Every fall, the NFL throws open its most valuable trove of data the player-tracking chips embedded in shoulder pads and footballs to anyone with a laptop and a knack for machine learning. The result is the NFL Big Data Bowl, a crowdsourced competition that has grown from a niche analytics experiment into one of the league's most important talent pipelines.
What the Big Data Bowl Is
The Big Data Bowl calls on professional and aspiring amateur data scientists to devise innovative, data driven approaches to a specific football challenge each year. It's powered by Amazon Web Services, and contestants work with traditional football stats alongside Next Gen Stats (NGS) the league's player tracking data to analyze trends, evaluate player performance, and help advance how the game is played and coached.
The competition has been hosted on Kaggle, the world's largest community of machine learning practitioners, since 2019. That platform choice matters: it means the contest runs on public leaderboards, shared notebooks, and open discussion threads, so the whole analytics community not just the eventual winners benefits from the ideas that get tested along the way.
How It Works
The call for participants typically goes out in the fall, with the competition running into early January. Entrants can work solo or in teams, and choose between tracks — including a University track, open only to groups or individuals made up entirely of undergraduate or graduate students, and a Broadcast Visualization track, where the goal is to produce the best animation, video, or chart depicting player movement.
Each edition centers on a fresh problem. Past competitions have dug into running backs, defensive backs, special teams, pass rush plays and tackling, and pre-snap tendencies challenging participants to spot the patterns both offenses and defenses give away before the ball is even snapped. Several of these projects have generated metrics that NFL teams have gone on to use in actual games.
This Year's Challenge
The 2026 competition the eighth annual Big Data Bowl asked applicants to use Next Gen Stats to generate insights around a new part of the play: what happens while the ball is in the air. For the first time, participants had to predict player movement using only data available before the throw, forecasting where players would move once the football left the quarterback's hand. Training data came from the 2023 and 2024 seasons, with predictions evaluated against real outcomes from weeks 14–18 of the season in progress. Signups opened September 25, 2025, with a shared prize pool of $100,000 and the chance to present to NFL teams at the Scouting Combine.
For the first time, applicants could also enter a public leaderboard that scored submissions by comparing predicted player locations to actual ones, adding a live, competitive layer on top of the usual proposal based judging.
The NFL also builds mentorship into the process. The program pairs experienced NFL analytics staff with newcomers through individual meetings and monthly group training sessions, culminating in a virtual forum where mentees present to analysts from all 32 teams.
The 2026 Winner
Lucca Ferraz, a Rice University student, won the 2026 Big Data Bowl with a project called Ghostbusters: Back off man, I'm a Data Scientist! His submission evaluated defender movement while the ball was in the air by modeling hypothetical distributions of "ghost defenders."
According to Rice University's own coverage, Ferraz competed independently and took the top prize for a defensive framework built around a new metric for measuring pass coverage impact. Using the tracking data collected from chips in players' shoulder pads and the football itself, his model captured how defenders affect passing plays even when they don't rack up a traditional stat line like an interception or pass breakup. He presented the work during Combine week in Indianapolis, in front of league executives, scouts, and analytics directors the kind of exposure that has made the Big Data Bowl one of the NFL's most visible talent pipelines, with some past projects later appearing in national broadcasts.
Why It Matters
The Big Data Bowl sits at an interesting intersection: it's a research competition, a scouting tool for NFL front offices, and a genuine pipeline into professional sports analytics jobs Rice's coverage notes a previous Big Data Bowl standout went on to become a quantitative analyst for the Philadelphia Eagles. For students and independent analysts, it's a rare shot at working with real, high-resolution NFL tracking data and having that work seen by the league itself. For the NFL, it's a low cost way to crowdsource ideas that its own analytics departments might not have time to chase several winning concepts have gone from a Kaggle notebook to an actual in game metric.
If you want to dig into the data yourself, the 2026 competition pages remain up on Kaggle under "NFL Big Data Bowl 2026 Analytics" and "NFL Big Data Bowl 2026 Prediction," and the NFL Football Operations site tracks each year's finalists and winning submissions.