In this week's Varsity Numbers, Bill Connelly revisits some measures and concepts: Adjusted Scores, Covariance, and momentum (or whatever you choose to call it).
11 Nov 2009
by Mike Tanier
Someone named Dominic Baronet allegedly "friended" a dozen women on Facebook, then impregnated them.
Dude must have one hell of an Internet connection.
According to reports, Baronet knocked two women up in the same day, a feat of reprehensible morality, enviable stamina, and amazing scheduling. The women felt betrayed and violated by the passionate -- but unsettlingly brief -- relationships. George Kokinis can relate.
Kokinis may not have been much of a general manager -- he never got the chance to do anything generally managerial -- but at least he wasn't the Browns GM who started the NFL's flame war fad. That honor goes to Phil Savage, who had the vocabulary right but lacked Larry Johnson's homophobic flair. Thinking outside the inbox got Savage fired, but at least he got to empty his desk; Kokinis didn't have time to fill it.
The Browns deny reports that Kokinis was escorted by security from team headquarters. It's more likely the Browns just deactivated his security chip, everyone's favorite petty, over-officious post-firing security measure. (Granted, it has to be done eventually, but doing it four seconds after the employee huffs out of your office is a vindictive cubicle dictator move). Shutting off access to the company email is almost as overdramatic and humiliating, though in cases like Savage's it's was a wise move. The Browns had to be cautious: they couldn't risk Kokinis sneaking out with a copy of the offensive playbook, which like most depressants has a high street value.
With the Kokinis threat neutralized, Eric Mangini is free to act as paranoid as he wants. He wouldn't reveal that Brady Quinn was his starting quarterback until Wednesday, forcing the Ravens to prepare for Quinn, Derek Anderson, and/or Brett Ratliff, preparations akin to childproofing the family room.
If my tone two paragraphs ago suggested that I've been on the business end of a few "forced resignations" then I've probably said too much, although Savage, Johnson, and Baronet might disagree. It's one of the Internet's many evils: it eliminates the important personal distance between general managers and cocky fans; running backs and guys with poorly-chosen profile pictures; vulnerable women and the jerks who stalk them; sportswriters and people who couldn't care less about their personal life. I'm as guilty as the next guy of sharing my breakfast choices and Covington rants with social networkers, though I somehow avoid the curses and slurs, even when reviewing Jason Elam or quoting Eminem. Mangini may be right: It's better to say nothing and be thought a fool. Though it's probably better to say nothing, win a Super Bowl, and be thought a genius. Or to win a Super Bowl, say everything in a wise, authoritative tone, and be thought Tony Dungy.
Dungy is back in the news; I missed the press conference in which he was appointed to a cabinet position, but why else does everyone still listen to him? Dungy thinks Michael Vick, professionally unfulfilled by his two-draw-play-per-week workload, should play for the Bills next year. Dungy's words briefly dominated the Bills blogoverse, though that could be misleading: the only Internet hotspots in greater Buffalo are the student union at the SUNY campus and a Barnes & Noble on Elmwood Avenue.
Do we really care about Vick's job satisfaction? Shouldn't he be thankful just to be out of the license plate business? Still, Our National Conscience may be on to something: Vick may still have sins to repent, and the Siberian gulags outside Orchard Park are perfect for penance. I've seen Vick shiver under a Himalaya-grade parka on ordinary winter days; A few Buffalo Decembers would make him so contrite that even PETA would forgive him.
So the Internet begot Savage, who begot Johnson, and it also begot Baronet, who is guilty of a dozen cases of Internet begetting, which should be at least twice as illegal as Internet betting. On the plus side, it begot my writing career, and at least Baronet isn't affiliated with the NFL. Yet. Any more injuries and the Seahawks will have to hold open tryouts. Baronet is the kind of guy who knows how to score an event invitation.
I didn't pay much attention to the Cardinals in the first month of the season. I wrote them off after they lost to the Niners and Colts in the first three weeks, figuring that they would either flat-line or end the season as one-and-done token representatives of a bad division. Granted, that's what they looked like last season, too, but there seemed to be even less magic in Arizona this year than last.
They've gone 4-1 since then, beating some pretty good teams along the way. It's clear that they are going to win the NFC West, that they may once again clinch early, and that while they are still unpredictable, they are in good position to be playoff noisemakers.
With the Cardinals playing well, it was time to get caught up on my homework, I watched the Cardinals-Bears game tape, focusing on the Cardinals offense, specifically Larry Fitzgerald. With Anquan Boldin hurt (physically and emotionally), Fitzgerald has become even more important to the Cardinals offense than he was last year. I wanted to see what the Cardinals were doing to protect Fitzgerald from double coverage: special formations, unusual route combinations, pre-snap motion, or some other wrinkle to keep defenses from loading up against a receiver with 87 targets.
It turns out that the Cardinals aren't doing anything special. They are splitting Fitzgerald wide, running simple plays, and allowing him to use his exceptional athleticism and technique to get open.
