Thursday, July 17, 2008

Attack of the Sack | by Pat

Now then, enough Swarbrick for the time being. Back to football.

We're getting closer and closer to the 2008 season and before indulging in any predictions, it's perhaps wise to take another look at the ugly truth from 2007.

The list of things that went wrong last season is a long one indeed. But if forced to pick just one thing that stands out above the rest, the record 58 sacks allowed by the offense is a solid candidate for #1. Last season wasn't just a bad year for protecting the quarterback, it was leaps and bounds worse than usual. Cataclysmically bad. Consider that in 2005 and 2006 combined, the Irish only allowed 52 sacks. And don't think the numbers were skewed by the Georgia Tech or Michigan game alone. Only two teams failed to sack the Irish more than their regular season average and even those two were effectively tied with their season average (BC and Duke both sacked ND twice while averaging 2.46 and 2.08 sacks a game, respectively). Everyone got their shots in last season.

This post will attempt to look back over those sacks and see what, if anything, stands out and provides a few clues for what to look for next season. Should we expect immediate improvement, or were the mistakes so pervasive that a huge leap in eliminating sacks is unrealistic?

Warning: the following is not for the faint of heart. If you are experiencing chest pains, lower back problems, or are pregnant, please be advised that BGS is not to be held liable for any consequences of reading the following.

Still here? Ok then, let's dive in. First let's take a look at some of the game situation stats involved with the sacks and look for any particular noteworthy outliers.

The Irish attempted at least 447 passing plays if you factor in the 389 passing attempts listed in the season-long statistics plus the 58 sacks. Taking a percentage, we find that 13% of Notre Dame's called pass plays resulted in a sack. Ouch. That compares with 4% in 2005 and 6% in 2006. (There were also some quarterback scrambles for positive yards that were likely called as passing plays, but this post doesn't take them into account and they shouldn't change the final numbers by very much.)

Breaking down the sacks by down, we find the following.

Situation# of sacks# passes% of total
1st down
26
172
15%
2nd down
15133
11%
3rd down
15
124
12%
4th down
2
18
11%

First down stands out slightly from the rest as the only down to feature sacks at a higher rate than the overall average. I'm hesitant to try and read too much into these numbers other than the fact that it's not surprising the offense was so ineffective given how often the Irish found themselves in 2nd and greater than 10.

Shifting from when in a series a sack occurred to when in the game they happened, we get the following.

Situation# of sacks# passes% of total
1st Quarter
11
87
13%
2nd Quarter
15120
13%
3rd Quarter
14
107
13%
4th Quarter
18
130
14%

Nothing terribly noteworthy here as the numbers are pretty uniform. On the other hand, one might have expected to see the 3rd and 4th quarter sack percentages a bit higher given that ND frequently was trying to pass its way back into games and the defense didn't have to worry as much about the run. The unfortunate truth though is that ND didn't usually wait until the 2nd half to fall far behind on the scoreboard.

Slicing the sack numbers a different way, here is the score differential at the time of each sack.

Score Diff# of sacks% passes
Winning by 1+ pts
2
10%
Tied
1311%
Losing by 1-7 pts
9
15%
Losing by 8-14 pts
8
8%
Losing by 15+ pts
26
17%

Once the Irish started to fall behind the sack percentages not surprisingly start to jump up. There is a bit of a dip between being down one and two touchdowns, but once the Irish were losing by 15 or more points, it was open season on Irish quarterbacks. It's hard to dig yourself out of a hole when nearly every fifth pass call results in a sack. Once the Irish lost the threat of the run, teams were even more effective at dropping the quarterback behind the line of scrimmage.

One last stats breakdown before we move on to the analysis of personnel on the sack total. After the abbreviated Demetrius Jones experiment (he was sacked once), Evan Sharpley and Jimmy Clausen both got their shots to lead the Irish and both had their ups and downs. Here's a breakdown of the sacks that occurred when each player was under center.

Quarterback# of sacks# passes% of total
Jimmy Clausen
34
279
12%
Evan Sharpley
23163
14%

Interestingly, Sharpley was sacked at a greater percentage. It should be noted of course that Sharpley was the QB for most of the game against Georgia Tech (1st in the nation in sacks) and all of the Southern Cal (4th) game. Then again, Clausen got some snaps against the Yellow Jackets too and had to face off against Penn State (3rd), Stanford (20th), and Michigan (31st).

