Predictable Offense & Prevent Defense: No Team Blows Second Half Leads Like the Jaguars

By Warren Sharp

It’s hard to win games in the NFL.  It’s harder for a team like the Jaguars, led by Blake Bortles, to win games.  If fortunate enough to play the game successfully so as to hold a lead in the second half, a team like the Jaguars must play intelligently with a lead.  That’s because it’s easier for a team like the Jaguars, led by Blake Bortles, to lose that lead.  A team won’t win many games when playing with a second half lead and using highly predictable offense and extremely soft prevent pass defense.  Especially a team like the Jaguars.  But that is exactly what the Jaguars did last year.  And what happened?  The Jaguars blew second half leads in a NFL-high 5 games from week 7 onward last year.

The Jaguars started out 2-3 last season, with narrow losses to the Packers (4 points) and the Ravens (2 points).   They actually held a second half lead over the Ravens before blowing that lead.  Once Nathaniel Hackett took over as the offensive coordinator, the Jaguars vaulted into the most pass heavy team in the NFL when leading in the second half.  They passed the ball on nearly 60% of their offensive plays.  It was a ridiculous rate considering two things.  First, the NFL average was only 45% pass.  Second, they were terrible when they did pass.  Bortles was successful on only 37% of his passes and delivered a terrible 58.6 rating when passing in the second half.  When leading in the second half, the NFL average was a 45% success rate and a 92.3 rating on passes.

Typically a team with the lead in the second half may face more third downs than normal, because they are less focused on early down efficiency and more focused on running out the clock.  This puts extra emphasis on 3rd down efficiency.  It makes it more vital to be productive on 3rd down, because if you fail, you’re likely kicking the ball back to your opponent.

So how did Jacksonville fare on 3rd down in the second half with a lead?   They were music to a defensive coordinator’s ear.  Examine these numbers:

  • With 3+ WRs on the field, they passed the ball 100% of the time and were always lined up in 11 personnel.
    • The predictable nature made it easy on defenses, and Bortles delivered just a 20% success rate and a 42 passer rating on 3rd down in the second half with a lead.
  • With less than 3 WRs on the field, they ran the ball 100% of the time, from either 12 or 13 personnel.
    • The predictable nature was so terrible that they recorded a 0% success rate  on these runs.

Jaguars 3rd down play calls

Defenses knew exactly what Jacksonville was going to do on 3rd down when leading in the second half simply based on who they sent onto the field.  And as a result, they were the least successful offense in the NFL in these situations.  They were successful on just 20% of their third downs when leading in the second half.  This is why their opponents always had opportunities to come back in games.  This is why the Jaguars lost a NFL-high 5 games from week 7 (when Hackett took over) to the end of the season despite leading at in the second half in those games.

Jags losses once leading in 2H

The maddening thing about their third down performance was that on early downs, they were actually better than average.  On early downs with a lead in the second half, the Jaguars were successful on 48% of their offensive plays, which was better than the 47% league average.  They faced third and an average of 6.5 yards to go, which was actually ranked 6th best in the league.  The average was 7.2 yards to go on third down in the second half when leading.  While that may sound bad (6.5 yards to go seems unmanageable), only the Steelers, Redskins, Cowboys, Cardinals, Chargers and Broncos were better offensively last year.  Jacksonville averaged 4.8 yards per play on these early downs, which was above league average (4.69).  They were not very explosive on these early down calls, thus they did not bypass 3rd down as often (recorded new first downs on 17% of their early down calls, 8th worst in the NFL and below the 21% average).

Jacksonville must be more creative when leading in the second half.  In 2016, Hackett tipped his hand like you wouldn’t think is humanly possible.  On any down, if they had fewer than 3 WRs on the field, they ran the ball 100% of the time, gained just 1.7 yards per carry and they posted a success rate of just 29%.  In large part due to the Jaguars predictability, the defense knew what was coming, and it directly cost them.  The NFL average with less than 3 WRs was 74% run, with a run success rate of 40% and 3.2 yards per carry.

