BGA Weekly: A Recipe for Stuffing

Over the course of the offseason, I’ll be writing a weekly analytical piece, looking at statistical data from the 2011 season and revisiting some of my analysis from 2010 to see if any trends identified are still holding, or whether I can identify any new ones. As always, I would appreciate any suggestions for future BGA pieces.

As we get closer to the season, one aspect that has been discussed and deliberated in depth is the potential for the new defense to run more four man fronts. However, I’ve been wanting to look at a defensive package, which – even though you might only see it 15-20 times per season on average – can have a dramatic effect on the outcome of a game. How about a SIX man front? Yes, this week, I’m going to look at the Jets goal line defense.

After the jump, I’ll be looking at how the Jets approached goal line situations on defense, the personnel they used and their success in running such packages over the last few seasons. I’ll also be doing what I can to evaluate this performance against other teams.

The Goal line Package

The standard goal line package the Jets run is a 6-2-3. That features, as you’d expect, six down linemen, two linebackers and three defensive backs. The Jets will typically run this inside the five yard line, but not exclusively. I’ve seen them run it on first and goal at the five, but I’ve also seen them not run it on first and goal from the two or third and goal from the one. I guess that depends on the prowess of the other team’s running game, as well as the personnel they enter into the game.

The job of the tackles on the inside is often just to get penetration, even if the run comes in their direction. It’s up to the ends to pinch in or the linebackers or safeties to meet the ball carrier in the hole – the size of which is likely to be determined by how well the tackles penetrate. That penetration might involve meeting a lineman head-on and trying to drive him back or could involve trying to shoot a gap. It’s often enough just for a lineman to hit a gap and dive into the backfield because that can slow a runner down by redirecting him enough to affect the angle at which he hits the hole and the forward momentum he can create.

Exceptionally, if an interior lineman is able to get enough leverage, he’ll be able to shed the block and react to the direction of the run to make the tackle himself or redirect the runner further. There will usually be one defensive back on each edge. They can use their speed to either pursue a run to the outside or to get upfield and prevent the outside run by forcing the runner inside. The two linebackers and the other defensive back (or sometimes a third linebacker) have to react quickly to be able to stuff runs inside the tackles and also potentially chase to the outside to string out any outside runs. It generally takes more than one guy to bring down the ball carrier in short yardage situations, so they also have to get to the ball carrier as quickly as possible once the first tackler gets there. Although they have the defensive backs on their outside shoulder, the ends still have a crucial role in terms of setting the edge and ensuring they don’t get caught on the inside, especially since the defensive backs might be drawn off the edge by a tight end leaking out for a potential pass.

Personnel Packages

When the Jets go to their six man front, this will comprise four interior linemen and two ends. One of the ends will typically be a DE/OLB hybrid like Bryan Thomas and Calvin Pace, so the fact they don’t always have six active defensive linemen doesn’t necessarily matter. David Harris and Bart Scott would typically be the linebackers and the defensive backs would be Revis and two safeties. In theory, you could end up with Cromartie being replaced by another lineman and otherwise the personnel could be no different from the starting lineup.

Let’s look at who they used in these packages last year:

Defensive linemen (number of snaps in parentheses)

Pouha (19), Dixon (19), Wilkerson (17), DeVito (14), Pitoitua (13), Pace (13), Tevaseu (7), Ellis (6), Thomas (4), McIntyre (2)

This breaks down as follows:

Interior linemen:

Pouha (19), DeVito (14), Dixon (12), Wilkerson (11), Pitoitua (7), Tevaseu (7), Ellis (6)

Note: Tevaseu and Ellis got all their playing time when DeVito and/or Pitoitua were out injured.

Ends:

Pace (13), Dixon (7), Wilkerson (6), Pitoitua (6), Thomas (4), McIntyre (2)

Note: Thomas replaced Pace on this unit when he got hurt. They did not play together.

Linebackers

Scott (19), Harris (14), Mauga (6), Bellore (1)

Note: Harris would normally be on this unit all the time, but he missed three goal line plays against San Diego in week seven and two against Miami in week 17 with injuries. Mauga replaced him for those five snaps. Mauga and Bellore each also had one rep where the Jets went with two DBs instead of three and an extra linebacker.

