On a passing play that does not involve a penalty, there are two ways to generate yards: through the air and after the catch. Generating yards after the catch is conditional on the fact that the pass through the air was completed and the receiver of the pass can attempt to move their team further down the field. Over time, a quarterback’s average depth of target has become synonymous with a quarterback trying to be aggressive while yards after the catch have become linked to a receiver’s ability to make plays in space. Recently, however, both of those links to their respective positions might be weaker than previously accepted.
Over the past couple of seasons, the way NFL passing offenses have generated their production has changed to match the schematic shift of defenses. Using expected points added (EPA) – a statistic that determines what the expected points of an offensive drive will be based on yard line, down, distance, and time remaining – we can see how the average EPA through the air and after the catch has changed per season. Yards after catch (YAC) EPA per dropback has continually risen while Air EPA has declined (something that Robby Greerre has researched as well). This makes schematic sense as the rise of the Vic Fangio two-high safety structure predicated on taking away deep passes has continually risen through the league with disciples such as Brandon Staley spreading the defense across teams. Because of this, the ability to generate yards after the catch has become more important and may only become more valuable in football.
In the 2022 regular season, the three teams with the highest EPA per pass attempt were the Kansas City Chiefs, San Francisco 49ers, and Detroit Lions. Those three teams also had the highest yards after catch per dropback in the league. As mentioned earlier, defenses are changing the way they are being played, and offenses are learning to take what is given to them. This led those three high powered passing offenses to take advantage of high probability passes and getting yards after the catch. While air yards and yards per catch are negatively correlated, it is clear those offenses are rising above the trend-line having a much higher YAC based on what we would expect from their ADoT.
3rd and 9, Vikings are dropping out of double mug to trips bunch. Lions take the space given pic.twitter.com/Q3lT86PAMC
— Shawn (@SyedSchemes) December 12, 2022
As an offense, if you were told you would get 10 yards per completion with Option #1 being a completion that goes 9 air yards and 1 yard after the catch and Option #2 being a completion that goes 4 air yards and 6 yards after the catch, Option #2 should be the preferred choice. This is because, all other things equal, a 4 air yard pass has a 71% chance of being completed compared to a 9 air yard pass having a 60% chance and yards after the catch are dependent on that the pass being completed. Football is not always this black-and-white, but through the quarterback, play caller, and offensive weapons, we can see why these teams are generating yards after the catch and the efficient offense it’s leading to.
The next question is how these teams are generating YAC. Kyle Shanahan is regarded as one of the best play callers in the league, so we can see a path for the 49ers to lead the league in YAC per dropback despite starting three different quarterbacks. Patrick Mahomes is regarded as the undisputed #1 quarterback in the league, so there is a path for the Chiefs to show up so highly as well. For other teams, however, it can get more complicated.
To break down the YAC contributions of each group for each team, we can build a mixed effects model that takes into account air yards, pass location, and a team’s rushing ability as the fixed effects, and a team’s play caller, quarterback, and receiver as a team’s random effects (similar to Ron Yurko’s research on a similar topic). This can divide up the responsibilities for who is generating YAC for their team throughout the season. Instead of just the receiver getting full credit for the 3 EPA generated from a long after catch run, 1.5 can go to the quarterback, 0.5 can go to the play caller, and 1 can go to the receiver himself. Each of those splits can be bucketed as a participant’s “YAC EPA Added”.
While the model can help us divide the contribution for YAC EPA amongst members of the team, there are still some areas for improvement. While air yards is a factor of the model, it is possible it is not being accounted for enough as YAC EPA Added and ADoT will still be correlated. Additionally, while the mean YAC EPA Added can be an actionable takeaway, most play caller’s and quarterbacks’ effects have a confidence interval that crosses 0 meaning that they can’t be deemed as statistically significant.
When looking at each groupings distribution for mean YAC EPA Added, we can see that quarterback has the highest average, followed by play caller, and then receiver has the widest tails. This makes sense as there are 32 play callers, 40 quarterbacks, and 65 receivers included in this model. For receivers, we have our Deebo Samuels on the far right that are adding the most YAC EPA in the league and our D.K. Metcalf’s on the far left that seldom add after the catch.
Looking specifically at play caller, the model supports the position that Kyle Shanahan, Andy Reid, and Ben Johnson are three of the top offensive minds in the league. These three coaches added between 0.02 and 0.04 YAC EPA on average. Doug Pederson, in his first year with the Jaguars, also did a great job calling plays that could maximize the YAC opportunities that were to be had by his team. Ken Dorsey ranking highly is also interesting to note given the Bills’ offense was predicated on a high ADoT and low YAC. When looking into their YAC contributions specifically, Dorsey did a good job setting up avenues for the passing game to advance the ball once passes were completed.
From a quarterback perspective, Patrick Mahomes takes his rightful spot on top of the graph contributing to the most YAC EPA added amongst passers. 49ers quarterback Jimmy Garoppolo and Lions quarterback Jared Goff appear behind Mahomes even with their play callers receiving so much praise. This could mean the quarterbacks had a bigger impact than given credit for despite being in offenses that are seen as quarterback friendly.
As mentioned above, there still is correlation with air yards and YAC EPA Added and that is notable when examining the picture from a receiver standpoint. Deebo Samuel added 0.15 YAC EPA on average to the 49ers making that the highest mark in the league but did so partly due to a low average depth of target (ADoT). What is just as impressive is receivers like Jerry Jeudy, Terry McLaurin, Jaylen Waddle, and Davante Adams producing similarly high YAC EPA on a higher ADoT. On the other end, we see receivers like Drake London and D.K. Metcalf add their value bringing in deeper passes instead of tacking on big YAC numbers.
The 49ers offense received much praise in the 2022 season as they were able to maintain high efficiency – thanks in part to a YAC-based offense – with multiple starting quarterbacks. When broken down, it can be seen that offensive architect Kyle Shanahan had the most to do with this as he added 12.3 total YAC EPA to his team. The difference between Jimmy Garoppolo and Brock Purdy is shown as Garoppolo added 0.06 YAC EPA on average while Purdy averaged 0.02.
Down safety buys the run and Deebo Samuel gets rolling. Love Brandon Aiyuk's blocking down the field pic.twitter.com/7BA0DuE3V0
— Shawn (@SyedSchemes) March 7, 2023
While yards after the catch is just one aspect of what can help offenses score points, it has shown an increase in importance over the past couple seasons as defenses devote resources to stopping deeper passes and accept shorter passes that make up longer drives down the field. Who the YAC is being driven by is also as significant as why it’s happening so the components that make up the full picture can be broken down. As mentioned earlier, it should be noted that YAC EPA Added is still influenced by a variety of factors outside of the model’s control as well as features in the model that might not be taken into account enough (e.g. air yards). However, it can still give us a direction in the building blocks that make up good YAC-based offenses and give teams a guide of what to look for if they choose to go down that road.