The problem for the Jaguars wasn’t entirely related to Hackett and the third down play calling, although inevitably there is a spillover effect from poor offense into a defense.  From week 7 onward in the first half, take a look at the Jaguars defense in the following situations, first visually, then in tabular format:

Jags DEF in 2H - 2

Jags DEF in 2H

The Jaguars defense was the #1 defense in the first half and they were #1 in the second half when the Jaguars were trailing.  However, when they were leading, while their run defense was close to their first half rate of success, their pass defense dropped all the way to the NFL’s worst, allowing a whopping 57% success rate to opponents.  If there ever has been a way to visualize what a prevent defense looks like, it is on the graphic above.  Examine how truly mind blowing it is for the 3rd ranked first half pass defense and the second ranked second half pass defense (when trailing) to drop to the league’s worst when leading.  This again relates to play calling, but in this case it is the fault of play calling from defensive coordinator Todd Walsh.

On both sides of the ball, play calling was a prime element which cost the Jaguars 5 games from week 7 onward that they lost despite leading at some point in the second half.  In these games, there is no doubt that play calling was a major reason the Jaguars gave up these second half leads.  But it is not entirely the fault of the coaches.  The players made mistakes and need to clean up elements of their execution when playing with a lead late in the game.

But as Jacksonville’s training camp opened the focus has been on quotables from the team about winning or making a trip to the Super Bowl.  Calais Campbell said “all the pieces are in place” to win the Super Bowl, and Malik Jackson took it a step further and said he expects the Jaguars will win the Super Bowl this year.  But to win the Super Bowl, the Jaguars must win games.  And to win games, they have to figure out how to play with a lead in the second half.

This week, Head Coach Doug Marrone gave the following response when asked how many times a game quarterback Blake Bortles will attempt passes:

 “Zero.” [He said he wasn’t joking about this, and he wasn’t smiling when he said it.] “For me, I’d like to run the ball every play.  I want to go back to the old way. I want to change the game.”

This is concerning from the very fact that the Jaguars, as described above, appear to have extremely little creativity on play calling, and are very single-minded based on personnel groupings in the game.  There is no doubt minimizing the influence of a poor quarterback like Bortles would be in the Jaguars best interest.  But that means all the more reason for the Jaguars to involve disguise and deception with their play calling.  They clearly did anything but that in 2016.

In 2016, the Jaguars recipe when playing with a lead in the second half was highly predictable formations and play calls offensively, and utterly painful prevent pass defense.  That combination caused them to lose 5 games that they led in the second half.  And that cannot repeat itself in 2017 if the Jaguars want any hope for a winning record, let alone winning a Division Championship, Conference Championship and the Lombardi.

About the author:  Warren Sharp ( @SharpFootball ) owns and operates SharpFootballAnalysis.com and SharpFootballStats.com

The NFL’s Best and Worst Shotgun Runners

By Anthony Staggs

In evaluating situational rushing performance from various formations, it is apparent that not all NFL backs are as good running from spread and shotgun formations. The most popular formation is “11 personnel” which is used by NFL teams on 60% of offensive snaps.  No other formation is even used 20% of the time.  [11 personnel is 1 running back, 1 tight end and 3 wide receivers.]  Of the over 20,000 offensive plays from 11 personnel, teams lined up in shotgun on 81% of these snaps.  When they were in shotgun from 11, they passed the ball 76% of the time.  When they were under center from 11, they ran the ball 65% of the time.

The shotgun formation is becoming ever more popular in the NFL with the league wide average usage at 63.5% last season. We have heard the narrative of guys like DeMarco Murray struggling when asked to run out of spread shotgun formations, and there are pure football reasons for this beyond just the statistics. Running from the shotgun takes a slightly different skill set than the I-formation sets many runners are accustomed to employing.

Patience is a key trait to see the hole in an outside or inside zone and make their move. Running backs also need to have that acceleration to burst through the hole once it is seen. While these may seem like key traits to any running back, they are even more important to spread rushers as they typically try to attack the edges of defenses and get their backs out in space.

