Hi, I’m Lexie. I’m a third year Mathematics student studying at the University of York. I’ve always had a passion for sports statistics, so when the opportunity came along to take part in a paid internship based at the University, I was very excited.
This past summer, I completed ten weeks of independent research focusing on data analysis in sport, supervised by two of my lecturers. I really enjoyed the experience of learning how different statistical ideas can be used in the context of sport.
My research subject
I took part in a national competition based around predicting the Paris Olympics gold medal table. Every year lecturers at York host a sport themed event and students are encouraged to take part. It’s a great way to become familiar with working with large-scale applied problems.
How did I approach my research?
I decided to approach the competition mainly using statistical knowledge taught at undergraduate level as I was keen to explore my courses in more detail. I was also intrigued to see how well uni degree methods would compare to the more complex methods that an expert in the field may produce.
My research methods
Some methods I applied throughout my prediction entry included k-means clustering and linear models. K-means clustering is a technique used to find unknown patterns in data. Using data from all of the countries that take part in the Olympics I could find groups of countries that had similar medal counts. This helped to suggest which nations would be placed close together in the final medal table.
In contrast, linear models involve predicting future results through the use of a modelled equation. I used this technique when dealing with countries in the same cluster. Here, I added variables such as host nation, GDP and number of athletes a nation takes to a particular games into the equation as I felt that they would impact how many gold medals a country would earn. However, some factors cannot be measured and having a knowledge of a particular sport is vital when looking at sports analysis.
In my case, I love athletics and so I chose to really focus on this. Especially as several countries are known to only get medals in this event. For example, I predicted the Dominican Republic to win a gold even though they had never won a gold medal at the Olympic games before.
What I gained from my internship
This project was a fun opportunity to look beyond my degree by expanding my mathematical knowledge. As well as being an excuse to watch as much sport as possible!
I love that the lecturers here at York show a genuine enthusiasm towards students’ interests whatever they may be and are happy to support you to reach career aspirations. Through completing my internship with the University, I learned many valuable skills that will be essential when finding jobs in statistics in the future, such as working with real life data and learning to proficiently code in the programming language R.
Whether you love sport like me or have very different interests, there is so much to get involved in at the University of York.
Leave a Reply