I’ve always enjoyed watching sports, especially events like the Olympics and the World Cup. I loved the energy of cheering for teams, the excitement of close games, and was personally motivated by thinking about how much effort and dedication athletes put into their performance. After moving to the United States, I became even more fascinated by sports culture. People here are deeply passionate about a lot of different sports, including college sports, and identify strongly as fans of specific teams. Experiencing that environment in person was really fun, and I naturally became part of it.

Discovering a New Kind of Fandom

What started as cheering along with friends gradually turned into becoming a dedicated fan of the college football team myself. As I learned more about football through games, YouTube, and documentaries like Quarterback, I began to notice how much strategy goes into every play. Teams make decisions based on specific situations, and many of those decisions are supported by data and analytics. At first, it felt complicated, but that just made me want to understand more about how teams actually use data to make decisions and plan strategies.

That curiosity eventually led me to join an analytics project supporting the Duke softball team. The project is part of a collaboration between the MQM program and Duke Athletics, led by Professor Ryan Burk. Through this collaboration, student teams work on real analytics problems for Duke sports programs, gaining hands-on experience while supporting the teams’ needs. Students applied to join the project, and a small group was selected to work directly with the Duke softball team this year.

Learning About Softball Analytics

When I first heard about the Duke softball project, I was immediately interested. Given my growing curiosity about how data is used in sports, the idea of applying analytics to a real team felt especially meaningful. Using data to help a real sports team make decisions also felt very different from typical class assignments. Instead of analyzing data just for a report or homework, I would be building something that coaches could actually use.

The team already had a large dataset tracking players’ performances. This data recorded many details about each at-bat and allowed coaches to measure important performance metrics, such as how well players perform in certain game situations. However, to make this information more useful, the insights needed to be easily accessible.

At the beginning of the project, I didn’t know much about softball analytics, so I spent time studying the dataset carefully and documenting how each metric worked. Understanding the logic behind the numbers was important to ensure that the dashboard calculations matched what the coaching staff already trusted and used.

Piecing Together a Prototype

After I learned about the metrics themselves, I built a simple prototype dashboard using Plotly and Dash before our first meeting. The goal was not to build the final product yet, but to see whether we could turn the spreadsheet data into something more interactive and easier to explore.

After the prototype, our team worked together to expand the system based on the first dashboard that I built. One of the key improvements was organizing the dashboard in a way that allowed coaches to quickly find the information that mattered most to them. For example, we created side-by-side views that allowed coaches to compare practice and game performance, and added filters so they could explore the data in more useful ways.

Putting Our Client First

One of the biggest lessons I learned from this experience was that building analytics tools is not only about writing code or calculating metrics. It is about understanding the people who will look at the dashboard and how they make decisions.

Through conversations and feedback from the softball team, I learned how coaches think about the game and what kinds of comparisons actually help them make decisions. For example, coaches often want to know things like which players perform best in certain game situations, how a player’s performance changes over time, and how players compare against each other in key metrics.

Understanding these questions helped my team understand how coaches would use the dashboard, which allowed us to make the system more intuitive and useful. This process taught me how to translate raw data into insights that stakeholders can quickly understand and act on.

Advancing My Skills

Looking back, this project taught me two important things about data analytics.

1. Data becomes most valuable when it helps people make better decisions in real situations. Before, I focused mainly on analyzing data and producing results.
2. Building something useful requires more than technical skills. It requires understanding the users, how they think, what they need, and what decisions they are trying to make with these insights. Watching the dashboard evolve based on feedback from the coaching staff showed me how important communication and iteration are.

This experience made me more confident in applying data analytics to real-world problems, and it reinforced my interest in building tools that can have a meaningful impact.