how that works. Each student had something to say about teamwork. Gilbert sums it up: “I have learned that making a game with five people is pretty difficult because each person has different ideas and beliefs” (Gilbert). Guizar elaborates on the issue:

When we first started this project, I felt like we could get a lot done with all of us working together on it. We got together to brainstorm ideas for what this game could be. A lot of ideas were thrown out, and we voted on which one we should go with. At later meetings, all of us tried to design the game at the same time. This lasted for about two weeks. We had too many ideas that clashed with other ideas. It was also hard to pick an idea that everyone was on board with. It felt like any idea that made it into the game had to be approved by everyone, if one person didn’t like it, then it wouldn’t make it into this game. Some nights we just sat in the room together and didn’t get much work done because of this.” (Guizar)

Perry seconded her teammates: “this project has also helped us discover a lot about how our team functions together. We started off with everyone trying to co-design the game together and that bombed horribly. There were too many people with too many different visions for the project and that led to a lot of backtracking and lost time in the early part of the semester. I think we’ve each sort of found our spot in the team and what we can offer, so now we can use that knowledge to figure out how to work more cohesively and more efficiently” (Perry). And Kirkpatrick concurred: “our group dynamic lacked cohesion. We learned that all five of us designing one game was creating too much chaos and the resulting game felt like a disjointed, unengaging mess. Along with dozens of other design lessons, we learned that too many cooks in the kitchen will ruin a design” (Kirkpatrick).

Teamwork was the number one issue that the team struggled with outside of design. But through the process they learned what did not work and what did work. Guizar notes:

Eventually we did get better at working with each other. It kind of felt like we were each trying to be the lead designer. We decided that we should stick to our specific roles and the lead designers should make the final decisions whether one person likes something or not. I think overall we learned how to finally work together. When we finally sat down to talk about how it’s not working, I felt relieved because now only one person could focus on the actual design instead of all of us. We started getting better prototypes after that and some of them could have been our final game. (Guizar)

The role of a lead designer was something that we built into the syllabus, but the group dynamic of close friends shifted that original intention to one of communal design. While they had to struggle with realizing that did not work, they eventually did follow the model of a lead designer. Perry appreciated this learning process for team dynamics: “this project was really good for helping our team understand how each of us works, both when designing and when working on a team. I feel like that knowledge is going to be very beneficial as we transition towards independent work outside of school where we’ll be responsible for our own progress” (Perry). Their team organizational chart is visible in Appendix A.

Enhancement of Skills: Individual skills and specific discipline learning through personal discovery.

In struggling to find out how to best work as a team, they increased their individual skills and discipline specific knowledge while each fulfilling their own part. Guizar was responsible for the art, Kirkpatrick for the design, Gilbert for the narrative, and Perry for the project management and PR. Knowing they were responsible for their specific role to the team helped spur personal learning in their own specific area. For example, Guizar joined a few groups on Facebook and Reddit for graphic design and illustration for board games, as well as Facebook groups for game designers. He notes, “I have been paying attention to how art and board layouts work together more. I’ve also taken intro to graphic design this semester and plan on taking an advanced graphic design class next semester in preparation for this project” (Guizar). Kirkpatrick focused on design and industry skills, stating, “I now know how to properly pitch a game, who to pitch games to, the right methods for publication, the kinds of things publishers want to see in games, better ways to playtest games, how to better interpret feedback on games, how to better find the fun in games, what kinds of things gamers are attracted to in games, and the list goes on” (Kirkpatrick). Perry learned she could not do all jobs but instead had to specialize:“During this semester I’ve found out that I can’t do all major design related jobs at once. I want to design, develop, and act as the Space Lion manager, but I found that I can only seriously do two of those at once. In the future I’m going to need to switch between design and development while I act as the Space Lion manager” (Perry). She created and manages the team Facebook, Twitter, and Instagram accounts, actively using them for establishing a company brand.


            While the physical objective of the project was to create a board game, the team also wanted to use this project to help establish their own company—Space Lion Games. They now have established a decent social media presence with 85 Twitter followers and 69 Facebook followers. Perry assert that, “These followers mean that we’re successfully establishing an audience that’s interested in us as a design group, which will be crucial to gathering funds and support during future design endeavors.” (Perry). Perry posts to their group Twitter page regularly and to their Facebook and Instagram accounts semi-regularly. In addition to social media, the team produced physical objects for marketing: Guizar designed the blue space lion logo and each team member has specialized business cards. They were able to share their social media contact and business cards with industry professionals when they attended gaming conventions.


I have always been a strong proponent of exposing game design students to the networking that happens at gaming conventions; therefore, as part of their project, I required attendance to at least two conferences or conventions. They participated in six:

  • Dice Tower Con in Orlando, Florida – July 2017
    • Metatopia in Morristown, New Jersey – November 2017
    • Design Retreat with Waitress Games in Dallas, Texas – December 2017
    • Design Retreat with Waitress Games in Atlanta, Georgia – February 2018
    • Unpub 8 in Baltimore, Maryland – March 2018
    • Undergraduate Research Day in Montevallo, Alabama – March 2018

The board and card game industry is a fairly intimate group, with published designers accessible to new designers during conventions.Perry asserts:“Attending cons has been especially helpful this semester at establishing connections and an audience for our designs. If we keep going to cons and playtesting print and play games for other designers like we’re currently doing, we’ll be well on our way to the inner circle of the industry.” (Perry).

Learning Process

Through it all—conventions, teambuilding, iteration after iteration—the students came away from the UR project accomplishing more than I could have listed in the course objectives.  “I think that working through a group design for this project has been extremely helpful in figuring out how Space Lion Games will move forward in the future. We’ve established a working development model, designed a good base game, and have established a presence and connections within the industry” (Perry). She continues, “Overall, I think the project was a success. We learned a lot about design, each other, and the industry, and I think that knowledge is worth more at this point in our design careers than one really good game. I think we have a wider selection of tools to work with now when approaching game design, and that should be really helpful as we transition to our own studio environment and away from an academic one” (Perry). Kirkpatick concurs: “This semester has been a series of lessons learned both the easy way and the hard way. It’s been long, difficult, stressful, and incredibly rewarding. My abilities as a game designer have evolved more this semester than ever before and my ambition to be a professional game designer seems closer than ever before. I wouldn’t be anywhere near the game designer I am now if not for the experience of this project, and for that I am extremely grateful for the opportunity” (Kirkpatrick).

The students not only took away a lot of lessons learned, but also a desire to continue, as Guizar notes: “What we ended up with was a prototype for the base game. I think we are all pretty proud of what we have at the moment. Hopefully one day when we get it more fleshed out it can be a published game. Because we don’t want to stop working on it now that the semester is over. I think we have all gotten attached to the game and want to see it turned into something great.” (Guizar). Gilbert sums up their experience: “This was not an easy process to do but it was rewarding and a journey.” (Gilbert) What more could a professor ask for from her students?

Appendix A