Expected points added is an interesting is ephemeral concept, and one that has yet to be fully embraced by a sport in football that has been more averse to analytics than others, particularly baseball. Basically, the general premise of EPA is that it can determine the likelihood of a play positively or negative contributing to a team being able to score, and by how much.
Pro Football Focus recently published a list of every team in the NFL
, ranked by the EPA per pass play that they posted during the 2018 season. The Pittsburgh Steelers unsurprisingly finished in the top 10 in that list, placing ninth overall.
“No offense passed the ball more than the Steelers in 2018, as they dropped back to pass on 65.1% of their offensive snaps and Ben Roethlisberger led all quarterbacks with 675 pass attempts”, Ben Linsey wrote. “On a per-play basis, they managed to be effective despite a drop-off in play from Big Ben, who dropped from fourth in overall grade from 2013 to 2017 to 15th among 32 qualifying quarterbacks last season”.
“The Steelers’ playmakers picked up Roethlisberger in 2018”, however, he went on, “notching 6.1 yards after the catch per completion (fourth-highest rate in the NFL). The departure of Antonio Brown leaves a large hole to fill at the position, but JuJu Smith-Schuster appears ready to fill it. His 21 contested catches last season trailed only Deandre Hopkins”.
The eight teams who placed ahead of the Steelers in expected points added per passing play, from first and on, in order, are the Kansas City Chiefs, the New Orleans Saints, the Atlanta Falcons, the Los Angeles Chargers, the Seattle Seahawks, the Los Angeles Rams, the New England Patriots, and the Tampa Bay Buccaneers.
I don’t think there is much in the way of surprise in that list. And remember, it’s important to keep in mind that this is not about the overall passing game, but rather a performance on a per-play basis, so it’s qualitative, not quantitative. It’s also not necessarily predictive of success if a team has a bad defense, for example.
PFF doesn’t actually give the EPA numbers, so we don’t know where they quantified it, but I think we can say that the team’s usage of the short passing game as an extension of the running game may have somewhat artificially lowered their expected points added.