The goal of this project is simply to test and learn how to embed Shiny Dashboards into a website. The specific dashboard isn’t necessarily anything particularly fancy – specifically, it is one I threw together quickly for a homework assignment for my Sports Analytics class last June – but it will be suitable to serve as a test for integrating dashboards into my site, which I’ll hopefully be able to do for more interesting DB’s going forward.
If you are curious about this one: basically the goal of this project was to use a combination of unsupervised and supervised learning to determine if TNT was getting a fair shake for a schedule of NBA games they’d get to televise based on the quality of matchup. It was pretty standard stuff, so I probably won’t make a dedicated post for it in detail, but again, will use this quick dashboard for testing purposes. The first tab illustrates the matchups based on win probability and total Elo of competing teams, color coded by a matchup tier that was determined by the analysis (1 to 4 is best to worst). This can be filtered by TV Network using the check boxes to the right. The second tab illustrates the historical Elo rating of the team selected from the dropdown menu. The third tab is a fixed chart, which I ultimately removed from my analysis due to time constraints and access to relevant data, but features some loose info about how each network might break down based on social media buzz.