Tuesday, January 27, 2009

The Lions' new coach Jim Schwartz: football pragmatist?

The Detroit Lions hired former Tennessee Titans defensive coordinator Jim Schwartz to be their new head coach. A daunting gig, to be sure: achieving success in Detroit might be beyond any coach's realistic hopes. But, insofar as they might have a shot, this is an excellent hire.

There are the obvious and ESPN-ready reasons to hire Schwartz: (a) that he has been defensive coordinator with the Titans under Jeff Fisher, running one of the league's best units there, and (b) that he has worked with Bill Belichick, which come NFL hiring time is like holding a golden ticket.

But there's a better reason, and it is one that should give Lions fans at least a glimmer of legitimate hope: the guy has a brain. Yes, he has a degree from Georgetown, which puts him ahead of most NFL coaches, but more importantly he has proven that he has an inquisitive, analytical mind, which is all-too-often in short-supply in the NFL.

This past fall, the New York Times ran an article on Schwartz saying he was like the NFL's version of Billy Beane, the empirically minded general manager of the Oakland A's made famous (and in some circles, infamous) in Michael Lewis' great book, Moneyball. Beane, as you may remember, helped revolutionize baseball by favoring detailed statistical analysis to aid him in determining his draft picks, batting order, and pitchers. It famously led him to pick up and use guys no one else had any interest in or had even heard of.

(A running theme in Moneyball was Beane's repeated failed attempts to trade for some then-unknown minor league player for the Red Sox that he nicknamed the "Greek God of Walks" for the player's ability to repeatedly get walked more than just about anybody, while also driving up the pitch-count and consistently getting on base. He even had trouble getting him because the then Red Sox's front-office couldn't even remember that he was on their roster. That player? 2008 All-Star Kevin Youkilis.)

The other thing Beane did was win against the odds. The A's repeatedly made the playoffs despite having a payroll a mere fraction not only of juggernauts like the Yankees, but most other teams in their division and around the league. Lewis' answer to the question "How was Beane doing it?" was that Beane was outsmarting his opponents. It was not necessarily that he was smarter, but his approach was: the A's were willing to do away with "common wisdom" and even the kind of impressions most scouts give regarding a prospect: "Wow, look at the guns on him. He just looks like a baseball player." As a result, the A's routinely beat teams with payrolls twice theirs. And, now, the so-called sabermetric revolution has almost entirely swept through baseball. Even teams that don't rely on it as heavily as the A's still have some guys with laptops and Ivy League degrees slipping around their front offices these days. (The Red Sox too are now somewhat considered a Moneyball based organization, though one with a rather large payroll.)

But football is a different animal. On the one hand, football coaches and aficionados were engaged with advanced statistics long before baseball. Virgil Carter, former quarterback under Bill Walsh, actually computed the "expected value" of field position back in the 1960s. (He actually did it while enrolled part time in Northwestern's MBA program while also a player with the Chicago Bears.) Yet, as I have previously written, football is the most complex sport of all. You cannot model the game as a series of one-on-one battles as you can with baseball; indeed, the goal for both offense and defense is often to get two on one or three on two. But that has led far too many coaches, fans, and commentators -- maybe it is the machismo, maybe it is just the complexity -- to denounce and deride statistics out of hand, without basis.

Enter Schwartz. As the New York Times reported:

Schwartz, now the defensive coordinator for the Tennessee Titans, had an economics degree from Georgetown University, an abiding fascination with statistics and a preference for watching game film over television. That made him a kindred spirit with his first N.F.L. boss, Bill Belichick. But when Schwartz told Belichick his findings from an early N.F.L. research project almost 15 years ago, Belichick said he did not believe him.

“Fumbles are a random occurrence,” Schwartz said he told Belichick. “Being able to get interceptions or not throw interceptions has a high correlation with good teams. But over the course of a year, good teams don’t fumble any more or less than bad teams. Bill didn’t agree. He said, ‘No, good teams don’t fumble the ball.’ But actually, they fumble just as often as bad teams.”

With the Titans, Schwartz once encouraged the former offensive coordinator Norm Chow to run more on third-and-short because his research indicated that it was more effective than passing.