AS02-1-1

Last year, the average NFL team ran 187 plays to what is deemed the outside (tackles and out) and 214 plays to what is deemed the inside (guards and in). On inside runs, the league-wide average was 4.01 yards per carry. Outside runs were much more effective averaging 4.74 yards per carry. With ends and outside linebackers getting lighter and lighter by the season, this is something offensive coordinators can exploit with spread runners. Chip Kelly, for all the flack he got at the NFL level was one of the best at attacking the edges in the run game and letting his runners barrel through smaller defenders or take it to the outside and use their speed. Last season the top five rushing attacks belonged to the Buffalo Bills, Dallas Cowboys, Tennessee Titans, San Francisco 49ers, and Atlanta Falcons. Here is how those teams ranked in yards per attempt to the outside: Buffalo (5th), Dallas (11th), Tennessee (4th), San Francisco (6th), and Atlanta (9th). Attacking the edge is an increasingly important factor in today’s NFL, and teams that do it best usually have successful run games.

Last season, NFL teams ranged from 50 (Cardinals) carries to 444 (49ers) carries out of the shotgun formation. From under center, San Francisco had just 15 carries while the Atlanta led the Falcons with 423 carries. The average NFL team last year attempted nearly 100 less rushes out of the shotgun formation despite the league being more effective as a whole out of the set. The average NFL team averaged 4.82 yards out of shotgun formation last season and a lowly 3.63 out of under center, that’s a difference of nearly 1.2 yards per attempt! One reason to believe teams are so much more effective running out of the shotgun is because of run pass splits. When we examine the shotgun split league-wide we see that NFL teams run out of the shotgun formation just 23.5% of the time and throw it 76.5% of the time when they are lined up deep. Conversely, in under center formations teams run the ball 68.3% of the time and throw it on just 31.7% of plays. NFL teams know this split, giving the benefit of the doubt to the runner and opening up wide lanes for them to crash through.

AS02-2

Lets also take a look at play calling splits from under center and in shotgun when in the opponent’s redzone. This is particularly valuable for end efficiency, which includes fantasy touchdowns, as most rushing TDs originate from red zone runs.

AS02-3

The shotgun is used slightly less frequently here at just 61% of the time. When in shotgun teams still throw the ball on 75% of plays. Under center, teams run the ball more often on 75% of plays. Using the shotgun run could be just as effective in the redzone as it is in the open field with defenses knowing these tendencies. Spreading the field and running from the shotgun set can be a key edge for any team, especially one with a heady quarterback who can run RPO’s (run pass options). Looking at the careers of the top-40 fantasy running backs last season there is a remains a stark contract between effectiveness between running out of the shotgun (4.68 yards per carry) and from under center (4.17 yards per carry). The top-40 running backs last season have season 28.6% of their career carries from shotgun formation. Now that we have set the stage league wide, lets examine some players who over the course of their careers (to increase sample size) have shined and struggled while running out of shotgun formations:

 

Effective Shotgun Runners

LeSean McCoy has the most attempts of any runner in my database out of shotgun sets, checking in at just fewer than 1,000 carries out of the set with 992. Even Jamaal Charles and Adrian Peterson have just 312 and 130 carries from the shotgun in their respective careers. In fact, he has twice as many carries as the next guy Matt Forte out of shotgun. In his career, 52.4% of McCoy’s carries have come from the shotgun. In shotgun, McCoy has averaged 4.9 yards per attempt, 0.3 over his average from under center. His 23 touchdowns are ten more than the second highest running back in Lamar Miller. While McCoy is effective running from under center, putting him back in the shotgun will give defenses nightmares.

Jordan Howard was dynamite when in the shotgun set last season, blowing away all other runners busting off runs at a 7.0-yard clip. While Jordan Howard’s sample size may look small, with 94 carries from shotgun last season, Mark Ingram has just 96 and he has been in the league since 2011. The entire top-40 from last year’s career average is just 185 carries, so while smaller than we would like it was still a big season. Howard gained 657 yards on his 94 carries but did not score a touchdown. Howard averaged a solid 4.2 yards per carry last season, but his 2.8 yards per carry difference between splits was the largest measured in this sample.

Carlos Hyde was run almost exclusively out of the shotgun last season under Chip Kelly, much to his delight. 213 of his 217 carries last season came from shotgun and he thrived to the tune of 4.6 yards per carry. He has been even better than that in his career averaging 4.8 yards per carry on 297 attempts. Ten of Hyde’s 13 rushing touchdowns have also come out of shotgun. Issue is, under center Carlos has struggled to the tune of just 3.1 yards per carry on 118 attempts. Projecting forward, this could be the scheme fit issue John Lynch and Kyle Shanahan have spoken about this offseason. As previously mentioned, Atlanta, Shanahan’s former squad, ranked first in under center rushes and the 4th lowest in shotgun rush attempts, maybe all that talk about a surprise cut isn’t too far out of whack.

