261 rushes, 1,307 yards, 5.0 yards per attempt, 11 touchdowns.
121 targets, 91 receptions, 721 yards, 7.9 yards per catch, 4 touchdowns.
Those numbers put New York Giants star running back Saquon Barkley in the top five of just about every common rushing and receiving stat for NFL running backs — and he is in the top three more often than not.
So taken at face value, it would seem that whatever the Giants did to help Barkley become the reigning Offensive Rookie Of The Year, they should keep on doing it.
Right?
The folks who dive deep into the data beyond the box score, look at the context formed by thousands of similar plays, suggest that the Giants should rethink how they use Barkley if they want to run a consistent offense and get the most out of him.
To fans and NFL personnel of a certain point of view, the idea that nerds looking at a spread sheet can know more about how to use a generational running back talent than ‘Football Guys’ is nothing less than rank pencil-necked geekery, and should be dismissed as such.
But before dismissing the idea that winning football is about resource management and maximizing efficiency, I’d like to share a name with you: Ernie Adams.
Most football fans don’t know who Adams is, but he might be the most important man in NFL history. Three times he has been a trader on Wall Street, but he is also a close friend of Bill Belichick going back to grade school.
In 1979 Ray Perkins hired Adams to the Giants when he became the Giants’ head coach, and Adams quickly convinced the Giants to hire Belichick. Eventually, Adams followed Belichick to the Cleveland Browns in 1991 and then to the New England Patriots in 2000.
With the Patriots he holds the title “Football Research Director,” and he is intimately involved in almost all areas of the Patriots’ machine. On Sundays he sits in the press box and advises Belichick on tendencies of opposing teams and in the offseason he assists the scouting department in building its value chart. But mostly his job entails studying the game from an analytic point of view. For instance, Adams was one of the first to study fourth down tendencies, coming to the conclusion that teams punted far too often — and in 2018 the Patriots were one of the most aggressive teams in trying to convert fourth downs. Adams keeps a low profile and nobody (save Belichick and the Patriots’ brass) truly know all of what Adams does. But given the utter dominance of the Patriots over the last two decades, trying to see the game in the same way — maximizing every advantage and avoid inefficiencies wherever possible — seems to be prudent.
So, what does all this have to do with Saquon Barkley?
In the chapter dedicated to the Giants of his 2019 Preview e-book, Warren Sharp says, “It’s been proven time and time again that running is devalued. I’ve discussed it ad nauseum and won’t put you through that again. But it’s a fact. And understanding the devaluation of the running game is the measure of a quality pro-football analyst. Running is favorable in short-yardage situations and some areas of the red zone, but it is -EV virtually everywhere else.”
Sharp is blunt (no pun intended) in his statement, but the data backs him up: The teams that are better at throwing the ball are the ones that win more consistently.
In these two charts from ThePowerRank.com, we can easily see the relationship between passing (and defending the pass) and winning.
And if you think about why it should be that way, it makes sense. It all comes down to basic math: You need a great running back (or a great running game) to average 5.0 yards per carry, but the league average yards per pass attempt in 2018 was 7.5, and the average completion was 11.3 yards. If you only have so many offensive plays in a game (the NFL average is about 63 per regulation length game), you are going to want to make the most of those opportunities.
This graph from Chase Stuart, writing for FootballPerspective, shows just how much more efficient throwing the ball is than running it.
The last time it was a better idea to run the ball than throw the ball on a given play was 1977. By 2018 there is a stark divide between teams that have “throw first” philosophies and those with “run first” philosophies.
But per SharpFootballStats, the Giants ran the ball on 52 percent of their first downs in 2018. Basically, when the Giants had their druthers, before the flow of the game could influence their play calling, they wanted to run the ball. The problem was that they weren’t terribly good at it. As great as Barkley is, the Giants were only successful on 44 percent of those plays, good for 27th in the NFL. For reference, the Giants were eighth in success rate in those same circumstances with Orleans Darkwa and the pre-rebuild line of 2017.
Getting the most out of Barkley on the ground
If the Giants are bound and determined to lean on the run in 2019, they should probably consider ways to get better results.
The most obvious way to improve the running game is something I’ve touched on before: Force the defense to not devote defenders to stopping it.
The Los Angeles Rams fielded one of the greatest rushing attacks in the history of the NFL through the first half of the 2018 season. To the point where Barkley himself said that Todd Gurley should be the MVP for the season and NBCSports stated that the Rams might have had the best run blocking offensive line of all time.