Unorthodox thinking like that has earned Schwartz, 42, a reputation as one of the N.F.L.’s leading practitioners of statistical analysis — “Moneyball” for the shoulder-pad set — using them in coaching the defense for the league’s only unbeaten team . . . . Belichick regards Schwartz as one of the smartest coaches he has been around.


As the Times points out, however, in the NFL, being known for your analytical skills is, strangely enough, not always a plus:

But being known as a “stats guy” is not necessarily a compliment, because statistics do not hold the romantic place in football that they do in baseball. Although every coach uses plenty of data — the Titans’ Jeff Fisher tracks how long his team takes to break the huddle — football is unlikely to bestow statistics-driven celebrity on anyone the way the baseball book “Moneyball” did on Billy Beane, the general manager of the Oakland Athletics.


(Of course, as Salon's King Kaufman points out in his article "Ignorance is not a sportswriting skill," baseball isn't always that enlightened either.)

In a previous article I discussed the anti-stats view, which I said can be described as nothing but neanderthal in nature. But sometimes it is just inertia and an unwillingness to be beholden to anything that doesn't seem "up front" or real; to these people, they feel like they have the experience and perception to "just know" -- Hey, it's common sense. But as Lord Keynes warned, what many call "common sense" is often just some past blowhard's own shoddy analysis or comment preserved and repeated over time, without examination.

"Practical men, who believe themselves to be quite exempt from any intellectual influence, are usually the slaves of some defunct economist. " - John Maynard Keynes


Think of the myriad examples, like "balance," play-calling, spiking the ball, or going for it on fourth-down. This is why Schwartz offers some hope. He may not succeed in Detroit, but, to me, he appears more likely to do so than anyone else. He will no doubt attempt to go in there and put together the best possible plan, not just a collection of truisms and cliches: Orwell's advice about writing can be applied to putting together a football team and organization; he discusses reading a writer who, in his opening paragraph, appears to have something important to say. But the rest of the writer's piece collapses because whatever fresh thought he began with was quickly replaced with a series of tired cliches and overused metaphors. The final product was thus imprecise, impersonal, and banal.

The analogy works for coaches. You might be talented and have a vision, but if all you can say is that we're going to "outwork our opponents," "we're going to have balance," "we will establish the run," or "we will be more disciplined than our opponents," then you're in trouble. All those are worthy goals (mostly), but they aren't always particularly constructive. Indeed, although in high school, maybe you can do these things because you might know the game more and be better organized, the NFL is a different animal. In the NFL, if a guy doesn't work he's cut; if a coach doesn't win he's fired. Rod Marinelli's Lions rarely lacked effort. They simply lacked wins.

In gleaning other hints about Schwartz's mindset and approach, I saw his name arise in the context of another Michael Lewis piece, this one about Texas Tech's Mike Leach:

At least one N.F.L. defensive coordinator, Jim Schwartz of the Tennessee Titans, had stumbled upon Texas Tech accidentally and said, Oh, my. The surprise runner-up in the search earlier this year for a new San Francisco 49ers head coach, Schwartz had scrambled to answer a question: if he got the 49ers job, whom should he hire? He was just in his mid-30's, and his football career stopped at Georgetown (where he graduated with honors in economics), so he really hadn't thought about this before.

The 49ers had not bothered to interview college coaches for the head-coaching job in part because its front-office analysis found that most of the college coaches hired in the past 20 years to run N.F.L. teams had failed. But in Schwartz's view, college coaches tended to fail in the N.F.L. mainly because the pros hired the famous coaches from the old-money schools, on the premise that those who won the most games were the best coaches. But was this smart? Notre Dame might have a good football team, but how much of its success came from the desire of every Catholic in the country to play for Notre Dame?