Ty Montgomery has the look of a prototypical spread runner, with requisite bulk and acceleration to attack the edges of a defense. Montgomery’s pass catching ability is already up there with the top running backs in the league, and working out like a running back will help him in his first full-season at the position. Montgomery was a monster on shotgun runs last season, rushing for 347 yards on 53 carries, that’s a 6.5 yard per rush clip and also scored all three of his rushing touchdowns. From under center, the former wide receivers attempts were limited gaining 124 yards on 24 carries. This doesn’t even mention how good Ty is as a receiver out of the backfield, where he is sure to be a big factor this season.

Bilal Powell has been hyped this offseason as maybe taking over more of the rushing role, and if they implement it effectively, it could pay dividends for a team expected to struggle almost across the board on offense. In his career, Bilal has rushed for 1,333 yards on 258 carries, a 5.2 YPC clip. When under center, Powell has rushed 275 times for 990 yards, just 3.6 yards per rush clip. Powell’s ability to run from the shotgun could be important if the Jets choose to run a spread formation with either Josh McCown or their young inexperienced quarterbacks in Bryce Petty or Christian Hackenberg.

Other notables: Duke Johnson, James White, Rashad Jennings, and Jeremy Hill.

 

Less Effective Shotgun Runners

Ezekiel Elliott was a revelation for the Cowboys last season, but he wasn’t quite the runner out of shotgun as he was from under center. On 53 attempts out of shotgun, Zeke compiled 234 rushing yards and just two touchdowns. From under center, Zeke buttered the bread, racking up 1,396 yards on 269 carries and 13 touchdowns. This came as a surprise as Zeke came from a spread offense in college and had Dak Prescott’s rushing threat to keep defenses honest on the backside. In Dak’s second season in the league, Zeke could be lined up in shotgun more than you may expect, so he needs to improve in this area. In total, The Cowboys ran 541 plays out of the shotgun and 533 from under center. When under center the team had the highest run percentage in the league at 77.9%. In shotgun, they had the sixth lowest run rate at 20.7%. Some more variation in formations with an athletic quarterback at his side could make Zeke even more dangerous, but if not power football worked like a charm last season.

DeMarco Murray is always the name brought up when you talk about running from under center and out of shotgun, and while it might be slightly overblown, it has some merit. Under center, Murray has been excellent, with 1,081 career rushes for 5,037 yards and 32 touchdowns. Out of shotgun, Murray has just 339 career carries for 1,478 yards and 11 rushing scores. Half of Murray’s rushing attempts from shotgun came in his 2015 season in Philly and he averaged just 3.6 yards per attempt on them that year. On the other 173 career carries from the shotgun, Murray has compiled 884 yards or 5.11 yards a rush. So while Murray’s numbers don’t look great because of 2015, that may not be indicative of his true talent as a spread runner. Last season Murray averaged 5.4 yards per carry from the shotgun and 4.2 from under center. Murray continues to be an underrated performer at the running back position.

LeGarrette Blount does one thing and one thing well, run from under center; he does not provide a dynamic backfield threat however. In his career, just 8% of Blount’s runs have come from shotgun sets. On 97 career carries, Blount has rushed for 374 yards, just 3.9 a rush and scored three times. Under center, Blount is a beast, on 1,070 career carries he has rushed for 4.4 yards an attempt and scored 37 times. Blount’s new team the Eagles were in the shotgun 13% more than the Patriots and had nearly double the amount of their runs come from the shotgun formation. Besides obvious touchdown regression, maybe there are more reasons to be concerned about LB this season.

Tevin Coleman is a surprise show on the list, you’d think a player with his speed and acceleration would be great out of shotgun sets, and maybe he is, but he was never really given a chance to show that off with just 34 attempts to date from shotgun. On those carries he has gained just 94 yards and scored two touchdowns. New offensive coordinator Steve Sarkisian can either buck his trend for us, or prove that the team doesn’t see him in that light. In his NFL career, Tevin has 816 yards on 171 attempts from under center with seven rushing touchdowns. Tevin has major upside to his game, but will need to work on the nuances to bring it all together, and showcasing his skills from shotgun will be an even bigger part of his game.

