“Daddy, where do fumbles come from?”
Startled, I nearly spit out my coffee. “Well … when a mommy and daddy fumble love each other very much …” I stammered.
He interrupted me, “Dad!”
I fixed a sober gaze on my son. “Do you really want to know?”
“Look at the chart.” I swiveled my laptop around to face my son and showed him this, aggregate fumble data pulled from all plays from 2009 to 2016:
Note: these categories are parsed from play-by-play data and may contain minor errors.Strip Sacks
Sacks can be game-changing plays in their own right, but the possibility of a strip sack adds another level to the impact pass rush can have. More than a third (35%) of forced fumbles occur on sacks, even though sacks represent only 3% of plays. A fumble occurs on 14% of sack plays, by far the most of any play category, and a little over half become turnovers.
Unsurprisingly, the list of forced fumble leaders since 2002 consists predominantly of edge rushers, with one notable exception: former Chicago Bears cornerback Charles Tillman (“Peanut” put in middle of name). Non-pass rushers will occasionally put up big fumble-forcing seasons, but they rarely do so consistently. Tillman possessed a rare ability to punch the ball free, forcing fumbles on nearly 5% of his tackles, more than double the league-wide average. Tillman, fellow defensive backs Charles Woodson and Brian Dawkins, and linebacker Derrick Johnson are the only non-edge players with 20 forced fumbles since 2002.
So while Atlanta Falcons safety Keanu Neal impressed with five forced fumbles in his rookie campaign, we shouldn’t expect a repeat performance in 2017. Miami Dolphins cornerback Byron Maxwell had four forced fumbles in 2016, but forced only five fumbles total in five prior seasons in the NFL. That’s not to say Neal and Maxwell were “lucky” in 2016 – undoubtedly they are trying to deliver big hits or rip the ball out – but that doesn’t mean they can do it again. History suggests they won’t.
If sacks are the leading cause of fumbles, we’d expect quarterbacks to be the leading fumblers – and they are. Of the top 50 fumblers since 2002, 46 are quarterbacks, with Frank Gore leading the non-QBs at 39th overall. Some of that is the longer careers quarterbacks have, but even looking at just one year, 2016, all 21 players with more than five fumbles were quarterbacks. QBs have other opportunities to fumble too, on snaps (about one in 500 snaps are fumbled) and if they scramble – certainly a factor with #4 Michael Vick.
Five-time MVP Peyton Manning excelled in a number of obvious ways, but his propensity to avoid sacks – and therefore fumbles – is a subtler way that he contributed to his teams. For his career, Manning took only 303 sacks in 9,683 dropbacks, and thus only coughed up the ball 77 times. Contrast with peers Tom Brady (417 sacks, 107 fumbles) and Drew Brees (358 sacks, 96 fumbles), who took more sacks and fumbled more in fewer drop backs. Brees and Brady are both above-average in avoiding sacks and fumbles, but Manning was other-worldly in this respect, averaging a miniscule 0.29 fumbles per game for his career.
No current quarterback can match Manning in this regard, but reigning MVP Matt Ryan averages fewer than one fumble per 100 drop backs and 0.35 fumbles per game, both outstanding rates. Andy Dalton of the Cincinnati Bengals, Matthew Stafford of the Detroit Lions, and Cam Newton of the Carolina Panthers also excel at avoiding fumbles. Newton’s track record is particularly noteworthy, since he also runs the ball often. In 2016, Newton took 36 sacks and had 90 rushing attempts, but only put the ball on the ground three times.
It is remarkable how consistent other plays are when it comes to fumbles. The fumble rates on catches, runs, and kickoff and punt returns all sit between one and two percent. We see higher rates on lateral plays – usually desperation circumstances at the end of game – and muffed punts, which are recovered by the receiving team at higher-than-normal rates. On the low end, kickoffs are rarely muffed, and fumbles on snap exchanges are uncommon. The consistency between types of plays has logic to it – fundamentally, all the plays where we have a ball carrier and a bunch of defenders looking to tackle him equalize around the same fumble rate.
This analysis can help us predict some regression, both positive and negative, for teams who showed a disconnect between their sacks and forced fumbles:
The obvious outlier here is the Oakland Raiders – they somehow led the league in forced fumbles while finishing dead last in sacks! That’s obviously unsustainable. On the other hand, teams like the Tennessee Titans, who only forced four fumbles despite a healthy 39 sacks, might poke a few more balls out this season.
The Luck of the Bounce
There are countless examples of fumbles changing the complexion of a football game, but they remain unpredictable events. This analysis makes clear that the best way for a team to up its fumble count is to sack the opposing quarterback more often. That might not be the kind of answer that would satisfy my five-year-old son, but it exposes hidden value that edge rushers can contribute to their teams.