Jasper Stephenson

UX engineer, friendly ghost.
Header image – Positivity: Using Bracket Data to Encourage Low-Placing Players

Positivity: Using Bracket Data to Encourage Low-Placing Players

This is a project that was born out of frustration and self-disgust. Born out of feeling dejected, and worthless, and never wanting to come back. It turns out that going to a Super Smash Bros. Melee tournament isn't always as cheery as it sounds.

No matter how friendly the competition, losing out 0-2 in a tournament sucks. Like, really sucks. Even if you don't consider yourself a competitive player, it's the kind of thing that hits you in a strange spot of pride that you didn't know you had. When I found myself in that position, wanting to run and hide even while I was among friends, all I could think was, "I have to do something about this."

This is about searching for redemption for all of us down there, bravely grinding away in the loser's side of the bracket. This is a story of positivity, and of gradual growth, and of being proud of what you achieved, even if you lost 0-2. This is one for the underdogs.

Wait, Super Smash What!? Super Smash Bros. Melee is a video game released by Nintendo in 2001. It's gained prominence as a 1v1 competitive esport in the years since its release, and you can find tournaments ranging from 10 to 2000 players all over the world. The weekly meetup I go to most often is called Friday Night Melee at Tokyo Game Bar in Akihabara, Tokyo.

Yeah, yeah, that all sounds dramatic, and I've defined a problem. But how the heck do I start to approach this as a project?

A tool meant to help people absolutely has to be based in understanding of those people's feelings. With that in mind, I decided that I needed to conduct some user interviews, and to do that, I needed to go to the source.


So there I was, the guy at the Friday Night Super Smash Brothers meetup conducting user interviews while playing games. It's the only form of user interview I've ever done where I couldn't look my interviewee in the eye; we were always both too focused on the game onscreen.

This was quite a challenge. On the one hand, I have to do all of the standard interview things like keep the conversation flowing in a direction that gives me useable insights into what I want to know. On the other hand, I have to actively play the game and try to put up a good challenge so they'd feel engaged and want to keep playing. All the while, I have to memorize their answers long enough to scramble for my notepad after the interview and scribble down what I had just heard as reliably as possible. It was the most engaged and human set of interviews I've ever done.

My standard interview setup.
My standard interview setup.

Traditional interviews can sometimes feel like a struggle to inject humanity into the conversation. Here, though, we were communicating through the game on a competitive and fun level, with the push and pull of the game providing a natural flow of tension and resolution to the conversation. It felt like a scene from a movie, where the main characters duel while trading quips. It made for a fantastically engaging set of interviews, and a relaxed and open set of interviewees.

My interview topics focused on a few key points of interest:

  • In the Moment: What factors shape a player's feelings and actions after a loss? Do those feelings change over time?
  • Pride: What makes a player feel good about their own performance at a tournament? What makes a win or a loss particularly impactful?
  • Social: What kinds of rivalries or friendships exist in your relationship with this game? How does it feel to see your peers succeed or fail?

Getting Real

Something that stood out in my initial research was when players would tell me things like "well, I lost to him, but he always wins around here. He even beat [a top player] once." I started to think about how I could surface that kind of background information for players who weren't regulars in the scene and might be discouraged to lose to players they have no context around. I also realized that tournament bracket data probably contains a lot of opportunity for tracking a player's growth and achievements over time.

With that in mind, I started by seeing what kind of data was available from the tournament's online bracket API. I kicked off a github repository called simply "positivity."

First github commit. Gotta start somewhere.
First github commit. Gotta start somewhere.

Without worrying about style, I just wanted to get information on the page to see how it felt to play with point systems and what kind of information I could surface from the data.

You can see how aesthetics really took a back seat to just putting things on the screen.
You can see how aesthetics really took a back seat to just putting things on the screen.

// dev side note The API for our tournament bracket system doesn't have a field for tournament host! I want to automatically find all tournaments from a specific tournament series, so I use Phantom.js to render the tournament page on my Node server so I can parse out the host's username once it gets AJAXed in after page load. Now, the user doesn't have to manually enter every tournament they've ever been in.

Prototype Testing

Laptop in hand, I went back to the meetup on a snowy January evening, ready to run some live demos. It's always amazing how just sharing a prototype or running an interview can kickstart motivation.

The version of the prototype I showed to players that night.
The version of the prototype I showed to players that night.

I heard from a player named Michael, who told me about his emotional experiences as a world-class Yu-Gi-Oh card game player. A guy who goes by "h0p" told me in a clear and calculated tone about his self-improvement notebook that he always writes in after matches.

Having heard from them and many others about how they engage with competitive situations, I was energized. I sat down and sketched out some key user groups targeted specifically toward differences in competitiveness, rivalries, and feelings post-tournament.

Those groups are:

The Analyst

  • doesn't take any player more or less seriously than another
  • often studies past performances
  • always focuses on personal improvement
  • driven by desire for personal ability
  • almost like playing alone in a crowd
  • not particularly social
  • mindful of their emotions and keeps them in check

The Emotional Player

  • driven mostly by a desire to WIN
  • rates players mentally as "should beat" or "tough to beat"
  • very disheartened by losses to "should beat" players
  • quick to mention past triumphs
  • very likely to feel rivalry
  • goes with the flow of their emotions

The Community Player

  • not especially competitive
  • it's not cool to be overly competitive
  • doesn't see themself as a fantastic player
  • goes to tournaments for fun + social reasons & likes to know people there
  • very engaged in the community
  • quick to mention personal achievements

I got to work modifying my app to answer each of their needs. For example:

The Analyst focuses on always leveling up. I added a graph to show cumulative points that by nature cannot ever go down. I also grouped points into a few different categories (based on participation versus based on achievements) so analysts can always feel like they're growing in some way.

The Emotional Player feels the impact of their victory or defeat based on the skill of their opponent. To answer this, I highlight when their opponent plays particularly well at a specific event, as well as showing the relative strength of their opponent based on historical data, taking this opportunity to either congratulate the emotional player when they topple a giant, or to console them with the fact that their opponent is actually really good.

The Community Player likes to put themselves alongside the other players in their scene. I added an easy way for a player to graph any friend or rival's points side-by-side with their own, showing exactly when each player went to different tournaments and how their achievements match up.

I also threw in a few quality-of-life improvements such as in-app tournament search.

What things looked like after those adjustments.
What things looked like after those adjustments.

// dev side note Respect. Your. Data. Every day I continue to learn the value of a solid data hierarchy. If I had started this project thinking more about the structure and flexible usability of my data, my entire development time would have been cut in half, and I would be able to focus better on improvements for the user, rather than on wrangling unwieldy object properties.Also, understanding proper object and array referencing as well as how to safely duplicate data can save you hours of headaches. Hours.Since I was dealing with multiple APIs at once, I ended up structuring my backend similar to how graphQL works, delegating generic requests to a set of middleware "concept" handlers (players, tournaments, etc). Those get necessary data from lower-level service-specific handlers that take care of the data fetching, parsing, and (where necessary) server-side rendering.


It's a bit of a cliffhanger to leave it here, but that's where this positivity project is at, currently. It remains a work in progress, so keep an eye out for updates coming this year!

My future goals for it include:

  • Host it live and make it accessible to anyone
  • Add smash.gg bracket compatibility
  • Interview multiple users over the span of a few tournaments to see how successful the targeted encouragement has been
  • Find new ways of analyzing data to present varying types of encouragement to players depending on their recent success/failure
  • Actually put my graphic design degree to work to make it look somewhat less awful

Thanks for reading!