NFL Offenses Turned to the Pass in 2019 and Turned the Corner

By Warren Sharp

We will look back on the 2018 NFL season as one that broke many offensive records.  There were a lot of them that can be discussed.

  • Most touchdowns (1,371)
  • Most passing touchdowns (847)
  • Highest passer rating (92.9)
  • Highest completion rate (64.9)

The 2018 season saw the second-most points in history scored as well.

But we didn’t just luck into this happening.

It happened because offenses got more aggressive than ever before and finally opened up their eyes to three of the most seductive words for any fan of offensive production:

First down passing

Last summer, after the Rams signed Todd Gurley to a $60 million extension with $45 million in guarantees, some thought this was a turning point.  After all, in 2017, teams ran the ball on first down in the first half of games 52.8% of the time (avg over the last 20 years was 51.8%).

I argued strongly against that logic.  I argued in favor of taking things to the extreme in favor in the opposite direction.  Last summer I wrote:

My contention is that teams are actually running the football too often.  And they haven’t learned their lesson over the last two decades. Teams are stuck in the past.  They feel it’s of-value to establish the run.  This is a -EV strategy.  Teams should not run on first down in the first half to “establish the run”.  We need to change the philosophy of “establish the run” to “establish the lead” and that comes by passing early and allowing teams to run late.

First-and-10 pass plays are more successful, they gain more yardage and will have a higher floor in 2018 than they did in years past… First-and-10 in the first half needs to become more of a passing down.

Despite a high-water mark in first-and-10 run rate in 2017 league-wide, and the bank being broken for the most expensive RB signing in history, thankfully, play callers around the NFL listened.

Historical Pass Rates on First-and-10

In 2018, the NFL saw a record 51.5% pass rate on first-and-10 in the first half.  These plays averaged 8.2 yards per attempt, another record.  And they averaged 67.6% completions, a third record.

Last year I created the below graphic that stepped from 1997-2017, studying play calls and production on first-and-10 play calls in the first half.  I’ve updated to insert the 2018 season for comparison:

first and 10 play calls

During the 2010 NFL season, the league modified penalties on passing plays to keep quarterbacks and receivers safer.  This allowed an easier pass to passing efficiency.  Looking at every year since the mid-season rules changes of 2010, it’s evident how the 2018 season stands out: 

first and 10 play calls4

Offenses became more aggressive with their first-and-10 play calls.  They intelligently opted to shift to the pass in a very strong manner.

Aggressive passing on first downs early in the game opened up the run game for many teams as well.  As a result, the run game was more productive.

Smarter Rushing Using Personnel Groupings

In 2018, 55% of plays on first downs in the 1st half were called from 11 personnel (1 RB, 1 TD and 3 WRs).  Compare to 48% in 2017 and 52% in 2016.  First down runs from 11 personnel are extremely efficient.  In the first half, they can paint the picture of a pass play, with 3 WRs on the field, and create a lighter box to run into.

Compare the efficiency of first down runs in the first half based on personnel (using the 5 most-common groupings used)

  • 11 personnel (3 WRs): 52% success, 4.8 YPC
  • 12 personnel (2 WRs): 48% success, 4.3 YPC
  • 21 personnel (2 WRs): 54% success, 5.3 YPC*
  • 22 personnel (1 WR): 34% success, 2.5 YPC
  • 13 personnel (1 WR): 41% success, 4.6 YPC

*21 personnel numbers are heavily skewed by the Patriots, 49ers and Saints. These three offenses combined to total 37% of the total rushes from 21 personnel, and all 3 were extremely strong when running from 21 (NE: 55% success, 5.4 YPC, NO: 68% success, 6.8 YPC, SF: 47% success, 6.1 YPC)

Running from 1-WR sets on 1st down was extremely unproductive, as the defense was very likely playing run.  Running from 2-WR sets had more success, but for most teams around the league, running from the 3-WR set of 11 personnel was most productive.

And in 2018, on first downs in the first half, more runs occurred occurred from 11 personnel than in any of the last 3 years:

  • 2018: 49% of all runs from 11 personnel on first down in the first half
  • 2017: 41%
  • 2016: 44%

First down passing is more efficient than first down rushing, but when teams run the ball, they need to do so intelligently.  We saw more of that in 2018, which produced much more successful rushing numbers league-wide.

