The holy grail of the box office industry is to predict how much a movie will make accurately.
Technology has significantly enhanced movie accessibility. Aside from the rise of streaming platforms like Netflix, HBO Max, Apple TV+, and Disney+, which have significantly changed the way movies reach audiences, there are numerous dynamic factors in predicting a movie’s financial success.
Aidan Gerber, a Rice University undergraduate student double majoring in computer science and statistics, created a model for the STAT 450 Senior Capstone Seminar to predict the daily box office gross of a film over the entire course of its time in theaters.
The quantitative approach is a robust time-series technique based on daily, weekly, and seasonal multipliers that update as new data becomes available. The model is trained on a dataset of over 3,000 movies and their box office grosses from 2015 to 2025.
Gerber’s data-driven analysis and subsequent paper, “The Gerber Method: Using Multipliers for Daily Box Office Prediction,” was selected for publication in the CoFES White Paper Series.
“The U.S. domestic box office is one of the largest, most well-reported, and transparently published revenue streams for films. Studios report daily domestic box office figures to industry outlets, such as Deadline, The Hollywood Reporter, and Variety,” said Gerber. “[Analytical] figures are also published by The Numbers and Box Office Mojo.”
Gerber is a movie aficionado. He is deeply interested in the business side of films as a top form of entertainment for people worldwide and as an art form that can reflect and shape societal values and beliefs.
For the past 9 years, Gerber has rated and kept track of the movies he has seen. His playing Fantasy Movie League in middle and high school inspired much of his capstone project.
“The fun I had competing with movie lovers evolved into my becoming proficient enough at statistics to implement the strategies I’d only dreamed of as a kid.”
Gerber’s capstone research project aims to formalize these hobbyist-inspired strategies to predict box office gross of all types of movies.
Gerber says, “Accurate box office predictions have many possible use cases: theater owners looking to decide how many screens to dedicate to a movie, studios figuring out what productions to green-light and the most profitable release dates, investors looking to buy or sell stock in a studio, gamblers hoping to place accurate bets on box office performance.”
Gerber’s model dataset includes film budgets, daily gross, reviews, production company, spoken languages, release types, production methods, theater counts, and social media analytics.
A key finding of The Gerber Method is that this strategy is effective, and the results set a new standard to build upon for predicting daily box office. Several avenues exist to expand upon the model, such as awards, influential websites that offer predictions, additional competition dynamics, and marketing strategies.
Gerber’s mentors on the project, Dr. Michael Jackson, Dr. Elizabeth McGuffey, and statistics doctoral student Arya Muralidharan, were all critical in helping him bring the paper to fruition.
Using his previous experience as manager of data analytics for BET+, Gerber said Dr. Jackson suggested many impactful data sources. Dr. McGuffey and Arya Muralidharan both provided valuable advice on creating a polished and statistically interesting final project.
Gerber will begin work as a software engineer at the molecular diagnostics company, BillionToOne, this August.
- Shawn Hutchins, Communications and Marketing Specialist
