SpringSim-ANSS 2019, April 29-May 2, Tucson, AZ, USA; ©2019 Society for Modeling and Simulation (SCS) International
DRAFTING AGENT-BASED MODELING INTO BASKETBALL ANALYTICS
Matthew Oldham
Andrew T. Crooks
Department of Computational and Data Sciences
George Mason University
4400 University Drive
Fairfax, VA 22030, USA
moldham@gmu.edu
ABSTRACT
The growth of sports analytics (SA) has raised numerous research topics across a variety of sports, including
basketball. Agent-based modeling (ABM) has great potential to assist and inform SA, but to date it has not
been utilized. To support the use of ABM in SA, a model of a basketball game, which considers most
fundamentals of play, is presented. Additionally, player behavior is partially predicated on assessing the
length of a player’s shooting streak (testing the “hot-hand” effect) and the consideration a team gives to a
streak and their franchise player. The model’s output is used to calibrate and validate it against statistics
from the National Basketball Association (NBA). Via a set of experiments, the model indicates that an
increased belief in the franchise player leads to increased scoring action, but a belief in the hot-hand a minor
effect. Thereby, demonstrating the utility of ABM to SA, thus opening a new research field.
Keywords: agent-based modeling, sports analytics, hot-hand effect
1 INTRODUCTION
The burgeoning field of sports analytics (SA) has presented researchers from varying disciplines a rich
stream of topics to investigate, as detailed in Section 2. The increased availability of high-fidelity data has
fed the growth of SA, with the data utilized in unison with behavioral theories to provide insights into
understanding and improving performance. SA, initial success came in baseball, before spreading to other
sports, including football, golf, motor racing, and basketball. In basketball, numerous areas of research have
arisen including: at the macro level the distribution of scoring activity is a mixture of random walk processes
and power-law behavior, and at the individual level players often exhibit cognitive biases. An example of
a bias is whether players consider a hot-hand effect and their reaction to such a phenomena. As detailed in
Section 2.3, the hot-hand effect refers to the behavior of people who erroneously consider the length of
successive success in the past when forming expectations regarding the probability of future success.
As the field of SA expands, it has become apparent that the application of innovative approaches is required.
This point is made by Merritt and Clauset (2014) when they indicate that while game theory can help
explain dynamics in some circumstances, it is of limited value in assessing complex games. One method
that has great potential to assist and inform those engaged in SA, but which to date has not been utilized, is
agent-based modeling (ABM). The advantage of creating agent-based models is that it allows researchers
to assess, in silico, the micro-level interactions that give rise to verifiable macro outcomes. This outcome
is achieved by employing a bottom-up modeling approach, where heterogeneous interactive agents make
decisions, adapt and evolve to meet the requirements of their environment.