|
Activity Number:
|
413
- Section on Statistics in Sports Cpapers
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics in Sports
|
|
Abstract #330750
|
Presentation
|
|
Title:
|
Recreating Plays - Testing Shot Policies in Basketball Using Non-Stationary Markov Decision Processes
|
|
Author(s):
|
Nathan Sandholtz* and Luke Bornn
|
|
Companies:
|
Simon Fraser University and Sacramento Kings and Simon Fraser University
|
|
Keywords:
|
Markov Decision Process;
Simulation;
Sports;
STAN;
spatio-temporal
|
|
Abstract:
|
Last year, the Cleveland Cavaliers took 329 contested mid-range jump shots with over 10 seconds remaining on the shot clock. What could've happened if they had taken these shots 20% less frequently over the season? We attempt to answer these types of questions by modeling plays from the 2015-2016 NBA regular season as Markov chains realized from team-specific Markov decision processes. To account for the dynamic nature of a basketball play over the shot clock, we model the transition probabilities as a tensor exhibiting correlation in time. The draws from the transition probability tensor posterior distribution then serve as inputs in our regular season simulator for the 2015-16 NBA regular season. We validate our simulation method by showing that we accurately recover the 2015-16 transition counts for all intermediary and terminal states when simulating under each team's observed shot policy. To culminate, we simulate seasons under "altered" shot policies proposed within the basketball analytics community and explore the net changes in production under these alternative shot policies.
|
Authors who are presenting talks have a * after their name.