One final note is the performance against the blitz. All year long we heard that the Irish offensive line couldn't handle the blitz. And while this was certainly true, the number of sacks that came against a blitzing defense versus a standard four (or three) man rush was rather surprising. 33, or 57%, of the 58 sacks last season happened when the opponent was blitzing while 25, or 43%, occurred when the defense did not. That's a very disheartening number as it paints the Irish offense in an even worse light. It's one thing to give up a sack when you are out-numbered in the backfield. It's another when you have more blockers than the other team does rushers.

Now let's move on to the more murky aspect of this analysis. I took a look at all 58 sacks and tried to see what went wrong with each one. As much as possible I tried to find the breakdown that led to the sack and assign "blame" accordingly. I'm going to stress right now that the following should be considered a rough guideline more than any sort of 100% accurate recap of what truly went wrong. One of the major things working against me was identified by Charlie Weis in a past interview with Mike Frank over on Irish Eyes.
“For example, offensively you talk about protection breakdowns. What were the cause of protection breakdowns? To a fan in the stands it just looked like somebody was just getting beat the whole time. In reality it might’ve been that guy was expecting a double team. He might’ve been expecting outside help. So he’s playing with inside leverage because he expects somebody outside of him that’s helping , and that guy outside of him might not be there. So there’s a lot more to it than meets the eye in several instances.
Caveats aside, let's move on and see how the numbers broke down.

Position at fault# of sacks
Quarterback
24
Offensive Line
26
Running Backs
3
Tight Ends
2
Solid defensive play3

Quarterback - 24 sacks

Good news, bad news. The good news is that our offensive line wasn't at fault for every sack. Over half of the sacks from last year weren't necessarily their fault. The bad news is that there are still Irish players at fault, in this case primarily the quarterbacks.

Usually when a quarterback is to blame for being sacked, the major mistake is holding on to the ball too long in the pocket. It's tempting and certainly admirable to want to hang in there until an open receiver can be found. But each and every successful quarterback has a finely tuned internal clock that let's them know when to get rid of the ball before taking a hit for a loss. ND's quarterbacks struggled with this last season.

The flip side to this is that if receivers aren't getting open, the quarterback has to wait longer to throw the ball and that also leads to what are commonly called coverage sacks. So it stands to reason that while Sharpley and Clausen are getting blamed here for these sacks, the wide receivers and their inability to get open at times deserve some of the blame as well.

Splitting the total between Sharpley and Clausen, Clausen was responsible for 16 sacks compared to 7 for Sharpley and 1 for Jones. Sharpley's major mistakes tended to be hanging in the pocket just a bit too long.
It's clear he's a fighter and will keep trying to make the play. And he did have some exciting escapes from near sacks that he turned into completions and first downs. However the tendency to hang in there as long as possible is a disadvantage as much as it is an advantage when under constant pressure. For every great hustle play to keep things alive, a quarterback is likely to get sacked more than once.

Clausen on the other hand was to blame at times for reasons that can be explained as (freshman?) mental mistakes. He picked up six sacks alone by running out of bounds behind the line of scrimmage, most notably in the Stanford game when he did it three times. That's over one third of his sack total that could have been avoided had he tossed the ball into the stands before stepping over the line. As with Sharpley, it is certainly admirable to try to make the play down to the bitter end. But there are times when a QB needs to learn to cut his losses and start over again. Hopefully this is one area where experience will make an impact. That he was still doing it in the last game of the season isn't exactly comforting, so hopefully he's spent plenty of time this off-season in the film room.

If you're looking for examples of coverage sacks where the receivers apparently aren't getting open to the satisfaction of Clausen, here is video of two Penn State sacks and a UCLA sack. Notice in all three clips that Clausen has time to throw to his primary target, but doesn't like what he sees. (Also note that on many sacks, more than just one player screwed up. But in the interest of keeping this as simple as possible, I stuck with one player per sack for nearly every sack.)

Looking to how this applies to this coming season, I do expect Clausen to manage games at a much better level in 2008 as compared to when he was thrown into the mix as a banged up freshman. Even if the offensive line doesn't improve, Jimmy's improved decision making should cut down on the number of sacks allowed by the Irish next season. He's also just mobile enough to turn a few sure sacks into positive yardage. Improved route running and experience by the wide receivers likewise will help to cut down on coverage sacks. Even if he stops running out of bounds behind the line of scrimmage that will have a large impact on the sack total.

