Thesis or dissertation
Date of this Version
We define what it means for an National Hockey League (NHL) team to have been successful in drafting, and use this retrospective framework to determine if any teams exhibited a sustained edge in drafting from 2000 to 2009. At a high level, we compare actual draft outcomes to teams’ perfect draft outcomes. The perfect draft can be thought of as what would happen if a general manager (GM) could redo a draft with complete knowledge of prospects’ career values as well as other teams’ choices. A prospect’s career value is defined here as aggregate Point Shares (from hockey-reference.com ) through the 2015-16 NHL season, though any valuation metric can be used. When drafting perfectly, a GM picks the best player available as is commonly prescribed, but they start with their last pick and work backwards. This subtle but important distinction alludes to the game theoretic considerations of the draft. Simply put, the best time to pick a good prospect is just before any opponents do. Every year, each team has a different perfect draft value, which is the aggregate career value of their optimal picks. Having more and better picks means more potential value available and a higher perfect draft value. By calculating the percentage of their perfect draft value that a team actually extracts, we now have a simple draft efficiency metric that shows how teams are drafting relative to their best case scenario. This allows us to rank and compare teams equitably by eliminating some of the advantages of having high-value picks.
National Hockey League, amateur draft, game theory
Nandakumar, N. (2017). "What Does It Mean To Draft Perfectly? An Evaluation Of Draft Strategy In The National Hockey League," Joseph Wharton Scholars. Available at http://repository.upenn.edu/joseph_wharton_scholars/26
Date Posted: 14 September 2017