Entertainment, Sports & Gaming

Zachary Anderson

VP of Global Analytics & Insights
Electronic Arts

Estimating Customer Lifetime Value in Free to Play Mobile Games

At EA, we face a unique problem in our mobile gaming sphere: how do we run efficient acquisition for over 20 different Freemium games in 100 different countries via 100 different channels when only around 5% of users will ever end up spending? To do this, the budget is optimized based on the long-term revenue forecast and the acquisition cost of an average player from each of those sources, using only a few days or weeks of observations. We disintegrate the calculation of player lifetime value into the timing of conversion to a spender and the repeated spending probabilities following a conversion. Separate generalized linear models are used to fit the two components, both with mixed effects to capture game, platform, and channel specific characteristics. The estimated results are then convoluted with realized spending of existing players to arrive at the lifetime values. This presentation will walk through these complexities, from the LTV calculation itself to the breadth of campaigns that is supported.


Zachery is the Vice President of Global Analytics and Insights at Electronic Arts (EA), the world’s largest Video Game Company. He is responsible for leading Consumer Insights, UX Research, Data Science, Studio Analytics, and Marketing Analytics for EA. His team uses in-game behavioral data, traditional consumer research, lab work, and online advertising data to provoke and inspire EA’s development and marketing teams to think and act “Player First”. Prior to joining EA in 2007, Zachery was head of consulting and modeling for J.D. Power and Associates’ PIN group, Corporate Economist for Nissan North America, and Economist for the private investment company Fremont Group.

Zachery’s work has been highlighted in the Harvard Business Review and the MIT Sloan Management Review. His work has won many awards including the INFORMS Marketing Science Practice and while at Nissan he was recognized by the US Federal Reserve for the Best Industry Forecast.

Zachery’s undergraduate degree in Political Science and Communications is from Southern Illinois University. His graduate work was at UCLA, in Economics and Political Science, where he studied game theory with Nobel Prize Winner Lloyd Shapley.

Phil Bangayan

Director, Marketing Data Science and Analytics
Universal Studios Hollywood

Theme Park Analytics: Maximizing Attendance Subject to Capacity

This talk focuses on how Universal Studios Hollywood collects and analyzes data to maximize turnstile attendance and overall revenue while staying within capacity constraints. Like other entertainment venues (e.g. concerts and sporting events), theme parks have a capacity limitation. However, theme parks have the added complexity that tickets are purchased for a range of valid dates. For example, single day tickets and Annual/Season Passes are valid until expiration with the exception of blackout dates. Furthermore, since each type of ticket has different revenue, consideration is taken with regards to price and access. To manage attendance, Universal Studios Hollywood utilizes date-specific ticketing (introduced last year), continually optimizes blackout date calendars and offers Pass Members special Bonus Days. Data from these programs as well as historical take rates and external factors are used to forecast attendance and optimize revenue while maintaining a favorable in-park experience by staying below capacity.


Phil Bangayan is Director, Marketing Data Science and Analytics at Universal Studios Hollywood. In this role, he leads the team in analysis of marketing, sales and research data to develop insights for business decisions, such as smoothing attendance, maximizing channel profitability, and optimizing membership programs. Prior to joining Universal, Phil held finance and marketing roles at Disneyland Resort, where he wrote the business plan for Mickey’s Trick or Treat Party, and engineering roles at Rockwell Scientific, where he designed speech recognition systems based on Hidden Markov Models. Phil holds an MBA from the MIT Sloan School of Management and BS and MS degrees in Electrical Engineering from UCLA.

Maarten Bos

Research Scientist
Disney Research

Personalization at Scale

Personality types vary. This suggests that tailored advertisements may be appealing and effective. In a series of studies, we investigated both the image preferences of individuals as well as responses to a personality questionnaire. Our investigation consisted of 4 parts. In part 1, we had images rated by participants whose personality we also measured. We also had each image rated for specific features, such as presence of people in the picture, whether those people looked into the camera, etc. We then used computational methods to extract image features, ranging from contrast and color and content and image composition. Combining these data, we were able to predict whether a new set of participants would like certain images. Because our techniques rely on automated classification, we are able to tailor advertisement images on any scale.


