This post is a continuation from my previous post Can Fencers Bounce Back From Being Cut After Pools?. As I prepared that work to be presented at the CMU Sports Analytics Conference and the Midwest Sports Analytics Meeting, I made some small changes and even found some new results, which I’m excited to share in my poster below (click on it to expand):
A major change I made was switching the outcome variable (measure of performance) from percentile in event to indicator (points scored minus points received). I felt that change in percentile wasn’t easily interpretable, whereas change in number of points scored was. I could have also changed this metric to number of wins, but I felt like it wasn’t granular enough. Additionally, I realized I had enough data to look at the effect of being cut on Y-14 fencers fencing Cadet and Cadet fencers fencing Junior, so I could see how the effects varied across age groups! The results, however, remained the same. Finally, I calculated the Minimum Detectable Effect (MDE), which allowed me to rule out effects larger than 3 points in indicator, meaning that if there is an effect, it is likely to be very small.
