UC Santa Cruz undergraduate students and siblings Guy Haiby (B.S., Computer Game Design) and Roee Haiby (B.S., Computer Science) received an honorable mention in the competitive National Football League (NFL) Big Data Bowl undergraduate category for their project Decoding Audibles: Leveraging Pre-Snap Signals. The pair was recognized out of 7,000 teams for their innovative approach to analyzing football player movements.
The NFL Big Data Bowl is an annual data science challenge where participants analyze a dataset of past game measurements and statistics. This year’s competition focused on how pre-snap patterns relate to post-snap actions.
The Haiby brothers credit their Baskin Engineering coursework for equipping them with the skills needed for the competition. Through CSE 40: Machine Learning Basics: Data Analysis and Empirical Methods, taught by Distinguished Professor and Jack Baskin Chair in Computer Engineering Lise Getoor, they gained hands-on experience in data science and machine learning—skills they applied to analyze player movement in their project.
In football, pre-snap refers to the movements before a play begins, when teams line up and prepare their strategy. The post-snap phase occurs after the ball is in motion, when players initiate the play. Quarterbacks—players who initiate a play—often verbally call out last-second adjustments before the snap, known as audibles, to adapt to the defensive formation.
The Haibys’ project leveraged quarterback tracking data to provide insights into when and what types of audibles an opponent’s offense is likely to call. Their approach allowed teams to analyze play adjustments in real time and evaluate the effectiveness of their own audible usage.
Unlike other submissions, the pair’s project successfully identified audible patterns despite the dataset lacking explicit audible-related features. To achieve this, they developed an algorithm that detects audibles based on the quarterback’s body language, allowing teams to anticipate audible tendencies and gain a strategic advantage.
The Haiby brothers were drawn to the NFL Big Data Bowl through their shared passion for football and data science.
“Guy played football in high school and always enjoyed watching NFL games,” Roee said. “When I heard about the competition from my brother, I knew that we would enjoy working on this project and that it would challenge us to improve and apply our data science skills.”
Their experience in the competition highlighted the practical applications of their coursework.
“I enjoyed applying the skills we developed at UCSC to solve real-world problems, especially in the sports world,” Roee said.
Reflecting on their experience, the pair emphasized the importance of persistence.
“Everything we did was through trial and error, the key was committing to the project and learning from each mistake,” Guy said. “My advice to other students is to give it your best shot and do not be afraid to fail.”
The Haiby brothers will graduate in spring 2025 and plan to build on this success through their company Blitz, which uses an AI-powered tool combining game film and 3D simulations to enhance football player development.
“When I was studying game design at UCSC, I learned how to make 3D projects, which made me think about ways it can be useful for football teams,” Guy said. “We learned what technologies could be useful for the industry and we are aiming to provide them with Blitz.”
This block group hides your featured image, remove this block group to show your featured image again.