Maarten Bos is a Research Scientist at Disney Research, which is an innovation lab of The Walt Disney Company. He received his Ph.D. in behavioral science from the Radboud University in The Netherlands. Maarten started at Disney Research in 2013, after working as a post-doctoral research fellow at the Harvard Business School. He leads a group of behavioral scientists, with the mission to make people feel better and be better

Dave Schrader

Board Member
Teradata University Network

The Golden Age of Sports Analytics

This talk describes the current state of analytics for major professional sports – baseball, basketball, football (American), and soccer, as well as analytics used by trainers and strength coaches. It touches on the latest research, and shows results from several university projects underway in the USA and Germany.

Topics include:
The field of Sports Analytics is hot – what’s table stakes? What’s new?
Big Data in sports: what are the new data types, and data insight generation techniques?
How do analytics from traditional CRM marketing apply to sports business operations?
How do video and sensor technologies provide opportunities to improve team operations?
What new analytics for individual athletes as well as team play dynamics are in use?
What university sports analytics projects are underway, and how can students in stats, computer science, and math contribute to the success of their university’s sports teams?
If you want to do data science research in sports analytics, where can you find data sets and the best research? What kinds of research projects will provide the most value?


Dr. Dave Schrader retired in 2014 as a marketing director at Teradata, where he was responsible for marketing Teradata’s Big Data initiatives. Schrader helped Teradata customers understand how to use both traditional and big data to create analytical insights and predictive models for strategic and operational employees and systems. He continues to help faculty and students as a board member of the Teradata University Network, giving talks at universities and helping them with curriculum development for data analytics programs. In retirement, he’s created new teaching materials and homework assignments in the area of sports analytics for baseball, basketball, football, hockey, and soccer. Since 2015, he has given more than 47 talks at 25 universities to 2500 students, faculty, and coaches. In addition, he has helped launch 6 “Moneyball” projects, connecting business school faculty and students with their own sports teams. Schrader holds a Ph.D. in computer science from Purdue University, has published in the areas of customer management and pervasive business intelligence, and is a popular worldwide speaker at conferences on how to gain a competitive edge from using technology.


Aurelie Thiele

Associate Professor, Department of Engineering Management, Information and Systems
Southern Methodist University

Robust Player Selection Strategies for Baseball Analytics

This work investigates robust optimization and prescriptive analytics approaches to the problem of selecting free agents to sign for a Major League Baseball team. The decision maker may sign free agents for a one-year or multi-year contract or call up minor league players for one-year contracts. The goal is to maximize a performance metric combining wins above replacement and win probability added, subject to position requirements and budget constraints, as well as uncertainty on the players’ future performance, the salaries the free agents will accept, and the ultimate decision the free agents will make when presented with offers from competing teams. While baseball has been the subject of many analytics efforts starting with the Oakland Athletics in the early 2000s, optimization tools have not been used, opening teams to significant regret if their most promising and expensive hires do not meet expectations. Our methodology allows decision makers a better understanding of the players who are critical to team performance and has applications in R&D portfolio management and human resources management problems, as well as implications for robust knapsack problems. We implement our approach on the roster of the Los Angeles Angels prior to the 2013 season, using historical player performance and contract data. We show how to calibrate the model, in particular how to create the uncertainty set for players’ future performance, and compare our signing decisions with those actually made by the Los Angeles Angels.


Aurelie Thiele is an Associate Professor in the Department of Engineering Management, Information and Systems at Southern Methodist University in Dallas, TX, where she plays an active role in the Analytics research cluster bringing together SMU analytics experts and top professionals in the second-fastest-growing U.S. metropolitan area, home of companies such as American Airlines, Capital One, Southwest Airlines, AT&T and more. Her research is on decision-making under high uncertainty using robust optimization and advanced analytics. Prior to joining SMU she was an Assistant and Associate Professor at Lehigh University in Bethlehem, PA and a Visiting Associate Professor at the Sloan School of Management at MIT. She holds her MS and PhD in Electrical Engineering and Computer Science from MIT. At INFORMS she serves on the Analytics conference organizing committee and is or has been Chairperson of the O.R. and Analytics Student Team Competition, the Undergraduate O.R. Competition and the Professional Recognition Committee as well as a member of the Wagner Prize committee, among other responsibilities.