Last year, I introduced a new framework for understanding wide receiver play, quarterback decision-making and offensive potential. This framework relies on an XGBOOST model and PFF’s impressive collection of route-level data. Using machine learning and this breadth of PFF data, we can create models with the goal of predicting where a target should go on a given play.
The resulting metrics, Share of Predicted Targets and Share of Predicted Air Yards, are both more stable than their “actual” counterparts.
Divisional Round Recap
- Courtland Sutton: 9 targets, 22.50% target share
- Jauan Jennings: 5 targets, 19.23% target share
- Colby Parkinson: 7 targets, 17.50% target share
- TroyFranklin: 0 targets (injured after 13 offensive snaps)
Another solid showing from the model in the divisional round. All three players who logged more than 13 snaps earned a higher target share than they did in the regular season.
Potential Breakouts: Conference Championships
These are players who were open far more often than they were targeted in the divisional round. In general, players who show up on this list see an uptick in targets per route run and target share relative to both themselves and all players with similar target shares.
Top Divisional Round Predicted EPA Contributors
Using the predicted EPA model, originally explained back in Week 12, we can look at which receivers could have contributed the most to their teams whether they were targeted or not.
Jayden Higgins led all receivers by a wide margin. Texans wideouts totaled 123 receiving yards against the Patriots, with 59 from Higgins. He generated 4.02 EPA, which was only 27% of his potential EPA per the model.
After TroyFranklin and Pat Bryant exited in the first quarter, Courtland Sutton stepped up and helped the Denver Broncos beat the Buffalo Bills. The next-closest Bronco in predicted EPA was Franklin, who played just 13 snaps. Without Sutton’s contribution, the Bills might’ve won this game.
Divisional Round “Coach, I Was Open” Review
WR Puka Nacua, Los Angeles Rams
In a wild Rams–Bears divisional playoff game, Los Angeles had the ball with 39 seconds left in the first half, down 3, hoping for at least a field goal. Nacua ran a post–seam against Cover 3 match, typically favorable for the defense, and still earned a +1 play-level PFF grade by blowing past his coverage.
Matthew Stafford and the Rams maintained a clean pocket that ended with a checkdown to Kyren Williams for a gain of four yards. The play took 9 seconds off the clock, and a timeout was called after.
Given the weather conditions, this was a difficult throw, but it’s something Stafford has done many times throughout his career. Nacua’s predicted EPA was 1.94, the highest of the weekend, but his resulting EPA was −0.38. That gap was one of the largest in the divisional round.
WR Jayden Higgins, Houston Texans
In the 1st quarter on a third-and-6 while down by 7, the Texans had an opportunity to tie the game.
C.J. Stroud had a remarkably clean pocket while looking in the direction of Jayden Higgins, who was wide open for a touchdown. His predicted EPA on this play was 1.40, one of the highest of the week.
Stroud ended up holding the ball until pressure arrived and effectively threw the ball away in the direction of Woody Marks (-0.25 predicted EPA). The pass was incomplete, and the Texans settled for a field goal.
It is unclear why Stroud did not throw it to Higgins, but this single play was a 4-point swing in the Patriots’ favor. This snap may not have swung the outcome of the game, but 4 more points in the 1st quarter would have had ripple effects on the entire game.