Still More Room for Improvement

The efficiency brought from more first-and-10 passing in the first half was not a fluke.  But the league still must increase its usage.  In addition, teams can improve their personnel grouping deployment for even better success.

Much like rushing from pass looks is +EV, passing from run looks is +EV as well.  Compare passing efficiency by personnel grouping on first downs in the first half last year:

  • 11 personnel (3 WRs): 54% success, 7.5 YPA
  • 12 personnel (2 WRs): 54% success, 8.5 YPA
  • 21 personnel (2 WRs): 57% success, 8.4 YPA
  • 13 personnel (1 WR): 54% success, 8.8 YPA
  • 22 personnel (1 WR): 57% success, 9.5 YPA

Last year, 62% of all first down passes in the first half came from 11 personnel, with just 21% coming from 12 and 9% coming from 21.

Offenses need to look to pass more from 12 or 21 personnel on first downs.

Additionally, if we examine the way in which plays are called from these groupings, it’s clear that there is room for improvement:

  • 11 personnel sees a 57% pass rate and a 43% run rate
  • 12 personnel sees a 53% run rate and a 47% pass rate
  • 21 personnel sees a 58% run rate and a 42% pass rate

While these numbers seem logical, I would argue the NFL needs to investigate flipping them on their head. I’d be in favor of 12 and 21 personnel being used as pass groupings on first down.

Examine some of the teams that bucked the trends with the most in 2018 by passing from 12 personnel (NFL average was 47% pass from 12 personnel):

  • 67% – Denver: 64% success, 9.8 YPA
  • 62% – Indianapolis: 66% success, 9.8 YPA
  • 61% – Oakland: 65% success, 7.7 YPA
  • 59% – Kansas City: 43% success, 7.7 YPA (12.8 air YPA)*
  • 57% – New England: 88% success, 9.6 YPA
  • 57% – Atlanta: 50% success, 9.2 YPA
  • 56% – New Orleans: 66% success, 9.7 YPA

* KC often used 12 personnel for shot plays on first downs in the first half.  Their air yards per attempt of 12.8 was substantially higher than the NFL average, as well as their own passes from 11 personnel (8.0 air YPA) or 21 personnel (6.7 air YPA).  As such, their success rate was much lower on these plays.

Which Teams Called the Most Passes on First Half First Downs?

Many teams with great QB situations called passes early and often, but even teams with mediocre QB situations were passing early and often and seeing a lot more success in doing so than they did when they ran the ball.

first and 10 play calls3

Teams like the Broncos (Case Keenum) and Buccaneers (Jameis Winston or Ryan Fitzpatrick) took to the air on first downs in the first half at well above average rates and had substantially better success when doing so as compared to when they ran the ball.  When the Buccaneers passed on first-and-10 in the first half, they averaged 10.3 YPA and a 61% success rate, as compared to first-and-10 runs which averaged just 4.7 YPC and a 53% success rate.

The biggest surprise team was the Seahawks.  Despite having tremendous QB Russell Wilson in his prime, the Seahawks chose to run the ball 10% above average, at a 59% clip, the second most in the NFL.  While success rates were similar, passes averaged a full 4.0 YPA more than run plays (8.7 YPA vs 4.7 YPC).  Expect the Seahawks to continue their run-first ways since it led them to the playoffs last year and OC Brian Schottenheimer and defensive-minded HC Pete Carroll are still calling shots.  But their offense would have substantial upside if they let Russell Wilson pass the ball more on first down.

What team was balanced but blew it by not being more pass heavy?  The Chargers.  Despite me pleading with the team to go much more pass-heavy last summer, the Chargers still passed the ball at a below-average rate (50%).  These passes averaged 10.8 YPA and a 58% success rate, whereas runs averaged 4.8 YPA (league average) and a mere 44% success rate (7th worst).  If the Chargers went more pass-heavy on first-and-10 in the first half, they would be much more dangerous offensively.

Will Teams Continue to “Pass” on First Down Rushing in 2019?

Yes.  It should be a no-brainer, but nothing smart seems to be a no-brainer in the NFL.  That said, I believe it will continue.  More intelligent offensive minds are being hired or promoted to positions of importance around the league.  Better analytics are being made available and coaches are unable to deny the information, and the people with these analytics have more influence on teams and league-wide thought process.