But while Gurley certainly is a great running back, and the Rams have a good offensive line. But the other aspect which doesn’t get talked about as much is how they scheme light boxes for their running backs.
How good were the Rams at giving their runners the advantage? Per NextGenStats, Gurley and C.J. Anderson made up 50 percent of the list of running backs who had more blockers than defenders on the majority their carries.
Only four running backs face more blockers than defenders in the box on average (with at least 65 carries). Three of those backs will be featured on Conference Championship Sunday:
— Next Gen Stats (@NextGenStats) January 16, 2019
C.J. Anderson
Damien Williams
Todd Gurley#RamsHouse | #ChiefsKingdom pic.twitter.com/vx3HrFBr4x
And it wasn’t just devoting blockers to the running game either. The Rams were able to force defenses to play light boxes against Gurley, who faced six or fewer defenders on 44.9 percent of his carries per PlayerProfiler.com. That a running back of Gurley’s caliber saw a neutral (seven man) or light box on 85 percent of his carries simply defies conventional wisdom. After all, defenses are going to stack the box to make sure a great back can’t beat them, right?
As it turned out, Sean McVay had a life hack for NFL defenses: the 11 personnel package (1 RB, 1 TE, 3 WR). The Rams ran the 11-personnel package on an incredible 94 percent of their running plays (94 percent of all plays) before WR Cooper Kupp was lost for the year in Week 10.
11-personnel became something of a dirty word around Giants-land after the Ben McAdoo era, in which the Giants ran the package on 92 percent of their plays in 2016, and remained well above league-average before McAdoo was dismissed in 2017. But as it turns out, that was one of the truly insightful aspects of McAdoo’s offense — a fact that I didn’t grasp until digging into how, and why, the Rams were so good at running the ball in 2018.
For reasons we won’t get into, the Giants just won’t be able to field as scary an 11-personnel package in 2019 as they could in previous years. But as Josh Hermsmeyer suggests, defenses tend to react to personnel, not players. So the Giants will still be able to influence defensive personnel and alignment, even if they can’t take advantage of it as well as they could in the past.
The data suggests that long as the offensive line and running back execute, most of a run play’s yardage can be anticipated based on defensive personnel and alignment — as well as field position.
It’s a trend that shows up on tape as well. In studying film of new Giants’ guard Kevin Zeitler, it was pretty easy to anticipate the outcome of a given running play before the snap based on just the box count — if the defense put eight defenders in the box, the play wasn’t going anywhere.
The idea of taking advantage of a box count to make it easier to run the ball isn’t a new one. In fact, it dates back to the advent of the Run and Shoot offense and Kevin Gilbride in the 1980s. In it’s original form, the Run and Shoot would only run the ball against six (or fewer) man boxes, counting on their personnel packages, alignments, and explosive passing game to force defenses into nickel and dime packages.
In 2018 the Giants didn’t seem to care much about the box count, running Barkley against neutral or heavy boxes 74.4 percent of the time. But while Barkley was still productive and able to break off a long run at any time, wouldn’t it behoove the Giants to maximize his number of carries against light boxes? After all, on the 25.6 percent of his runs in which he saw six or fewer defenders, Barkley averaged 6.7 yards per carry — nearly as efficient as throwing the ball.
And if the Giants want to get the most out of Barkley on the ground, they will need him to be as consistently productive as possible. Over the last four weeks of the 2018 season the Giants relied on Barkley to be the engine which drove their offense, but struggled to get consistent production from him. He certainly had explosive, highlight reel plays, but on a down-to-down basis, he was inconsistent.
He finished strong against a Dallas defense looking forward to the playoffs, and in weeks 14 through 16, he had 244 yards on 49 carries (4.9 yards per carry). But taking a closer look, 130 of those yards came on two plays in week 14. So on his other 47 carries in that three week span, Barkley had 114 yards — just 2.46 yards per carry.
The other way to get the most out of Barkley on the ground is to give him the ball close to the end zone.
Turning back to the Patriots, they were one of the most run-heavy offenses in the NFL. In his preview of the Patriots, Sharp notes, “I urged the Patriots to shift to a more run-heavy approach in 2018. They listened. They replicated their 2016 run rate of 67 percent. And runs were tremendously productive. Run plays averaged a 65 percent success rate compared to passes which produced a mere 30 percent success rate. It was wise for the Patriots to go more run-heavy. Sony Michel’s 67 percent success rate and seven rushing touchdowns led the way.”
