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.
2025 Regular-Season Recap
Over the course of the season, we had 87 players show up as potential breakouts in our model. Those 87 players tallied a 20.8% target share, which was, on average, 3.0 percentage points higher than a random control group of players with at least 15 routes. A player’s target ceiling was also slightly improved (+1.5 percentage points) as well as their targets per route run (+1.7 percentage points).
As a reminder, the model excludes certain routes and plays for computational reasons. For example, pass interferences and most manufactured targets are removed no matter the result.In general, across any season, this model is capable of identifying players who will earn more targets in Week N+1.
Potential Breakouts: Wild-Card Round
These are players who were open far more often than they were targeted in Week 18. Overall, 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. If a team was resting starters in Week 18, that should be taken into strong consideration.
Josh Whyle and Christian Kirk both had a small increase in routes for Week 18, and neither has seen high route participation all season. I would be cognizant of this in any decision-making using this model.
On the other hand, Ladd McConkey and Josh Palmer have experienced significant participation this year and could very well see a boost in targets for their respective playoff games this week.
2025 Regular Season “Coach, I Was Open” Review
For this edition, I wanted to take a look back at two of the bigger “Coach, I Was Open” plays of the last few weeks via the predicted EPA model.
Week 15: WR Ja’Marr Chase, Cincinnati Bengals
Back in Week 15, the Bengals did not score a single point against the Baltimore Ravens, but they could have had the lead in the second quarter thanks to Ja’Marr Chase — had Joe Burrow seen him on this play. It was third-and-9, and the Bengals had the ball tied at 0-0. The Ravens ran a Cover 6, and the Bengals called the perfect play to gash them deep.
Andrei Iosivas and Mike Gesicki both ran “In” routes, pulling the safeties playing Cover 4 down, while Ja’Marr Chase ran a deep post route. This is often referred to as a “Pin” concept. Burrow had a perfectly clean pocket and seemed to glance at Chase, but decided not to pull the trigger and instead checked it down to Drew Sample for a seven-yard gain.
Chase had a predicted EPA of 2.65 on this play compared to Sample's -0.11 mark. The play resulted in a -0.34 EPA, and the Bengals would punt and proceed to not score for the rest of the game. The Bengals’ season was over at this point, but missed opportunities like this can have huge ramifications for both the game and across the entire league.
Week 16: WR TroyFranklin, Denver Broncos
TroyFranklin earned an incredibly rare +1.5 “play-level” PFF grade on this play. For reference, there have been 21,216 receiving snaps this season, and only 97 targets have garnered a +1.5 or better. However,Franklin did not earn a target on this play.
The Broncos were losing by 14 toward the end of the third quarter, and a score here would have been vital to their chances of coming back to win this game.
Bo Nix began the play with a clean pocket and no disruption; Franklin was running an out-and-up against Jarrian Jones. Nix drifted right and eventually began to sprint right into pressure from Emmanuel Ogbah. This was a very clean pocket for Nix to step up into, but instead he initiated pressure from the defensive end rather than avoiding it.
Given the Broncos’ No. 1-ranked team PFF pass-blocking grade, had Nix simply not moved or drifted left or center, this would have been a touchdown.
This is a great example of how all pressures are not the offensive line's fault, and how quarterbacks can miss some major opportunities by creating pressure on themselves. Per PFF Premium Stats, Bo Nix has allowed the 10th-highest rate of quarterback-assigned pressures.