Luke Braun's Film Room: Towards A Theory of Drops (We Have A Really Bad Understanding of Drops)
Luke Braun's Latest Film Room gets philosophical and asks: what is a receiver drop anyway?
What is a dropped pass?
No, seriously, how do you define a dropped pass? If you picture a receiver drop, you probably picture something you’d see from Troy Williamson - a wide-open receiver gifted a perfect pass that clatters off that receiver’s hands. That’s a popular definition - if you can get your hands on it, you can catch it with those hands.
With that in mind, a challenge: How many times has T.J. Hockenson dropped a pass in the 2023 season? Without looking it up, hazard a guess before reading ahead.
According to Sports Info Solutions, Hockenson has dropped three passes in 2023. Stats.com doesn’t have Hockenson on their leaderboard, which only shows players with four or more drops. Pro Football Reference, which uses Sports Radar for their advanced statistics, has logged four drops for him. The harshest, Pro Football Focus, credits him with five.
Based on my extremely scientific poll, there’s a decent chance that you at least agree with PFF, and may have had an even higher number in mind. Why is that? What is so elusive about the idea of a drop? We can mostly agree that a stretching, reaching one-handed catch isn’t a drop if the receiver can’t bring it in. But where do we draw the line?
What about something like this? Josh Oliver cleanly gets both hands on this screen pass, but by no means is it an easy ball.
Opinions are split. Some say that if you get your hands on the ball, you should catch the ball. Some forgive it for being too hard a catch to make. One interesting idea from the comments comes from Twitter user (and Locked On Vikings bold predictor) Malcolm Biggles, among others.
The pass follows the dictionary definition of a “drop” since Oliver got his hands on it. But if the pass is too inaccurate, should it count? There is no consensus. PFF, for example, did not count this as a drop, nor did PFR. SIS disagreed. For a relatively clean rep, even the established authorities can’t agree. Another reply to the Josh Oliver “drop” approaches the issue from a different angle.
I think Caleb is driving at something important here. Count it or don’t, but this isn’t a play we expect Oliver to make. Mullens threw a bad pass, and Oliver couldn’t bail him out. That’s on Mullens, not Oliver. A drop wouldn’t reflect fairly on Oliver.
The stats websites seem to agree. PFR, for example, doesn’t even have a definition for “Drops” in their glossary. Neither does SIS. They do, however, track “catchability” and whether passes are “on-target”. Those can be used as statistical proxies for degree of difficulty.
PFF leans on its grading system as a panacea for these sorts of nuances. If a receiver has a low catching grade but not a lot of literal “drops”, they’re trying to tell you that they still think the receiver did a bad job, regardless of the rigid statistical definition.
Can Contested Targets Also Be Drops?
What about a reasonably accurate throw that is affected by a defender? In this play, the Raiders defender restricts T.J. Hockenson’s arm and Hockenson is unable to bring it in. It did hit him in the hands, but none of PFF, SIS or PFR logged this as a drop.
So if a pass is contested, is it automatically disqualified from a drop statistic? More directly, should it be? My instincts tend to steer me away from absolutes in football, but stats companies don’t get that luxury. They have to be rigid and consistent in order to be fair to everyone in the NFL, so they’ll have to come up with some line in the sand and ignore exceptions.
To solve this, these stats websites have a separate statistic for contested catches. By counting this play as a failed contested catch, they can label it more accurately and reserve their “drop” statistic for more egregious offenses. But, in this case, that runs counter to our intuition. That Hockenson play feels like a drop, doesn’t it?
Keep reading with a 7-day free trial
Subscribe to Wide Left to keep reading this post and get 7 days of free access to the full post archives.