ideas 2 and 3 - #15
Conversation
Update Data 400 Fall 2024 Syllabus.pdf
There was a problem hiding this comment.
This is a very practical idea. My only comment is that you should be mindful of how the results might impact player or team perceptions. For example, if your model predicts that certain players have a lower probability of achieving specific performance metrics, this could potentially affect how they’re viewed by coaches or teammates.
There was a problem hiding this comment.
This is an interesting idea. And you might want to provide more detail about how you’ll set up the simulation. What random variables will be modeled in each simulation? For example, will you introduce randomness into shooting percentages, turnovers, or other performance metrics?
added presentation
|
Great topic. I think you should definitely look into distinguishing outdoor and indoor quality. My research deals with chess data a lot, and I know papers out there using chess data to distinguish it. You can check it out: https://pubsonline.informs.org/doi/10.1287/mnsc.2022.4643 Of course you should try to find data sources that would enable you to do such merging. (You need some indoor air quality data that is publicly shared, and outdoor data is public of course) And lastly, definitely look into a finer grid of air quality data by looking into shapefiles. That way you can figure out how things are close to the big emitting sources like factories or higways and disinguish them from all potential areas that get affected by external forces like wildfires smokes that come from long distances. |
3 participants
No description provided.