Coaches should look to pass even more on first down in the first half than they did in 2018.  Coaches should spend time this offseason devising more pass plays with high-floors (aka high completion rate) for use on first down.  Coaches should look to 21 personnel in particular as an edge in the passing game.  The extra tight end provides a huge mismatch and allows there to be less predictability pre-snap and can provide situations were there are more max-protection passes or better downfield blocking with TEs releasing to become blockers after high percentage passes to WRs.  And coaches should look to run more from 11 personnel rather than 2 or 1 WR groupings.

We will see more passing on first down in the first half because it’s more efficient, has more upside and because the rules promote it.

Teams can try to establish the run and make the playoffs, like Chicago, Dallas, Seattle and Houston did last year.  These teams went 1-4 in the playoffs, with the lone win when Dallas beat Seattle (solely because Seattle ran the ball so often).  These 4 teams didn’t make the playoffs because they established the run.  They were good teams that would have made the playoffs regardless of whether they went over-the-top run heavy or not.  The ceiling is much higher for the Dallas offense if they pass more on first downs (56% success, 8.0 YPA) rather than run it more (46% success, 4.2 YPC), and the same is true with the other run-heavy teams.

Running as often as some of these teams did is like fighting against the tide:  it will work when the tide is low and you’re playing an opponent who is beatable with any style, so it convinces you to continue with this -EV strategy.  But when it is high tide, fighting against the rules and modern day efficiencies which favor the passing game when taking on a tough opponent won’t often end well.  These teams would be well served to reassess their 2018 tactics.

 

Why the Rams Lost the Super Bowl

By Warren Sharp

I am not a Rams fan.  I am a fan of efficiency.  I left my job as a Professional Engineer to study NFL analytics.  Last year I began consulting for an NFL team.  And since that team didn’t play in the Super Bowl this year, I used my models and analysis to break down the Super Bowl and forecast how best the Rams or Patriots could win the game.

I wrote a 40-page report on the subject which I distributed to media a week before the game.  The Patriots offense was fortunate.  They didn’t need to alter much to beat the Rams.  They would have success running from 21 personnel (2 RBs, 1 TE, 2 WRs) and if they won the matchup in the slot (Julian Edelman vs Nickell Robey-Coleman, the most critical one-on-one matchup of the entire game per my analysis) they should be able to do enough offensively to win.

But I soon realized the Rams offense was not going to be competitive unless they tweaked and modified things.  They would not be able to “do what we do” and beat the Patriots.  They NEEDED to adapt their offense to be able to defeat the Patriots.

I isolated several key elements for the Rams.  They were “Ideas to Maximize Efficiency”.   I watched the Super Bowl to see if the Rams utilized any of these ideas.  They scored 3 points and after obtaining the film and charting data, I analyzed the results.

I wrote the following Twitter thread after the Super Bowl based on my review of the film and data (click below to read the thread):

Rams SB77

Because I was asked to put this into article form, I’m going to walk through these the ideas I isolated pre-game which would help the Rams increase their offensive efficiency based on what we knew of the Patriots defense.  And then I’ll look at whether they did or did not utilize any of these ideas and the results.

 

1. The Rams must pass the ball more from 12 personnel than their season average

What I said pre-game:
Rams SB1

What happened in the Super Bowl:

The Rams had 36 passing dropbacks from 11 personnel (1 RB, 1 TE, 3 WRs).  But only 6 dropbacks from 12 personnel (1 RB, 2 TEs, 2 WRs).

When using 11 personnel to pass:  5.6 YPA and a 29% success rate

When using 12 personnel to pass:  8.8 YPA and a 60% success rate

The splits aren’t close.  But they were predictable, considering the Rams were better out of 12 than 11 and the Patriots were the #3 defense vs 11 personnel passes but were #26 vs 12 personnel passes.

One benefit to using 12 personnel is added pass blockers.  While Gerald Everett is far from a true pass blocker, he can occupy or double long enough to create an advantageous downfield matchup.

The Rams 5-man offensive line was routinely whipped by the Patriots defensive line, which typically rushed only 4 men and still recorded pressures or sacks.  Instead of shifting out of 11 personnel to add an extra TE, the Rams continued using it.

The Rams used 12 personnel with both TEs blocking only 3 times the entire game.  The first time they used it was with 3:42 left in the 3rd quarter.  To rephrase – the Rams went over 2.5 quarters before using 2 TEs to help block the frenetic Patriots pass rush.

