Key Stat: There appear to be two peaks in catch probabilities on pass interference calls, about 55% and 75%, which suggests the feasibility of a two-level foul system.Wu and Gu merged statistical modeling techniques with potential changes in penalty calls and receiver catch probability to consider new standards for defensive pass interference and defensive holding.Peter Wu, Brendon Gu, Carnegie Mellon University – DIRECT: A two-level system for defensive interference rooted in repeatability, enforceability, clarity, and transparency Key Stat: In the play above, Carr released the pass only 0.3 seconds after Holton beat his defender, suggesting the quarterback knew he was going to Holton before the receiver broke free. One example highlighted a 64-yard touchdown pass from Derek Carr to Johnny Holton to show how Holton’s speed and direction indicated that he was moving towards an open space well before he looked open on the field. Burris used metrics like player speed, direction and acceleration to chart the space occupied by the 22 players on the field.Kyle Burris, Duke University – A trajectory planning algorithm for quantifying space ownership in professional football Key Stat: The flat-in-post route, a staple of the Steve Spurrier days at the University of Florida, was the best three-receiver route combination.Once grouped, these patterns were compared to win probability added (the change in the offensive team’s chance of winning the game before and after the play). Sterken treated receiver routes as an image recognition problem, using a neural network to categorize each route.Nathan Sterken – RouteNet: a convolutional neural network for classifying routes Key Stat: Through effective pass route combinations, an offense could control roughly 70% of the field.Using a technique called model-based clustering, the group found several complementary pass route patterns that could consistently yield positive outcomes, even when accounting for defensive formation and behavior. The group modeled play success rate and expected points under various passing route combinations.Matthew Reyers, Dani Chu, Lucas Wu, James Thomson, Simon Fraser University – Routes to Success Kevin Zatloukal and Ben Gretch – Unusual effectiveness of crossing routes.Jason Phelps – Proof of concept for a receiver separation model using long short-term memory recurrent neural networks on Next Gen Player Tracking Data.

Jacques Kvam and Karl Pazdernik – Receiver-route combinations that maximize separation.

Jesse Fischer – Identifying best receiver-route combinations.Kyle Douglas, William Haynes, Momin Ghaffar, Farzad Vafaee – Missed extra punts.Zach Stuart, Brandon Deflon, University of New Mexico – Q-Sep: The NFL’s Metric for Quantifying Route Quantity.Ali Kozlu, University of Pennsylvania – A framework for identifying route concepts using route trajectory data.Dylan Blechner, Zak Koeppel, Will Friedeman, Cameron Johnson, Syracuse University – Route clustering.Mitch Kinney, University of Minnesota – Untitled.

Charles Gelman, Duke University – Identifying best receiver-route combinations.

Congratulations to the teams and individuals that earned honorable mention at the 2019 Big Data Bowl – see below for an overview of their reports. While only a handful of finalists advanced to the NFL Combine, a group of honorable mention entries impressed with their explorations into how football is played and coached. The NFL thanks all who entered the league’s inaugural Big Data Bowl.

Kaggle nfl play by play