Other notables: Adrian Peterson, Mike Gillislee, Mark Ingram, Theo Riddick, Jerick McKinnon, and Latavius Murray.

About the author:  Anthony Staggs ( @PyroStag ) was crowned co-winner of the Sharp Football Stats 2017 Writing Contest.  He will share articles featuring his analysis throughout the 2017 NFL 

Quickly Improve Red Zone Efficiency through Personnel Groupings and Play Calling

By Warren Sharp

The NFL is a passing league. So, it may sound counterintuitive, but I’ve mentioned for several years now that running the ball in the red zone is a way to steal efficiency. That is because most defenses fear the pass in that area of the field. And as Sun Tzu preached in The Art of War:

“Appear at points which the enemy must hasten to defend; march swiftly to places where you are not expected… You can be sure of succeeding in your attacks if you only attack places which are undefended.”

It’s for good reason they fear the pass, because teams are now 55% pass inside the red zone. Over 56% of the time, teams line up in traditional 11 personnel (1 RB, 1 TE, 3 WRs) inside the red zone, and they pass 65% of the time in these situations. However, even in 11 personnel, as we will discuss, running the football results in a double digit increase to success rate.

Running the ball is more efficient in the red zone, whether in short yardage or not. As I wrote in my 2017 Football Preview. In the red zone on 3rd or 4th and short, teams call pass just over 55% of the time, but a run in these situations produces a 12% success rate gain. Additionally, while over half of the pass attempts were unsuccessful, 8% resulted in turnovers or sacks, whereas there were zero turnovers on the run plays.

I’ll present the data on production from various personnel packages to see where the edges lie. Then we’ll dive into a grouping that is primarily used at the goal line, and discuss optimal play calling vs what teams typically do instead.

RED ZONE SUCCESS BY PLAY TYPE AND FORMATION

Teams pass the ball far too often in the red zone. As mentioned earlier, teams pass the ball 55% of the time overall. Inside the 10, it becomes a 50/50 split. And inside the 5, teams run the ball 55% of the time.

Yet regardless of the distance, RB runs in the red zone are successful far more often than any passes (to all positions).

WS25Jul17-01-1

[For the purposes of this analysis, I am removing QB runs, as QBs have an inherent edge to be successful in extreme short-yardage rushing due to their position on the field, as well as their ability to scramble on broken plays.]

This rushing efficiency translates not just to certain formations. For example, don’t think that rushing is more effective because in the red zone and as team’s approach the end zone, they suddenly shift into more compact formations which are far more effective:

The NFL average is 60% 11 personnel over the entire field. On pass plays, that jumps to 69%. And in the red zone, from the 3 yard line out to the 20, teams still use 11 personnel on pass plays 68% of the time, nearly identical to its usage over the field as a whole. The only sudden change is from 1-2 yards out from the end zone, when teams use 11 personnel on just 46% of their attempts (teams pass from a lot more 13 and 22 than over the rest of the field).

And even from these 11 personnel groupings, simply running the football instead of passing it would lead to much more success. Particularly when edging closer to the end zone. It’s not a slight improvement, it’s nearly 20% more successful to run the ball than to pass the ball from 3+ WR formations inside the 5 yard line.

WS25Jul17-02

Now allow us to compare rushing vs receiving from a variety of personnel groupings inside the red zone. As we can see, the efficiency edge for rushing the ball in this area of the field is not solely limited to 11 personnel. Play success in general increases as the ball gets closer to the end zone, because fewer yards are needed to qualify for a play to be successful. For instance: On the 20, on first and 10 the play needs to generate at least 4 yards to qualify as successful. On first and goal from the 5, that play needs to generate only 2 yards to qualify as successful.

As the next graphic shows, indeed rushing is more efficient regardless of the formation. A key takeaway from this graphic is the counterintuitive nature of wide receivers for passing in this area of the field. All passes are more successful from the red zone with fewer wide receivers on the field.

