Cosmic Crucible is an asynchronous player vs player game mode for the hit mobile game Marvel Strike Force.

I was initially tasked with designing a tournament bracket visualization for the new tournament feature that was going through its design phase. This was very tricky because, despite many examples of similar tournament systems existing in the market, ours was to use a Swiss tournament bracket, rather than a simple single or double elimination bracket. I was unable to find any comps in the competing market space using a Swiss bracket, and the ones I was able to find in other applications tended to be unwieldy, unintuitive, and visually dense with little hierarchy or visual style.

I set about iterating on designs, experimenting with varying levels of detail and informational fidelity. I also experimented with providing different areas of emphasis for different designs.

After considering over a dozen designs that all failed to hit the mark, despite small successes in each, I ran across a visualization for an obscure e-sports tournament that felt clear and simple and visually dynamic.

I realized that with this style of visualization, some of the information regarding who gets matched up against whom is hidden, which is one reason I had not gotten to this solution myself. But, upon reflection, I realized it was not necessary to surface this information to our players, and could therefore reduce the complexity of what I was asking them to interact with. All they needed to know was, given their starting point in the bracket, where would they end up if they won, and where they would end up if they lost.

With this simple mental model created, I had a design that was straightforward and would give players confidence in predicting what the outcomes would be from their engagement. The similarity to a map, across which the player would move, gave me the idea to use the existing patterns from our raids system, with which the players would already be intimately familiar. This gave me a design that would fulfill the goals delivered by the system design team of creating a simple and intuitive interface, emphasizing the impact of wins and losses above all other measures of performance, emphasizing the path of the tournament winner, while motivating players to try to win, and to keep engaging even if they didn’t. User tests showed that they were able to easily understand and interact with the system.

I realized that this visualization was fertile ground for theming and could serve as the centerpiece and landing page for the entire feature. I began to flash out the rest of the screen as such, making use of existing patterns as much as possible to keep players comfortable in the context of this new paradigm.

It was also a goal of my own to add delight and visual power and excitement to a game that too often devolves to dense UI’s dominated by frames and boxes. I worked closely with the art director and the art team to build an exciting visual experience that would still support the UX I had designed. In support of this, I even created some rough visual concepts in order to define the visual expectations. Our talented team of artists delivered a final result that I was quite happy with.

With the highest risk piece of the feature settled, I proceeded to fill out the remainder of the screens for the feature. With the centerpiece already involving lots of new patterns and visuals, I emphasized reusing existing patterns from throughout the game as much as possible for the rest of the feature. Borrowing heavily from alliance war, raids, general leaderboard patterns, and more made the rest of the feature feel familiar to engaged players, allowing them to interact with the feature according to their already established expectations.

I modeled some simple player journeys to gain insight into the meta behavior of users adopting the new feature and incorporating it into their play patterns. This exercise highlighted a clear deficiency: the game’s home screen menu was expansive and cluttered, and simply appending the new feature would almost certainly result in low adoption, with users unaware of the feature’s cadence and when events would start and end so they can engage appropriately.

This led me to the start of another project where I studied our main menu and started formulating strategies for updating it in light of the current state of the game. While this was an informative and rewarding exercise, it unfortunately didn’t solve the immediate problems surrounding surfacing of the new feature. Any main menu updates would be a major feature in their own right and would take significant time to get through the development pipeline. As such, we released without adequately addressing the issue and consequently saw lower than desired engagement, both measured analytically and hearing anecdotally from users.

Over the next few releases we took a number of actions to address the engagement issue, as well as other issues that came to light from player feedback and analysis channels, and also improvements that were planned by the design and product teams.

The feature continues to undergo tuning and iteration. The main responsibility for these has now been handed off to another designer, who I consult and coach as they continue refining the experience.