Earlier in the preview, Sharp notes that the Patriots turned to the run on 73 percent of “second and short” plays, and when they turned to Michel on short yardage on first or second down, they were successful at least 60 percent of the time.
According to Josh Hermsmeyer, this tracks with what the data says league-wide.
Hermseyer wrote, “Effective running in the red zone, and especially at the goal line, is particularly valuable because this is the part of the field where passing is most difficult. As teams move downfield and get closer to the end zone, the field compresses and completion percentage drops. While the effect begins a little before the 30-yard line, league wide completion percentage drops from 57 percent to 48 percent [based on data from 2009 to 2016] as teams move from their opponent’s 20 to the 3-yard line. This decrease in passing effectiveness puts a premium on being able to run successfully. Teams that can consistently move the ball on short-yardage runs in the red zone — or runs on which a first down or touchdown is no more than 3 yards away [excluding kneel-downs] — give themselves the opportunity to score touchdowns more often, and they tend to win more games.
If the Giants are going to run their offense through Barkley, and at it certainly seems like that is the plan, the need him to be consistent on down-to-down basis. But picking their spots to run the ball doesn’t necessarily mean taking touches away from Barkley.
Barkley the receiver
One of the big arguments for why it was okay to select Barkley with the second overall pick wasn’t just his dynamic ability as a running back, but also his ability to be a dynamic receiver as well. And, once again, Barkley was a great receiving back last year, with the second most receptions behind Christian McCaffrey, fourth in receiving yardage, and third in receiving touchdowns.
Those numbers are all impressive, but what even more impressive is how difficult a position the Giants put Barkley in as a pass catcher. While his 721 receiving yards were good for fourth most among all running backs, and easily second-most among backs with at least 200 carries, Barkley actually had 768 yards after the catch. With an average depth of target of -.7 yards, forcing him to run 47 yards (over the course of the season) before his receiving yardage began to count.
When players say that they’re going to give 110 percent, that’s probably not what they mean.
If the Giants want to make the most of Barkley’s explosive athleticism and ability to create magic with the ball in his hand, they will need to reach further down the field.
As we can see, passes at or behind the line of scrimmage are worthless at best, and the value peaks between 15 and 20 yards downfield. This doesn’t mean that all passes should be there, but the Giants should make an effort to look further down the field than they did in 2018.
There are concerns about pass protection holding up, but Barkley clocked a 10-yard split of 1.54 seconds and a 20-yard split of 2.57 seconds at the NFL combine and Evan Engram clocked a 1.56 and 2.57, respectively. If the Giants can’t protect a quarterback long enough to throw that pass with anticipation (ie, before Barkley or Engram make their break or reach the catch point), they have much much bigger problems than maximizing their passing EPA.
The Giants also need to get better at sustaining their drives, and while they ideally will want to pick up first downs on first or second down, they will also need to improve the rate which they convert third downs.
Sharp notes that part of the Giants’ problem in sustaining drives is relying on Barkley as a “Third Down Back” — or rather, relying on a third down back at all, writing:
“Last year’s Giants ranked 26th in third-down efficiency, largely because of their reliance on Barkley as a third-down receiver. Saquon drew 32 third-down targets – just three fewer than team leader Sterling Shepard – but converted first downs on just 25% of them, by far worst on the Giants. Shepard converted 49% of his third-down targets into chain-moving gains, and OBJ converted 48%. Barkley was barely half as likely to convert. But that’s not a knock on Barkley. NFL running backs have converted exactly 32% of third-down targets into first downs over the last three years. Wide receivers and tight ends are far likelier to convert.”
Sharp added, “Let’s remove all third downs with 11-plus yards to go in the interest of fairness. Barkley lead the Giants in targets (12) on such plays. All 12 of those targets occurred short of the sticks and averaged -2 air yards, forcing Barkley to gain 13-plus yards on for a first down. And he went 0-for-12, averaging 4.4 yards per target.
Defenses took the “dump off” away from the Giants, especially on third and long, but Eli Manning kept throwing it. He threw 35-of-41 passes on third-and-long short of the sticks, generating a 6 percent conversion rate. His conversion rate on passes thrown to wide receivers on third and long was 67 percent, including incompletions”
The forward to the 2019 preview, written by Dan Pizzuta (recently of Big Blue View) and Evan Silva holds one answer for making Barkley even more dangerous as a receiver: Throw to him on first down.