Into the I
Last year's Cardinals used more four-receiver sets than any team in the NFL. They only used tight ends on 58 percent of offensive snaps, by far the league's lowest rate. They used single-setback formations 68 percent of the time (7th in the NFL) and ran a higher percentage of draw plays than any other team. They were, in short, a single-back team with a pass-oriented offense.
That's why I was surprised to see the Cardinals in so many offset I-formations against the Bears. Fullback Dan Kreider got a lot of playing time, and tight ends Ben Patrick and Anthony Becht were on the field for many snaps, alone or together, even on passing downs. The run-oriented personnel groups were partially a response to the loss of Boldin -- the Cardinals used more three-receiver sets early in the game against the Panthers two weeks ago -- but they also reflected Ken Whisenhunt's commitment to improve a poor running game.
|Figure 1: Cardinals Full House|
The Cardinals were so determined to run the ball that they sometimes used a full-house backfield (Fig. 1), with Patrick (89) and Kreider (35) lead blocking for Beanie Wells (26). The Cardinals ran simple isolation plays from the full house, though they certainly have some full-house passes in the playbook. The Cardinals ran effectively against the Bears, though most of their successful runs came from great individual effort by Wells, not from the blocking or the system.
I-formations and base personnel groupings also benefit the passing game by keeping the opponent's base defense on the field. For some teams, base-on-base matchups favor the offense: the Chargers, for example, can isolate Antonio Gates on a safety or Darren Sproles on a linebacker from their base package. The Cardinals don't get much of a matchup advantage from their tight ends and backs. What they do get is a compressed field: with a tight end and two backs in the backfield, they can open up a lot of real estate between the formation and receivers Fitzgerald (11) and Steve Breaston (15). All of that empty space forces cornerbacks to cover receivers with little underneath help, and with base-on-base, there are no extra defensive backs available for exotic coverage schemes. The formations make it easy to isolate Fitzgerald in single coverage.
Let's look at several variations on the same theme from the Cardinals game tape. In each of these plays, Fitzgerald runs a medium or deep route along the sidelines, while the other receivers essentially stay out of the way. The Bears are caught in some variation on man coverage in each play. The design of the plays is very simple; it's the execution that makes them special.
|Figure 2: Cardinals Fitz Post|
In Figure 2, Fitzgerald runs a skinny post. Breaston runs a deep cross to occupy the safety, while Becht (84) and Krieder run very shallow crosses. The Bears appear to be in Man-Free coverage: the free safety covers the deep middle, and the strong safety lurks in an underneath zone, but everyone else is in man coverage. Fitzgerald beats his cornerback with a sharp inside release and a smooth cut. The timing of his route and Kurt Warner's throw are excellent: Fitzgerald doesn't bend the route too far inside (where the safety can intervene), and Warner delivers the ball while the safety is still worried about both Fitzgerald and Breaston.
|Figure 3: Cardinals Fitz Deep Comeback|
Figure 3 shows a nearly identical play from a similar formation. Once again, Breaston runs a deep route, with Becht running an in route at about 15 yards. Krieder runs a short midfield curl, and Wells a shallow cross. Against zone coverage, these would be "hold level" routes designed to keep linebackers from backing up into deeper coverage. Against man, they have a similar purpose, opening up passing lanes for the deep receivers. Fitzgerald runs a comeback along the sideline at 18 yards, and his cut is so crisp that a good cornerback -- Charles Tillman -- has no chance of staying with him. Once again, Warner's pass is on time and on target.
When diagramming these plays, I shut off the "curved lines" feature for Fitzgerald. His cuts are so sharp that they have to be represented by angles, not curves.
If Figure 4 looks like a flipped version of Figure 3, that's because I used the software's Mirror function before making minor changes. The figure shows a short completion on a comeback to Fitzgerald in the second quarter; the defensive coverage was much tighter, but Fitzgerald made a diving catch. All of the principles of the Cardinals' offset-I passing game come into play in Figure 4: Fitzgerald is isolated on his side of the field, Breaston and the tight end run deep routes, the backs run short "level" routes. Fitzgerald and Warner can do this to you all day.
The Bears got caught in man coverage time after time against the Cardinals, but they tried to give Fitzgerald's defenders help. Tillman and the other Bears cornerbacks usually had high safety help, but a deep safety doesn't do much good against comeback routes along the sidelines. The Bears were shorthanded defensively and couldn't effectively run many elaborate zones.
|Figure 4: Cardinals Fitz Deep Comeback Part II|
They were also on the wrong side of the scoreboard: the Cardinals took an early lead and could dictate terms on offense, all the while adding to it. That's why the Bears weren't able to do what the Panthers did two weeks ago. Early in that game, the Cardinals had some success executing their somewhat-balanced offense. A Julius Peppers interception of a flat pass (from the offset-I formation) and a few big running plays, however, gave the Panthers a 28-7 lead in the second quarter. The Cardinals switched to the shotgun and the no-huddle, and once they started pressing, Warner started making mistakes. Most teams won't be able to build 28-7 leads on the Cardinals –- it's shocking that the Panthers did -– so opponents need a more viable anti-Fitzgerald plan.