The result of that pass?  It was the closest the Rams got to a touchdown all game, and should have been a touchdown but for a delayed release (he had so much time) by Goff:

The Rams 3 passes from 12 personnel where both TEs stayed in to block:

  • The above wide open missed TD
  • 16-yard completion on 1st down
  • 9-yard completion on 1st down

Only 3 plays.  All 3 generated open targets downfield which should have averaged 18 YPA and recorded a TD, but for the late Jared Goff pass.

The Rams seemingly didn’t care that the Patriots were significantly better defending 11 personnel.  They apparently tried to win by “doing what we do” and out-executing the Patriots.

The Patriots, on the other hand, modified their entire defensive strategy just for this game to capitalize on the Rams weaknesses.  I was asked on the radio yesterday if I though that was bold and dangerous.  I said “it was bold and intelligent” and I stand by that.  So do the Patriots, who said after the game:

Rams SB8

The flaw in failing to attack through the air more from 12 personnel likely cost the Rams the game.

 

2. The Rams must use play action often to slow the Patriots pass rush

What I said pre-game:

Rams SB3

What happened in the Super Bowl:

On the season, the Rams used play action on 35% of dropbacks. As mentioned above, I wanted the Rams to increase that rate in the Super Bowl.

But they used it on only 24% of their dropbacks in the Super Bowl.  They passed primarily from 11 personnel, and from 11 personnel, they used play action just 17% of the time.

The play action splits from 11 personnel?

Without play action:
• 5.3 YPA, 33% success, 3 sacks, 1 INT

With play action:
• 7.2 YPA, 50% success, 0 sacks, 0 INT

Looking only at early downs, from both 11 and 12 personnel, here were the play action splits:

Without play action:
• 6.5 YPA, 38% success, 2 sacks, 1 INT (10% sack rate)

With play action:
• 7.6 YPA, 56% success, 0 sacks, 0 INT

I’ve heard the Patriots “took away play action”… that would mean using play action in the 2nd half didn’t work any more. But actually, 2nd half only, all downs:

Without play action:
• 6.5 YPA, 39% success, 2 sacks, 1 INT

With play action:
• 8.0 YPA, 60% success, 0 sacks, 0 INT

 

3. The Rams must find a way to get productivity from the RB-pass game

What I said pre-game:

Rams SB2

What happened in the Super Bowl:

The Rams didn’t have a single target to a non-WR in the first half.

The Rams threw just one early-down target to Todd Gurley all game (4th qtr).

The Rams threw 3 early-down targets to C.J. Anderson, but the first was not until 4:30 left in the 3rd qtr.

A critical positional matchup edge wasn’t explored.

 

4. The Rams must run the ball more from 11 personnel, and focus on run types that mirror the Patriots weaknesses, which include outside zone (specifically from 11 personnel) and inside zone

What I said pre-game:

Rams SB5

What happened in the Super Bowl:

Heading into the game, the Rams used 11 personnel to run the ball 82% of the time.

They needed to use it at or above that rate against the Patriots, for the above reasons.  There was no reason to believe they would have an edge running it LESS from 11 personnel against the Patriots.

But on early downs (17 of 18 runs were on early downs), the Rams used 11 personnel on only 59% of run attempts.

In biggest game of the year, despite the need to run from 11, the Rams ran from 12 at their 2nd highest rate in any game this year.

The Rams gained more YPC and produced a higher success rate running if from 11 personnel in the Super Bowl, but ran too often from 12 personnel.

How about the run types?  Here’s a breakdown of the Rams run types by personnel in the Super Bowl:

Rams SB6

In addition to my belief that the Rams should run more from 11 personnel, I also thought the Rams should run outside zone from 11 rather than 12.

Indeed, in the Super Bowl, the Rams recorded more YPC when running outside zone from 11 personnel (9.0 YPC, 50% success) but ran outside zone from 12 personnel twice as often (4 times). These 12 personnel runs outside zone runs gained just 2.8 YPC.

 

5. Pace of play

I believed pace would be a huge element to this game.  The Patriots did not face very many fast-paced offenses in 2018, and the Rams were the #1 fastest paced offense.

I didn’t have any specific recommendations, but was eager to see if the Rams would utilize their pace to gain an edge for their offense.  Especially since it was nearly impossible to utilize it in the hostile crowd environment in the dome in New Orleans in the NFC Championship game.

What happened in the Super Bowl:

The Rams ran 18 plays where they snapped the ball inside of 5 seconds on the play clock.