They wrote:
“In 2018, there were 6,248 running back rushing attempts on first down. They averaged 4.5 yards per carry, minus-0.01 Expected Points Added per attempt, and a positive play rate of 41.3%. When teams threw to running backs on first down, they averaged 6.02 yards per target, 7.8 yards per reception, 0.08 EPA per attempt — slightly more efficient than the average of all passes regardless of down at 0.05 EPA — and a positive play rate of 52.3%.”
Simply making Barkley a “first down back” and throwing to him on first downs would go a long way toward harnessing the potential of one of the Giants’ most dangerous offensive weapons, while also helping to sustain drives.
Putting it all together
So then, do the Giants need to re-evaluate how they use Saquon Barkley and the running back position?
Probably.
Yes, Barkley was the Offensive Rookie of The Year and finished the season as one of the most productive running backs in the year last year. But the Giants still finished 5-11, with four of those wins coming against backup quarterbacks, scoring 9 points more per game against backups than against starters. And in all five of their wins the Giants finished with a +2 (or better) turnover margin.
That isn’t to disparage Barkley or his contributions to the Giants’ offense, but we also cant discount the role that circumstance obviously played in their record.
As good as Barkley was last year, the Giants will need him to be even better in 2019.
So let’s pretend we are Bill Belichick for a moment: What suggestions would Ernie Adams offer ahead of 2019?
- Go light to run, heavy to throw
Last year when the Giants used 11-personnel they threw the ball, and when they used 12 or 21-personnel, they ran the ball. As we saw above, lighter offensive personnel packages force defenses into lighter personnel packages and drew defenders away from the ball. Flipping the tendencies also has the advantage of making for a more efficient passing game as well, as noted by Dan Pizzuta in the forward to the preview.
Dan writes:
“As passing has increased around the league, teams have embraced 11 personnel (three wide receivers, one running back, and one tight end). In 2018, teams lined up in 11 personnel 66% of the time. That shift has forced defenses to change to a nickel package (five defensive backs) as its primary formation. Throwing from 12 personnel (two receivers, one back, two tight ends) happens less frequently yet is more efficient. A heavier personnel package with more tight ends invites defenses’ base personnel with more linebackers and fewer defensive backs. 12 personnel also serves as pseudo play action because defenses are expecting the run from that package. Whereas teams threw the ball 66% of the time in 11 personnel in 2018, they passed just 48% of the time in 12. From 11, teams averaged 7.16 yards per attempt, 0.03 EPA per attempt, and a positive play rate of 49.5%. Efficiency increased to 8.08 yards per attempt and 0.15 EPA with a positive play rate of 54% of out 12. Now add a play fake and teams averaged 9.19 yards per attempt and 0.15 EPA. 14 teams averaged at least 9.0 yards per attempt off play action from 12 personnel in 2018. That was only true for eight teams out of 11 personnel.”
- Throw in the open field, run in the Red Zone
Again, as we saw above, the most efficient place to run the ball is close to the goal line and in short yardage situations. In those field positions, the defense has much less ground to cover, the density of coverage increases, the game speeds up, and passing windows shrink. Teams that can run the ball well in short yardage situations are more likely to score and more likely to win.
- Get Barkley down the field
By and large, throwing to running backs is more efficient than handing the ball to them. Perhaps not as efficient overall as throwing to wide receivers or even tight ends, but still more efficient than running the ball. However, Saquon Barkley is no ordinary running back and enjoys an athletic advantage on par with — or even greater than — most receivers. Whether it be by lining him up as a wide receiver, slot receiver, using angle routes, wheel routes, or throwing to him on play-action, the Giants need to get Barkley the ball further down the field.
- Throw to Barkley on first down
We already saw that running the ball on first and second down is almost a wasted down outside of short-yardage situations. But that doesn’t mean that we can’t get Barkley the ball. Throwing to running backs on first down maximizes their impact and makes the most of their targets. Just remember to NOT dump the ball off — we saw enough the “Superstar And A Prayer” under Ben McAdoo.
So then, stepping back into ourselves as Giants fans, how likely are we to see the Giants take the advice offered by analytics and flip their tendencies. Based on last year, not very. In 2018 the Giants had a strong tendency to retreat to the basics of their offense, run the ball regardless of box count and in the passing game they relied on short passes and yards after the catch.
However, we also haven’t seen Barkley at all in a game this year and have barely seen Evan Engram. Occam’s Razor would suggest that this is because the Giants know their season is sunk without Barkley and likely so without Engram. However, it’s also true that if the Giants were going to flip their offensive tendencies from what we saw in 2018, there is absolutely no risk of tipping their hand if those two players aren’t on the field.
We don’t have long to wait to find out: regular season football is just a week away.