The pandemic forced most people to spend more time at home. Instead of seeing this as an obstacle, I saw it as an opportunity. When else would I have a chance to spend this much time working on projects and learning new skills to add to my resumé? I joined IUPUI’s Honors College this semester, which required me to complete an honors project in two of my courses. Instead of viewing the extra work as a hindrance, I saw it as another opportunity to add to my resumé.
As a student in Dr. Pierce’s Selling in Sport class, I wanted to tailor my honors project to my interests, while still exploring sales in sports. I began watching the UFC during the pandemic, as they were one of the first organizations to return to play. As a new viewer of the sport, I became curious about why some pay-per-view (PPV) events outsold others. In order to answer this question, I knew I would have to figure out which PPV event variables had the strongest correlation to PPV buys.
I collected available information for 201 previous UFC PPV events including 21 total variables, such as the fighters in the main event, whether the main event was a title fight, and the fighters’ win percentages. I also created a metric for demand for PPV events by dividing the PPV buys by the television ratings for the event. This takes the percentage of fans who are aware of and interested in the event enough to watch the free preliminary fights (TV ratings), who are also invested enough to buy the PPV (PPV buys).
Additionally, I proposed a dynamic price structure for UFC PPV events seen in the video above. In this proposal, events projected to have a lower demand would be priced lower, and events projected to have a higher demand would be priced higher. The goal of the new price structure would be to entice more fans to purchase a lower-demand event because of the cheaper price, as well as generate more revenue from a more expensive, higher-demand event. Overall, following the proposal would result in an increased revenue generated by these events.
My colleague Jack Gray used a multiple linear regression test to determine which of these variables were significant predictors of PPV buys, and I used Tableau’s visualization capabilities to test independent variables against demand using linear regressions. The number of former or current UFC champions fighting in the main event emerged as the strongest predictor of PPV buys, and the number of bouts (fights) on the main card was the strongest predictor of demand.
Click the picture below to engage with the interactive Tableau dashboard I created:
The research that went into this project was extensive, but rather than being discouraged by the amount of work, I became more motivated by each new finding. This was only my first experience in sports analytics, but it reassured me that I was pursuing the right career. The opportunities to collaborate with other students studying sports analytics for these types of projects wouldn’t be possible without IUPUI and the Sports Innovation Institute.