Defensive Backs

Pool (19), Smith (19), Revis (16), Strickland (1)

Note: It’s perhaps not surprising that the diminutive Jim Leonhard came out of the game on these plays(along with Antonio Cromartie, who is not known for his tackling).

I won’t break down the personnel in 2009 and 2010 in as much detail, but here are some observations:

2010 Personnel

- Pouha, DeVito, Thomas, Harris and Scott were ever-present
– The other three spots on the line were usually taken by Gholston along with two offensive linemen, Rob Turner and Wayne Hunter
– The defensive backs were mainly Eric Smith, James Ihedigbo and Revis

Although initially unplanned (after a Kris Jenkins injury early on in the first game), the move to include Hunter and later Turner in this group did make some sense. Hunter is a converted DT and Turner had played the same role at New Mexico in college and was also a defensive tackle on the field goal block unit.

One other interesting observation is that Shaun Ellis only featured on this unit once and Trevor Pryce not at all. The fact they had backups and offensive linemen in the game instead suggests the Jets saw this as a chance to rest some of their older players but did not feel this would adversely affect the unit’s efficiency. In every other way, Wilkerson’s 2011 role closely masked that of Ellis in 2010.

2009 Personnel

- Again Pouha, DeVito and Harris were ever-present. Scott and Revis were in for all but one play
– Howard Green, Marques Douglas and Kris Jenkins had the rest of the reps on the interior.
– Ellis did feature on this unit most of the time, sharing most of the reps at end with Thomas and Pace
– Rhodes and Leonhard were on this unit for most of the year, but were replaced by Ihedigbo and Smith respectively at the end of the season

2011 Numbers

Let’s see how the Jets fared when this unit entered the game in 2011:

In 19 plays, the Jets gave up six touchdowns.

However, if you stop a team on first down, but they score on the next play, that isn’t as good as actually forcing them to settle for a field goal (or even failing on fourth down or turning the ball over). So, let’s also consider whether or not they kept the team out of the end zone on the series where they used a goal line defense:

Six touchdowns, three stops and one series where they scored on third down against a non-goal line defense.

Before we try to determine how good that is on a league wide basis, let’s compare it with the numbers for 2010 and 2009.

In 2010, they had 15 plays and gave up 11 touchdowns. Disappointingly, that translates to 11 touchdowns and zero stops.

In 2009, they had 21 plays and gave up five touchdowns. Apart from the five touchdowns, they had four stops and one series that ended with a touchdown on third down against a non-goal line defense. There was one other series that didn’t really count because the other team opted to take a knee and then kicked a game winning field goal.

On the basis of the research so far, the goal line package clearly fared much better in 2009 and 2011 than in 2010, where they had no real success whatsoever. Maybe this is a place where the Kris Jenkins injury on opening day had a bigger effect than we realized. Also, the decision to go with Hunter and Turner while leaving the likes of Ellis and Pryce on the bench perhaps did backfire after all.

In both 2009 and 2011, the defense did a seemingly solid job of keeping the defense out of the endzone, although there were plenty of examples where they resisted at first but couldn’t ultimately prevent the touchdown. Until we look at some data from around the league, it’s not easy to evaluate how well those units did. Before we do that, let’s look at which individuals saw success within these packages.

Individual Success

While it isn’t always the best indicator of who does the best job, since some of the defensive tackles are not actively trying to make the tackle themselves, let’s consider who made the actual stops in each year:

2011: Scott 3, Pool 2, Smith 2, Several players with 1 (and three incomplete passes)
2010: Nobody had more than 1 (including one incomplete pass)
2009: DeVito 4, Harris 2, Several players with 1 (and three incomplete passes)
2009-2011: DeVito 5, Scott 4, Smith and Harris 3, Pool and BT 2, Several players with 1 (and seven incomplete passes)

2009 is where DeVito first started to emerge and make a case for himself as a potential starter, having backed up Marques Douglas during the season. He has always been good at shedding blocks and making plays, so it’s perhaps not a surprise to see him make more plays than anyone else, especially with four coming when he was an unknown quantity that teams may have been trying to exploit by running at him.

It’s also interesting to see if we can identify any trends that seem to show the package worked better or worse with certain players in.