Offensive Line - 26 sacks

Lineman# sacks
Paul Duncan
6
Sam Young
5
Eric Olsen
5
Mike Turkovich
3
Dan Wenger
2
Matt Carufel
2
John Sullivan
1
Chris Stewart
1
All OL
1
Some surprising and not-so-surprising numbers when we look at the sacks that were the fault of the offensive line. 11 of the 23 sacks came at the expense of the Irish tackles and that shouldn't be shocking. Sam Young and Paul Duncan usually faced the toughest pass rushers that opposing teams had to offer. It does raise concern for next year since Duncan and his team-leading 6 sacks allowed will be protecting Clausen's blind side. The good news/bad news is that not all of his sacks were the result of not being athletic enough. A few of the sacks he let up were the result of being bull-rushed or mental mistakes as he appeared to turn the wrong way or failed to engage the right rusher. Better technique through additional practice and film study will help with the mental mistakes while another year in the weight room will help him stand up to bull rushes. Still, there were two big examples of Duncan failing to pick up a speed rushers against Penn State that resulted in a downed Clausen. Until he proves otherwise, his ability to protect Clausen's blindside against these speed moves will be one of the biggest worries heading into the 2008 season.

Center John Sullivan was only at fault for one sack last year, which shouldn't be too surprising as the center is rarely the point of attack for opposing defenses. Still, while he had trouble with things like a shotgun snap, Sully did do a good job protecting the quarterback. His replacement for 2008, Dan Wenger, was at fault for two sacks but both of those came while he was at guard, not at center. They also happened in the first two games of the season and he didn't let up any more over the six games he played the rest of the season.

As for the guards, Eric Olsen let up the most on the line. Most of his were more related to not knowing who to pick up more than just getting overpowered on the line. In fact, his first two sacks of the season came when he was playing tight end (Georgia Tech) and right tackle (Michigan State) and just didn't know who to pick up. It's likely he didn't spend much practice time at those spots. Now that he's settled into guard hopefully the mental mistakes will be cut down. Turk had the next highest sack total, but at least one of them should probably be blamed on Charlie, not Turk. First and goal from the UCLA 1 yard line is not the time to call for a relatively slow guard to pull and be expected to pick up an All-American pass rusher like Bruce Davis. Not when Turk is far better doing something like this . The lone sack that Chris Stewart gave up came when he fell for a swim move against Duke.

Sam Young's time at left tackle showed that he did have trouble with speed rushers, so hopefully now that he is at right tackle he won't be placed in similar situations as often. It is a bit of concern that even as late as the last game of the year he was still showing poor technique at times. Perhaps that was a consequence of his injured wrist. Still, much is expected of him heading into 2008 with the move to the right side of the line, the weight and strength gain, and the stories of him meeting with Charlie to take on a larger leadership role.

Where does this leave us for next year? Well, as Charlie pointed out in his spring presser, experience is one of the most valuable traits to have in an offensive line. With another year of lifting, practice, and film review under their belt, the OL should cut down on the mental mistakes. If not, then Coach Latina's seat will get hotter sack by the sack. Physically, the left tackle spot is still a worry as Paul Duncan has yet to show he can handle speed rushers. Hopefully he, or perhaps Matt Romine, will be able to rise to the occasion. If Chris Stewart sticks in the starting lineup he will be question mark too as the only lineman yet to start a game. Teams no doubt will key on him until he proves his worth.

One thing that did stick out as a result of this OL sack review was the horrible cut blocking performed by the Irish. More than a few sacks were the result of an Irish offensive lineman diving at the legs of an opposing defensive lineman and just failing to slow him down in the least. Consider the worst sack the Irish gave up all year. The entire line tried to cut block Michigan's line and failed. The result was this. This is the one sack that I blamed on the entire offensive line.

The cut blocks did stop being a reason for the sacks after the Purdue game, which means that the OL either got better or ND just stopped using them as much. Coach Weis said after the season that even with full contact practices the offense will not be cutting the defense, so either the team is going to need to somehow keep getting better from non-contact drills (and I have my doubts if that is reasonable to expect) or the coaches will need to continue to scale back the reliance on cut blocks in 2008.

