.png)
In this session, we will explore roster construction through mixed-integer optimization.
In this session, we will explore roster construction through mixed-integer optimization. Using individual NBA player characteristics, we will formulate an optimization problem that constructs the ideal national basketball team subject to various constraints on team composition. This example will introduce a general framework for how to formalize various strategic problems (e.g. scheduling, pricing, staffing) as optimization models. These models will be implemented using JuMP in the Julia programming language. No prior experience with Julia is necessary.
If you are interested in this session, please take a look at Holly’s GitHub to download relevant code/data.
A global pandemic brought the shutdown of in-person live events around the world and a rise in the consumption of digital content. Digital content platforms like Netflix are shifting fan’s expectations to more personalized, customized, automated, and relevant experiences. Leagues, teams, and venues can utilize fan data to meet those same expectations.
"… all models are wrong; the practical question is how wrong do they have to be to not be useful.” (George Box) Athletic departments at schools with under-enrollment have been caught in a bad-model trap, where the NCAA's accounting conventions lead them to make harmful economic decisions in the false belief they are cutting costs. Andy Schwarz (PCL, OSKR) works through why the accounting gets the economics so wrong, provides detailed examples from schools that have made mistaken cuts in 2020, and outlines a solution to avoid these errors going forward.
.svg){kind=small}
