Kostya Medvedovsky joins the show to discuss DARKO, a well respected NBA player projection system. He begins with how a lawyer developed the data and math skills to develop DARKO. Then Kostya describes the two predictive systems that form the basis of DARKO, one of which was initially used in rocket science. Then he talks about how machine learning techniques combine these two factors. He uses Brandon Ingram of the New Orleans Pelicans as an example of how the system works. Finally, Kostya discusses the applications of DARKO to sports betting.