Running Backs -3 sacks

Given how young our backfield was, I was a bit surprised that the running backs only accounted for three sacks. I have two thoughts on this. One, the point that Charlie made about protection schemes probably applies here and I mistakenly applied a sack here or there to the OL when in reality it was the running back who wasn't where he was supposed to be. Two, Junior Jabbie deserves a bit more credit from Irish fans. Jabbie was the one most frequently in the game when ND went to the aerial attack and he did a very good job. Asaph Schwapp's name isn't on this list either which might surprise some folks. He certainly had his issues when it came to run blocking, but none of his whiff blocks resulted in a sack in 2008. James Aldridge, Travis Thomas, and Armando Allen all let up a sack apiece with Allen's being the infamous Ram Vela flying leap.

It's reasonable to assume the backs will all be better pass blockers in 2008, but at the same time, ND will likely miss the contributions of the unheralded Jabbie. Physically, Aldridge and Hughes have the strength to stand up any blitzing linebacker, so whether or not they can be effective will come down to how well they pick up the roles and responsibilities of pass blocking. With Coach Haywood having his plate filled with newly minted Offensive Coordinator duties this fall, hopefully he will still be able to consistently find time for individual instruction during practice. (I don't see any reason why not.)

Tight Ends - 2 sacks

Tight end is another category that somewhat surprised me. John Carlson had some trouble at the point of attack with a few linemen and linebackers in '08, but only twice did he directly lead to a sack. And even one of those was a bit iffy.

Getting back Will Yeatman for the fall is a big boon to the blocking capabilities of the tight end position. He'll probably spend a lot of time this season on the left side of the line helping out Paul Duncan. While he isn't the receiver that Carlson was, he is a better blocker and should help to keep Clausen's jersey clean. Mike Ragone will definitely see plenty of the field next season and will need to avoid the mental mistakes that usually accompany a big jump in playing time. He did show a passion for blocking, so with improved technique there isn't much worry about his ability to pass protect when needed.

Conclusion

At the end of this haunted horrors of a post I imagine the 2008 season predictions of quite a few Irish fans just went down by two or three wins. I don't blame you. The line was as bad last season as I've ever seen it. The play-calling at times failed to take advantage of what opposing teams were doing on defense. The quarterbacks showed their inexperience by taking unnecessary sacks. Through mental mistakes and sloppy technique, the line let defender after defender rush right by them. All in all it was borderline unwatchable. Even worse, the analysis reveals that the offense had nearly as much trouble with a normal front four as it did an all-out blitzing offense.

If you're looking for silver lining for next year, there are a few things we can point to. The experience factor is a big one. Another year of playing will have a positive effect on reducing mental mistakes. Jimmy likely won't run out of bounds with the ball in his hands nearly as much or try to force things that just aren't there. Hopefully an off-season in the film room will improve his ability to read the defense, anticipate the blitz, and if needed check out of a bad play. There will still be some growing pains at the quarterback position, but Clausen should easily be able to drastically reduce the number of sacks from 2007 where he was primarily to blame.

Another positive lies with the receiving corp. As they ingest more of the playbook, young players like Duval Kamara and Golden Tate will see the field more and provide a more athletic threat than some of the older players. Contributions from someone like Michael Floyd will only help as well.

The biggest target of criticism, the OL, definitely will have all eyes on them. Paul Duncan in particular will be under constant scrutiny. The coaching staff certainly can help the line by calling more running plays, but ND will need to pass the ball eventually. And when they do call a pass, a drastic pass blocking improvement over last year is paramount if ND wants to even be bowl eligible. After watching the line start fights during spring practice, hopefully the chip on their shoulder carries through to the season and they take out the frustrations from last year on the 2008 opponents.

There is definitely reason for optimism given their added size and strength, returning experience, and rumored improved chemistry. Still, they would probably be the first to admit they have a long way to go. Facing off against the suddenly blitz-happy ND defense in practice (hopefully at least some of the time in 1st team vs. 1st team situations) can only help the line build confidence and help slow things down during the game. But as with most of the rest of the offense, it's only logical to remain cautious. If you're looking for predictions, I'll stay away from any hard and fast number. The most likely result is that ND makes major strides in protecting the quarterback, but still makes enough mistakes to get Clausen knocked around at a rate greater than what Quinn had to endure in either '05 or '06. How that translates to wins or losses we'll just have to wait and see. Thankfully, we won't need to wait much longer.