2016 INFORMS Annual Meeting Program

TC51

INFORMS Nashville – 2016

TC49

3 - A Handicap System For Tennis Timothy Chan, University of Toronto, Toronto, ON, Canada, tcychan@mie.utoronto.ca, Raghav Singal Handicap systems are used in many sports and games to improve competitive balance and equalize the probability of winning a match between two or more players. In this paper, we develop a handicap system for tennis using a MDP model to quantify the appropriate handicap between two players of unequal ability. We apply the model to real match data to estimate professional handicaps. We also demonstrate how our handicap can be mapped to a tennis rating system, which can facilitate broader uptake at a grassroots level. 4 - Does The Current Golf Handicap System Bias Match Play Outcomes? Martin L Puterman, Professor Emeritus, University of British Columbia, Sauder School of Business, 2053 Main Mall, Vancouver, BC, V6T 1Z2, Canada, martin.puterman@sauder.ubc.ca, Timothy Chan, David Madras In amateur golf, lower handicap players “give strokes” to higher handicap players on the basis of the difference between handicaps so as to make head-to-head matches fair. In a match, the “standard way” to allocate handicap strokes is on the basis of the course hole difficulty ranking. Using a data driven simulation based on over 600 rounds, we show that standard stroke allocation and hole rankings favor the superior player. We investigate the simultaneous impact on match fairness of alternative hole rankings, allocations of handicap strokes to specific holes and awarding extra strokes to higher handicap players. 213-MCC Modeling Complex Systems in Education Sponsored: Public Sector OR Sponsored Session Chair: Roxanne Moore, Georgia Institute of Technology, 1, Atlanta, GA, 30333, United States, roxanne.moore@gatech.edu 1 - Applying System Dynamics Models To Steam Interventions Michael Helms, Georgia Tech, michael.helms@gatech.edu Education researchers implement education interventions in highly complex systems, where intervention outcomes depend heavily on the system attributes, actors and interactions over time. We developed a modeling process using system dynamics principles to better understand the complex interactions and mechanisms of a STEAM education intervention. In this presentation we will discuss the resulting causal loop diagram, and the implications of our modeling process in terms of intervention design, implementation and sustainability. 2 - Education as a Complex System: Opportunities, Challenges, and Perspectives Mirsad Hadzikadic, University of North Carolina, Charlotte, NC, United States, mirsad@uncc.edu Education is a large complex systems. It presents many challenges for its full understanding. However, if done well, it can bring tremendous benefits to people, society, and industry. It can be simulated and modeled from different perspectives, including the content of education, the modality of teaching/learning, improving access to education, school assignments, classroom assignmnents, seating assignments, within classroom interaction, student-student interaction, or student-teacher interaction. This talk will present an overview of various technologies to analyze/simulate/model the previously stated perspectives. 3 - Modeling Interventions In Complex Educational Systems Britte H Cheng, SRI International, Menlo Park, AL, United States, britte.cheng@sri.com The ability of complex social systems to resist policy mandates—to revert to the norm—has gained increased attention in educational research. This paper presents a perspective on the reasons behind the policy resistance of education systems including the lack of approaches for evaluating possible policies before implementation. We highlight two projects (one around STEM retention in undergraduate and one on K-12 formative assessment systems) that modeled educational systems to explore proposed policy implementations before they are put into practice. TC51

211-MCC Experiential Field Learning Sponsored: Education (INFORMED) Sponsored Session Chair: Benjamin Grannan, Virginia Military Institute, Lexington, VA, United States, grannanbc@vmi.edu 1 - A Survey Of Issues And Best Practices In Field-based Education Michael F Gorman, University of Dayton, michael.gorman@udayton.edu There is a growing literature on field-based education in OR/MS. I survey recent literature and share common benefits, issues and best practices from published works. 2 - A Toolkit To Facilitate Quality Work In Field Courses Patrick S. Noonan, Emory University, Atlanta, GA, United States, patrick.noonan@emory.edu Two beliefs about field courses are widely shared: 1. Good field experiences can greatly boost student ability to apply course concepts & tools to real-world challenges ahead. 2. Facilitating “good field experiences” is hard... for everyone. Tackling real-world problems is inherently difficult - that’s the point! But one central problem that educators can address is that few participants - students, faculty, TAs - have been trained in the consultative process. Adapting the toolkit of management consultants can improve the learning experience and the results for clients/problem-owners, and ease the teaching burden. We share experience & techniques from Emory’s “Management Practice” program. 3 - Supervising Undergraduate Field Based Analytics Projects Benjamin Grannan, Virginia Military Institute, grannanbc@vmi.edu Supervising undergraduate students working on real world analytics projects results in unique challenges. In this talk we discuss the experience of recruiting undergraduate students from multiple disciplines, selling the value of analytics to clients and helping students transition their classroom skills to the often messy real world project setting. Examples from both a summer five week field based independent research course and an extracurricular student-led analytics consulting club are presented. Chair: Stephen Hill, University of North Carolina Wilmington, 601 South College Road, Wilmington, NC, 28403-5611, United States, hills@uncw.edu 1 - Do Pitchers Differ on Batting Average on Balls In Play or Defense Independent Slugging Percentage? Matthew Hall, University of Alabama, Culverhouse College of Commerce and Business Administration, Tuscaloosa, AL, 35487- 0226, United States, mjhall5@crimson.ua.edu, James Cochran Using ANOVA and MANOVA, we systematically examine data for pitchers who have pitched at least 160 innings in five or more seasons since 1999 to assess whether they differ on Batting Average on Balls in Play (BABIP - common Sabermetric wisdom says they do not) or Defense Independent Slugging Percentage (DISP). 2 - Are You Ready For Some Football? An Analysis Of Monday Night Football Viewership Bhupesh Shetty, University of Iowa, bhupesh-shetty@uiowa.edu Analyzing data from 1993 to 2014, we conduct a three-pronged analysis of Monday Night Football (MNF) viewership. First, we present a model using ex post facto factors that explains over 90% of the variability. Then we present a model using only factors known in the April preceding the season (when the NFL schedule is announced). Finally, we use the predictive regression model to estimate objective function coefficients in an integer program formulation that maximizes expected MNF viewership. We conduct simulation experiments to determine the impact of forecast error on the optimal MNF schedule. TC50 212-MCC SpORts: Sports Analytics I Sponsored: SpORts Sponsored Session

319