10 Trevor Steunenberg
I joined the data solutions team because I am passionate about sports and math and wanted to gain experience doing work in a sports-related context. This was my first time doing sports-related work for an actual team; all the projects I had completed before were for my own enjoyment. I had an amazing experience this term and especially enjoyed gaining firsthand insight into how a varsity sports team operates.
Projects
Athlete Monitoring
This project was a collaboration between several data solutions members and involved creating a Shiny app to display the data collected throughout the week. IST student trainers collected survey data from players after each game and practice, asking questions like “How tired are you?”. Additionally, the IST sports science students collected data weekly from players on metrics such as their soreness, stress, and overall wellness indicators. We also collected jump data weekly from players to quantify their current physical performance level. Our dashboard presented this data so the athletic therapists could quickly check if any players were performing below their typical levels, as that player could be at an increased risk of injury. This dashboard also updates automatically, meaning we don’t need to manually update data every week.
Faceoff Dashboard
Our coaching staff emphasized faceoffs, especially as we entered the playoffs. We had hand-tracked our faceoff stats since the start of the preseason, but the coaches were also interested in analyzing our opponent’s faceoffs. To help with this, we collected faceoff data from games dating back to the 2022-23 season and then created a dashboard which displayed data on both our team and our opponents, as well as head-to-head performance between individual players. It also included filters for the faceoff side, zone, and strength (PP, PK, even strength). We initially just created it for a single opponent since that was simplest to create quickly, but after the season ended we upgraded it to allow coaches to select the opponent instead.
USports ELO Formula
Seeding at Nationals is partially based on ELO rankings. USports publishes the rankings at the start of most weeks during the regular season but does not post them during or after playoffs. The ELO formula is also not publicly available, which means that it is quite risky to scout an opponent for Nationals since nobody really knows who they are playing. We ended up figuring out the formula that USports uses, and thus what the first-round matchups will be at Nationals. This allows our coaches to get a head start on scouting our opponent.
Reflections
I had a blast this year. I really enjoyed the experience and I think a large part of that was that I tried to be pretty present. For example, I went to at least two practices every week which helped the players know who I was and allowed the coaches to see that I cared about the team and was committed. In that sense, I think a large part of what you will get out of this experience is determined by how much you put into it.
It took me a month or two to figure out how to work with the coaches and other data solutions members on my team. I don’t know if this is completely unavoidable, but I think it could be reduced by being proactive about building things before coaches/trainers start asking for things.
I am a strong believer that the data solutions members of the men’s and women’s teams for the same sport (e.g.: Men’s Hockey and Women’s Hockey) should work closely since a lot of what one team is doing can probably be done by the other team with minimal work.
There are some offseason projects that I think I will be interesting. I have some faceoff-related questions that I want to answer and I also want to get involved in the player-recruitment process. There is a surprisingly high (to me, anyway) amount of data available on high school players and I think I can do some cool stuff with it.