The NFL scouting combine is a rich source of data, featuring hundreds of aspiring players who participate in diverse drills, from the 40-yard dash to the vertical jump and short shuttle. For NFL executives, even the most experienced, navigating this wealth of information can be daunting. It is no less challenging for fans who dedicate months to becoming draft specialists.
The NFL’s NextGen Stats team has collaborated with Amazon QuickSight’s machine learning tool to enhance data accessibility through a feature called Combine IQ. This tool, now available on the NFL’s website, is designed to simplify and clarify the multitude of data collected during the combine. Mike Band, the senior manager for research and analytics at NFL NextGen Stats, explained, “We named it Combine IQ because we want to boost fans’ understanding of the combine.” He emphasized the team’s goal of helping fans comprehend the significance of the collected numbers, making the data both understandable and informative.
The dashboard showcases the results of all drills, incorporating tracking data from RFID sensors on each player. This data is then integrated into a player projection model, which considers college performance, physical attributes, and a consensus draft board from ten ranking sources. Players receive ratings between 50 and 99 in athleticism, production, and an overall draft score.
The NextGen Stats team has been correlating combine data since 2003, enabling fans to examine players’ potential over more than twenty years. These rankings align well with future NFL success, particularly in metrics like becoming a starting player or achieving Pro Bowl status. All data undergoes validation for accuracy by the NextGen stats team before being published on the website, typically within ten minutes after a player concludes a drill. This analysis is accessible to fans online and utilized by NFL Network during the combine broadcast.
The data can be organized to highlight top performers in different drill aspects, with spider graphs available to compare athletes in the same position group across several tests and measurements, including production and athleticism scores. Ari Entin, head of sports marketing at AWS, remarked, “These analytics were once exclusive to NFL teams, and now fans everywhere can access them, which is exciting. This is an innovative method for capturing, visualizing, and assessing athletic performance data in real time.”
For prospects who do not participate in certain drills, the model generates estimated athleticism scores using accumulated college data. The data goes beyond basic event timings like the 40-yard dash by also ranking players on top speed, 10-yard splits, acceleration, and speed at 10 yards. For instance, the dashboard reveals that Xavier Worthy set a new record last year in the 40-yard dash at 4.21 seconds, achieving the highest speed at each 10-yard segment.
The tool also provides insights into how various combine tests relate to NFL success, with athletic scores being more crucial for edge rushers and cornerbacks, while less so for centers and safeties. There are defined thresholds for achieving certain levels at specific positions, such as running backs attaining “good” with a time of 4.53 seconds or faster if under 210 pounds, or 4.58 if heavier. Aligning with these benchmarks leads to “elite” status at 4.39 for those under 210 pounds and 4.42 for those weighing more.
The tests have specific relevance depending on the position. For example, three-cone drills are important for running backs and edge rushers, and the vertical jump is crucial for pass catchers. Larger athletes, like offensive tackles, must meet criteria in the 40-yard dash, three-cone drills, and arm length based on decades of draft learning.
Band elaborated, “Our model is adept at identifying thresholds. If a player exceeds a certain rank or grade, the predictive analytics align below that level. Our intention is for our big board to somewhat mirror the consensus big board, reconciling any ties through advanced metrics like athleticism, production, and size.”