As I just mentioned, teams tend to stay in 11 personnel (3+ WRs) from the 20 up to the 3 yard line just as frequently as they do over the rest of the field. But the reality is that such formations tend to be mistakes. Passing out of 1 WR formations typically means 22 or 13 personnel. Inside the 10, for example, these passes have a 49% success rate. But targets to the TE from these formations have a massive 60% success rate, which is even stronger than rushing the ball in this scenario. [Perhaps in a future article I’ll discuss optimal positional targets in the red zone, based on personnel groupings.]

WS25Jul17-03

OVERALL TAKEAWAYS

The NFL is all about deception.  It’s about creating edges and attacking weak points.  Just because passing is more efficient than rushing over most of the field, that does not mean it continues to be so inside the red zone.  Rushing out of any formation is more efficient than passing, but balance is key.  Aligning in a formation which allows a team to pass or run, and will enable them to stay on schedule is a big edge.  That way, a team won’t find itself in 3rd and long or medium inside the red zone, leaving their only play call a pass to record a first down and avoid a costly field goal attempt.

If deciding to pass, particularly near the end zone, personnel groupings with fewer WRs and more TEs deliver substantially more efficient results.  And this makes perfect sense.  A team’s #3 DB against a #3 WR has less ground to cover and the #3 WR is less likely to get open quickly, which is how most passes in confined spaces need to be delivered.  However, a team’s #2 TE working his body positioning against a LB who doesn’t typically cover TEs often, let alone in confined spaces, is a massive edge.
JUMBO PERSONNEL

Some teams will load up in “jumbo” personnel, featuring no WRs, and use 3 TEs, a FB and a RB to try to gain short yardage near the goal line. How successful is this?

First let’s back up and discuss full-field jumbo packages. RB runs out of jumbo on 3rd or 4th down and 1 are successful 43% of the time. Compare that to RB runs with at least one wide receiver on the field in those same situations, which are successful 69% of the time, and you can see why it’s not particularly smart to trot on your jumbo personnel. Particularly because it gives zero threat to pass on 3rd or 4th down. [removing goal line plunge jumbo entirely, the RB success rates are 14% better in non-jumbo as opposed to 26% better from the 1 yard line, but this could be due to lower sample size producing higher variance.] Even the Patriots, who are one of the most efficient teams in the NFL, were 30% more successful (80%) when running their RB with at least one WR on the field on 3rd/4th down and 1 than they were when running their BR from jumbo formation.

So we’ve established that RB-runs from jumbo formations were successful just 43% of the time on 3rd or 4th down and short. On the 1 yard line, RB-runs from jumbo formations were successful just 38% of the time. And when in jumbo formation on the 1 yard line on 3rd/4th down, teams would run the ball 86% of the time. The few passes they attempted were never successful (0% success rate).

Add 1 wide receiver to give more of a threat to pass, and RB runs posted a 45% success rate (up from 38% without a WR). While teams primarily still ran the ball (79% run), passes became much more successful at 57% (up from 0% without a WR).

Add another wide receiver to place two on the field, and suddenly we’re at peak efficiency. RB runs posted a 73% success rate. And while teams were still run heavy (71% run), passes were 100% successful, with each pass recording a TD.

When teams added at least one more wide receiver, to play 3+ WRs on the 1 on 3rd or 4th down, the strategy switched to passing the ball, but that was a mistake. Teams ran on only 40% of attempts, much less than the 71% with 2 WRs or 79% with 1 WR. But these runs were much more successful than the passes: 60% of the RB runs were successful, while only 40% of all passes were successful.

WS25Jul17-04

What has this shown us regarding jumbo formations? A few key takeaways:

1) Jumbo formations are terrible in these situations. Stop coaches from using them, and feel free to shudder any time you see one trot onto the field on 3rd/4th down from the 1 yard line.
2) The optimal play from the one yard-line on 3rd or 4th down is a run play, but the formation is more key than the actual play call.
3) The optimal formation is 2 wide receivers, using either 12 or 21 personnel. RB-runs out of these formations are the best of any grouping, delivering a 73% success rate. In part this formation is successful because it delivers enough illusion for a pass. However, passes are also extremely successful from 12 or 21 personnel.
4) If a team wants to pass the ball, they should not add more wide receivers in hopes of playing “find the open man” thanks to more options. Passing out of 3+ WRs is a terrible strategy in this situation. However, rushing with a RB is successful 60% of the time, an extremely palatable rate.