These plays generated a paltry 1.4 yards/play (plus an INT).

The Rams ran 5 plays where they snapped the ball with 1 second on the play clock.

These plays generated 0.2 yards/play.

Snapping this late in the play clock played right into the Patriots hands.

Here is a look at every play the Rams ran (aside from plays after timeouts or change of possession, as well as the final drive of the game down 10 points) sorted by seconds left on the play clock at the snap:

Rams SB9

Aside from the two plays with less than 1 minute left in the first half and the hurry-up pass to Brandin Cooks to (smartly) prevent the Patriots from challenging a completed pass call, the Rams got off just 3 snaps (of 38) with more than 16 seconds left on the play clock.

Running the play clock down clearly was not effective nor useful.  Yet the Rams made minimal adjustments at halftime:

Their first 5 snaps out of the locker room saw them snap the ball with less than 10 seconds on the play clock every time (3 were with 3 seconds or less).

And of their final 9 snaps before the final drive, 7 of the 9 saw the Rams snap the ball with less than 10 seconds on the play clock, including 6 of 9 with 5 seconds or less.

 

The Fallout

The Patriots defense was tremendous.  They executed a tremendous game plan.  One that adapted and even offset the loss of S Patrick Chung in the 3rd quarter.

But there was a major opportunity missed by the Rams in creating their gameplan & adapting in-game.

Before the game, one could identify what the Rams needed to do based on the Patriots strengths and weaknesses coupled with an understanding of who the Rams are:

  • Pass more from 12 personnel and less from 11 personnel
  • Use more play action
  • Pass more to RBs
  • Run more from 11 personnel and less from 12 personnel
  • Use tempo

All of these attack weaknesses of the Patriots but also are not so foreign to the Rams that suggesting they implement them into the game plan would be impossible. Far from it. The Rams already do these things.  The strategy should have been to emphasize them more in this particular matchup.

The Patriots were willing to scrap everything that brought them success (man to man defense) and use something new (zone) in the biggest game of the year, primarily because they thought it game them a tactical edge.

What are you willing to risk to win the Super Bowl?

Because those elements worked in the Super Bowl.

  • Passing from 12 personnel worked
  • Play action worked
  • Running from 11 personnel worked
  • Tempo worked (as did the inverse: lack of tempo clearly didn’t work)

But all of these were severely underutilized.

(The only thing that didn’t work was early down RB-passes, but the sample size to Gurley and Kelly was 1 total target, so that wasn’t even explored enough to know if it would have worked)

The takeaway is that even a great, Super Bowl caliber offense must must adapt to attack their opponent flaws.  The takeaway heading into the 2019 season must be that while offense wins games, offenses cannot simply run the same plays and the same schemes and expect the same results over the course of the season.  Offenses, even great ones, must constantly (every single week) study with great detail the edges they have over the defense they are facing and adapt.  Their game plan entering the game must adapt and often times, adaptation must come in-game to circumstances they are facing.

The Rams didn’t have a creative enough game plan and were too tied to what worked for them most of the season.  They weren’t willing to get too far from their comfort zone.  The Patriots have no comfort zone and change anything if it is required to win a game.

I’m reminded of several quotes I used in my 2018 Football Preview book.

About being bold with your strategy to surprise your opponent:

Everything which the enemy least expects will succeed the best. – Fredrick the Great, 1712-86

Or about adapting your offense when strategies aren’t working:

Whenever you and your opponent become stagnant, you must immediately employ a different method of dealing with him in order to over come him. – The book of Five Rings, Miyamoto Musashi, 1584-1645

And finally, what Bill Belichick appears to have captured and used for success:

When Napoleon was asked what principles of war he followed, he replied that he followed none.  His genius was his ability to respond to circumstances, to make the most of what he was given – he was the supreme opportunist. – The 33 Strategies of War, Robert Greene, 2007

The Rams will learn from this.  You can blame the result partially on the loss of Cooper Kupp or an errant Jared Goff throw or simply the genius of Bill Belichick.  All of those play roles.  I prefer to accept the reality that with a different game plan or faster in-game adjustments, the Rams could have won this game even with the hand they were dealt.

I look forward to the time when all NFL teams become more willing and more open to adapting offensive strategies (even for Super Bowl caliber offenses) weekly to strike aggressively at opponent weaknesses based on pre-game study as well as making in-game adjustments in order to achieve the fastest path to victory.