9 Morgan Kurth
I joined the Data Solutions IST branch this year to diversify my experience working with data in high level hockey. I was pleasantly surprised by the breadth and depth of data available to us as analysts, and by the investment which the Women’s Hockey organization is willing to put into analytics. Another unforeseen aspect of this role has been its interdisciplinary nature. I had not previously known much about the IST program and did not anticipate the exposure I would receive to other areas of team development, including athletic therapy, strength and conditioning, sport science, equipment management, and team media administration. My Data Solutions teammates and I worked to establish a consistent and positive precedent for the role of analytics in the first year of its establishment within Women’s Hockey. I feel that our efforts contributed to the growth of the team and provided a genuinely valuable new resource for the coaching staff.
Projects
Athlete Wellness Monitoring
The concept for this project was straightforward: create an RShiny app that would allow coaches across multiple disciplines — athletic therapy, strength and conditioning, and sport coaching — to monitor the holistic well-being of their athletes and implement preventative measures before injuries occur. Realizing this vision required close collaboration and frequent meetings with coaches from each discipline to identify the key priorities and nuances of their respective fields, as each deals with a different aspect of athlete health: physical, emotional, mental, and developmental. From there, we worked alongside student IST members, who served as the primary point of contact for data collection. This brought us into morning strength and conditioning sessions to gather physical readiness data, into the training room with therapists for additional physical measurements, and into partnership with sport science students to track mental and emotional well-being for the team. My specific contribution to the Women’s Hockey app centered on early development, where I wrote R code to build a summary dashboard for physical well-being and performance, and regular coordination with coaches to collect and later incorporate their feedback.
Penalty Kill
At one point in the season, our special teams units struggled on both the power play and penalty kill. At the request of the Head Coach, I conducted a deep dive into our penalty kill performance with the goal of identifying trends that could inform improvements. To broaden my domain knowledge and ground my exploratory analysis, I also sat in on film sessions to better understand the coaches’ preferred penalty kill systems and structures. The analysis primarily involved descriptive statistics and the compilation of metrics relevant to evaluating penalty kill player combinations. I then developed graphics and collaborated with a teammate to produce and present a detailed report on season penalty kill performance. Because this was an open-ended exploratory analysis rather than a hypothesis-driven one, it was a valuable opportunity to learn firsthand what coaches prioritize when evaluating special teams and to lay the groundwork for more targeted PK modeling projects in the future.
Ad-hoc Player Reports
Throughout the season, I received individual requests from players seeking reports on their own performance analytics. Some requests were pointed and specific — a player curious about a particular metric such as plus/minus, time on ice, or shots on goal per twenty minutes — while others were broader, with players looking for general guidance on areas of improvement. As I completed these reports, I found that visual graphics conveyed information more effectively than raw numeric summaries, and so I shifted toward translating descriptive statistics into clear, interpretable visuals. An equally important dimension of this work was relational. Building trust and maintaining a professional rapport with players was essential to fostering open and productive dialogue. Their willingness to engage with analytics not only made these individual reports more meaningful but also helped strengthen the legitimacy and long-term role of Data Solutions within the IST.
Faceoff Project
At the request of the Head Coach, we dedicated time in the postseason to analyzing the faceoff performance of our competitors. Faceoffs had been a recurring theme throughout the regular season, as the coaching staff sometimes matched lines and wanted to know who was winning at the dotson any given gameday. While my teammate developed a faceoff dashboard, I produced individual team reports for the coach, shared through a mix of formal presentation and informal conversations. The focus of my later, deeper, faceoff analysis was a Bayesian inference model capable of predicting head-to-head faceoff win probability for any Waterloo player against any opponent in U Sports, drawing on data available through InStat. I incorporated this model into the faceoffs dashboard now used by the Head Coach. This model represents a step away from purely descriptive statistics, and I am confident I can meaningfully refine and expand it in the months ahead.
Data Source Exploration
Throughout the season, one of our ongoing responsibilities was organizing the wide variety of data sources available to us as analysts. As the coaching staff engaged with the university and administration, we continuously gained access to new datasets and tools. This required us to regularly adapt and expand our workflows. Much of our time was therefore spent navigating unfamiliar datasets and software, building the foundational domain knowledge necessary to use them effectively. Beyond what was immediately available, we also explored emerging technologies with the potential to enhance data collection in the future. This process of continuous learning not only deepened our technical expertise but also positioned us as a reliable resource for the coaching staff in managing the rapidly changing technological landscape of sports analytics.
Reflections
Overall, I am deeply grateful for my experience with the IST this year. It has been a privilege to work with such extensive data in U Sports hockey, and the exposure to the kinesiology side of sport was something I did not anticipate but came to genuinely value — sparking an interest I hope to continue developing. Collaborating across disciplines within the IST, in a structure that mirrors that of a professional sports organization, was also unexpected but rewarding and taught me how analytics fits into a broader team system.Though I stepped away near the end of the season to attend to a family matter, I was present for the majority of practices and games throughout the year. That time spent around the team reinforced something I now consider fundamental to this role: regular, meaningful contact with coaches and players is essential. Being present builds the trust and contextual understanding necessary to promote the role of analytics.
There are a few improvements which I believe could make next season even more successful. The holistic monitoring dashboard required significant coordination across multiple coaching disciplines, which was expected given the wide scope of the project. However, initiating those conversations earlier in the preseason would give coaches more time to provide meaningful feedback, ultimately allowing us to deliver a product better tailored to their needs. I think we should keep this in mind with future projects. Some administrative challenges this year also meant that a consistent structure for coach communication was not firmly established. Going forward, I think scheduling a dedicated and recurring meeting time with coaches would be highly beneficial. Given how busy they are, having even a brief window of their undivided attention would be far more productive than providing ad hoc updates in passing, although we should still be present and note their thoughts when they share ideas at practices. Finally, I believe that now that the Women’s Hockey analytics group is more firmly established there is an opportunity to formalize cross-team collaboration with Men’s Hockey, although I am presently unsure what the best structure might be.
Looking ahead, I anticipate improving our current dashboards to be run using server-stored data, rather than referencing CSV files. As that transition is completed and the dashboards will be largely finished, I am looking forward to doing higher-level modeling on the research and development side of the sport. I am intrigued by the idea of working with physical development data (athlete growth via SandC metrics) as well as studying player chemistry via exploration of line-specific metrics.