- In 2011, with Pace in, the Jets allowed just three touchdowns on 13 plays. His numbers in 2009 were identical, but in 2010 he only featured twice, with touchdowns scored on each play. Overall, that’s eight touchdowns on 26 plays (31%)
– In 2010 and 2011, they gave up 14 touchdowns on 19 plays with Thomas in. Overall, they gave up 18 on 34 plays (53%)
– They fared well when the backups played in 2011, as they only gave up one touchdown on Ellis’ six plays and one on Tevaseu’s seven
– On the opposite end of the scale, they allowed no touchdowns on the six plays that Pitoitua was out of the lineup, so with him in they gave up six touchdowns on 13 plays
– In 2009, changing the safeties over didn’t make a significant difference
– With Jenkins in the lineup they gave up four touchdowns and only had one stop. Three of the four stops in 2009 came after he went down for the year.

The last stat is perhaps the most revealing one. If we want to attribute the struggles in 2010 to Jenkins’ absence, we should at least acknowledge the fact that the goal line defense actually got better after his 2009 injury.

Where Do They Rank?

Here’s where it gets difficult because, as far as I can see, nobody tracks the data for goal line defenses around the league. However, we can do our best to piece together some bits of data to try and determine how well they performed.

The best data I was able to find for goal line defenses was from this fantasy football site. Clearly they were tracking how well teams performed against the run so that they could advise people whether it was a good idea to start the short yardage backs playing against those teams in any given week. As you can see from the data, at this stage – the halfway point of the 2010 season – these varied from the Bengals giving up seven touchdowns on nine goal line plays to the Redskins giving up zero in nine plays.

Unfortunately, all we have is that data, as I can’t seem to find complete data beyond that point. However, a half-season’s worth of data gives us 231 goal line plays (which another page on the site tells us is defined as plays in the red zone requiring less than three yards). This is a significant enough sample to at least give us a ballpark figure for what the expected success rate should be, provided we assume that this was representative of a normal year and wasn’t a down year for short yardage backs or anything.

Adding up the numbers, 44% of goal line plays resulted in touchdowns over that half of the season. So let’s build in a bit of a buffer and assume that the average falls somewhere between 40 and 50%. Clearly the Jets did better than this in 2011 (holding their opponents to a 32% success rate) and much better in 2009 (24%) but did an awful job in 2010 (73%). The three year average was exactly 40% with the Jets below average when Thomas played the end position, but above average with Pace there.

We could also consider how successful they were in terms of preventing their opponents from getting in the endzone on each series. They gave up touchdowns on 60% of their defensive possessions where they used a goal line defense in 2009 and 70% in 2011, but 100% in 2010. Perhaps the best indicator we can look at here is red zone defensive percentages. You’d expect the scoring percentages when teams actually got down to the goal line to be much higher than the overall red zone percentages but from 2010 to 2011, there were actually 10 teams that allowed an overall success rate of higher than 60% and one (the 2010 Eagles) that allowed a success rate of 78.62%. It shouldn’t be a surprise that the Jets red zone defense was worse in 2010 (59%) than 2011 (52%).

One final set of numbers we can use to try and evaluate their performance are the power success numbers from Football Outsiders. Power success includes third and fourth downs with one or two yards to go, but also takes into account goal to go situations at the two or closer. While it again doesn’t exactly match our data set because teams don’t generally put their goal line package in for a third or fourth and short play outside the red zone (in theory they could, but the Jets didn’t during any of the three years analyzed), this is another set of data to give us some reasonable expectation of success rate in goal line situations.

As you can see from the data, the league average was typically in the 60-65% range. Again this would mean that 2009 and 2011 were well above average. It would also mean that 2010 was below average. Sure enough, the rankings for power success reflect this with the Jets 7th in the NFL in 2011 and 2nd in the NFL in 2009, but 25th in 2010. Once again, these numbers don’t just comprise the goal line situations, but it’s clear the performance of the goal line unit made a significant contribution in each year.

Conclusions

It wasn’t easy, but I think I’ve compiled enough evidence to demonstrate how good the Jets goal line packages were in 2011 and 2009 and also that 2010 was a down year.

New defensive line coach Karl Dunbar now joins the team with most of the unit’s key personnel still in place, so it will be interesting to see if he can build on last year’s performance in this area. While you might only use this package once a game on average (and not at all in some games), it can make or break your season. After the 2010 postseason, if there’s anyone who knows the effect a goal line stand can have, it’s Jets fans.

Thanks to PFF for providing us with exclusive access to data not available to the public that was used to compile this research.