This was a group project for my OPMA 419 - Predictive Analytics class in the Winter 2021 semester. It was made in collaboration with Shaan Gehlot and Nevin Sangha. We decided to focus on powerlifting analytics as the three of us wanted to see if we could predict our own max deadlifts based on our own statistics!
This dataset is derived from the OpenPowerlifting archive.
The OperPowerlifting archive is a volunteer-run service with a team that consists of eight people.
Their goal is to remember every lifter at every level of the sport across every federation.
It is a public-domain of powerlifting history and events.
The data set is obtained from Kaggle.com and consists of over 1.4 million unique rows and 37 columns of competitor data from the OpenPowerlifting database from April 2019. The dataset can be accessed here:
https://www.kaggle.com/open-powerlifting/powerlifting-database.
For this class we used two software:
RapidMiner Studio
Tableau
For that reason, unless you have a RapidMiner license you can not actually see our processes in the backend. I have shared all of the processes via GitHub, but for the in-depth analysis, I have attached our reports down below.
This visualization was an exploratory piece of our analysis, to visualize the relationship between a person's deadlift to their age. From this visual, we can see peak age is between 24-39
Visualization showed compared the sexes (females on left, males on right), whether they were tested for performance-enhancing drugs (No/Yes), and what equipment they used for the lifts. This shows us that the highest total (bench + squat + deadlift) on average was performed by a male weightlifter, who had not been drug tested, and used multi-ply straps to assist his weight lifting.