The Uneasy Answer
The textbook way to stop Warner and Fitzgerald from picking you apart is to flatten Warner. The Bears didn't blitz much against the Cardinals, mainly because they were low on manpower with Tommie Harris, Brian Urlacher and several others out (including safety Al Afalava, who left the game early in the second quarter). When the Bears did try to blitz, they got burned.
Figure 5 shows Fitzgerald's second-quarter touchdown in the front corner of the end zone. The Cardinals are in another personnel grouping they had little use for in 2008: a two-tight end set, with Ben Patrick in the slot. Again, this grouping keeps the Bears in a base package, and the pre-snap read suggests Man-Free coverage. The strong safety is the only wild card; from his position, he could cover Becht or the back, drop into a zone, bracket Fitzgerald in double coverage, or blitz.
The safety blitzes, but the Cardinals have a max-protect blocking scheme in place. It's another advantage of two-back or two-tight end sets: extra protection. The three-route combination should look familiar: Breaston runs deep, Patrick runs a post to attack the high safety. Fitzgerald starts by running a post, but bends back into a post-corner route. His defender reacts hard to the inside move, and the free safety stays in the middle of the field to defend the post. Warner maintains discipline, watching the safety and only turning to Fitzgerald when it's time to throw. It's another simple play, designed to get the ball to the best weapon on the field.
|Figure 5: Cardinals Fitz Double-Move Touchdown|
Fitzgerald may be the best route-runner in the NFL right now. He always gets a clean release, dictating to the defender how he'll get off the line of scrimmage. If he wants the inside shoulder, he gets it; if he wants to work the sideline, he'll get outside his cornerback. Fitzgerald stems his routes very well, with sudden moves at the top of each route. Factor in his ability to make tough catches away from his body and scoops off the ground, and it's easy to see why he's targeted a dozen times per game.
When Boldin returns, three-wideout sets will also return, but the Cardinals no longer have to spread the field to beat opponents, and they don't have to keep Boldin happy to stay competitive. Their offset-I system, two tight end formations, and wishbone wrinkles give them a viable power game. They are no threat to power out 150 rushing yards against the Vikings, but they can run well enough to move chains and squat on the lead against the Seahawks, Lions, and Rams (twice). Those four opponents should get the Cardinals to nine wins; a .500 record against the Vikings, Niners, Packers and Titans could get them a home playoff game.
As for Fitzgerald, the only cornerback with a chance to slow him on the upcoming schedule is Coutland Finnegan. The Packers will try to neutralize Fitzgerald by blitzing Warner, the Vikings can get the pressure they need from their front four. Every other team will either tie its defense in knots to stop one man or simply accept the fact that Fitzgerald is going to do some damage.
The Packers have allowed a league-high 225 yards on 37 sacks this season. The NFL mean is this year 118 lost yards. The Packers have therefore lost 107 more sack yards than the average team.
The Packers have committed a league-high 62 penalties for 509 yards, the second highest total in the NFL (the Ravens are first). The NFL mean this season is 410 lost penalty yards. The Packers have therefore lost 99 more penalty yards than the average team.
The Packers allow 13.4 yards per punt return. The NFL mean is 8.45 yards per return. That means the Packers lose 4.95 more yards of field position on punts than the average team. Spread across 22 punts, that’s 108 more yards of lost yardage.
Add up the sacks, penalties, and punt returns, and the Packers have given up 314 more "hidden yards" than the average team. Prorate for the season and that comes to 628 yards: the contribution of a good slot receiver, or a pass-catching tight end. Or about a game and a half of total offense. That's what the Packers are handing opponents.
No wonder the Packers have a great offense and defense but a 4-4 record.
The sack problem is easy to correct. Like the Steelers, the Packers have a quarterback-line codependency. The line isn't very good, and Aaron Rodgers holds the ball forever. Mike McCarthy uses a lot of six and seven-man protection, so read progression is more an issue than scheme. Rodgers must stop sampling the Randall Cunningham highlight reel when the pocket collapses: three-yard dumpoffs are good, sideways scrambles aren’t.
Problems with the return game are also easy to correct. Kick coverage is all about assignment discipline and energy. Special teams coach Shawn Slocum should be allowed to draft a few more starters and run a few more practice reps to make sure the Packers don't give up any more 80-yard kick returns at the end of close games.
Penalty problems are the hardest to cure. Football Outsiders research suggests that few penalties correlate from year to year, but the Packers were flagged 110 times last season, so they can't wait for outside factors to take the penalties away. One great cure for holding penalties is a quarterback who dumps off three-yard passes instead of scrambling sideways. Maybe the Packers can kill two birds with one stone, assuming Rodgers doesn't hold onto the stone too long.
McCarthy is certainly emphasizing these "lost yardage" plays during the practice week. He may ultimately be judged by how well he solves these problems. The Packers are good enough to compete for the NFC North, but they are fumbling through their second straight season of missed opportunities. Many factors that affect a team's record are out of a coach's control: injuries, strength of schedule, the luck of the bounce. Penalties and special teams gaffes can be controlled. If McCarthy can't do it, the Packers will find someone who can.
54 comments, Last at 19 Nov 2009, 6:37am by countertorque