Looking for fresh coaching talent, Schwartz analyzed the offensive and defensive statistics of what he called the "midlevel schools" in search of any that had enjoyed success out of proportion to their stature. On offense, Texas Tech's numbers leapt out as positively freakish: a midlevel school, playing against the toughest football schools in the country, with the nation's highest scoring offense. Mike Leach had become the Texas Tech head coach before the 2000 season, and from that moment its quarterbacks were transformed into superstars. In Leach's first three seasons, he played a quarterback, Kliff Kingsbury, who wound up passing for more yards than all but three quarterbacks in the history of major college football. When Kingsbury graduated (he is now with the New York Jets), he was replaced by a fifth-year senior named B.J. Symons, who threw 52 touchdown passes and set a single-season college record for passing yards (5,833). The next year, Symons graduated and was succeeded by another senior - like Symons, a fifth-year senior, meaning he had sat out a season. The new quarterback, who had seldom played at Tech before then, was Sonny Cumbie, and Cumbie's 4,742 passing yards in 2004 was the sixth-best year in N.C.A.A. history.

... Whoever played quarterback for the Texas Tech Red Raiders was sure to create so much offense that he couldn't be ignored.

Schwartz had an N.F.L. coach's perspective on talent, and from his point of view, the players Leach was using to rack up points and yards were no talent at all. None of them had been identified by N.F.L. scouts or even college recruiters as first-rate material. . . . Either the market for quarterbacks was screwy - that is, the schools with the recruiting edge, and N.F.L. scouts, were missing big talent - or (much more likely, in Schwartz's view) Leach was finding new and better ways to extract value from his players. "They weren't scoring all these touchdowns because they had the best players," Schwartz told me recently. "They were doing it because they were smarter. Leach had found a way to make it work."

. . . This offense was, in effect, an argument for changing the geometry of the game. Schwartz didn't know if Leach's system would work in the N.F.L., where they had bigger staffs, better players and a lot more time to prepare for whatever confusion the offense cooked up. On the other hand, he wasn't sure it wouldn't.


The takeaway is not that Schwartz will be hiring Mike Leach as offensive coordinator. (In fact, he has hired Scott Linehan, which -- despite some outcry -- is generally a good choice. Many coaches, including Urban Meyer, still think Linehan is a bright, bright guy, and as a coordinator his offenses always put up points.) But the fact that he'd even consider back then should be heartening: it means Schwartz is looking for results backed up by the numbers, appearances and cliches be damned. Leach's offense looks screwy: the linemen are linemen split out wide, four receivers line-up on nearly every play, yet it gets results, and results with inferior talent at that. Beane was derided and mocked when he'd pick up a pitcher with a funky sidearm delivery where the ball was released only a few inches from the ground -- a release no scout could ever condone. Yet the derision would fade when they realized that few hitters could hit that crazy delivery, at least for a time. Beane was shopping for discounts.

Schwartz, as head coach of a struggling team, no doubt will be looking at the bottom line, and he too will have to shop for discounts. Traditional or different, he wants results. He appears to be a pragmatist. In that job, he'll have to be.

Winning football games on Mondays through Saturdays

One part of the Times piece on Schwartz particularly struck me:



"Sometimes, [being statistics-driven is] an easy thing for people in the media to use against you,” Schwartz said. “ ‘Oh, yeah, he can’t adjust; he’s just a stats guy. They don’t really understand the game.’ That’s why sometimes, the whole stats thing is a dirty word.

“If you ask me, Would you rather have a great fly-by-the-seat-of-your-pants guy on Sunday, a guy who can dial up plays and he’d be the best in league, or a guy who is best in the league from Monday to Saturday preparing, I respect the guy who prepares. You’re not always going to be rolling 7, 7, 7 and be hot every week. But if you prepare well during the week, you’ll be consistent from week to week.”


This exact sentiment formed the gravamen of Walsh's west coast offense. Quotes from Walsh:

I have been afforded the experience that allowed us to conceive an offense, a defense, and a system of football that is basically a matter of rehearsing what we do prior to the game. . . .

What we have finally done is rehearse the opening part of the game, almost the entire first half, by planning the game before it even starts. . . .

Now why would you do such a thing? I know this, your ability to think concisely, your ability to make good judgments is much easier on Thursday night than during the heat of the game. So we prefer to make our decisions related to the game almost clinically, before the game is ever played. We've scouted our opponent, we have looked at films, we know our opponent well. . . . To be honest, [in the heat of the game] you are in a state of stress, sometimes you are in a state of desperation and you are asked to make very calculated decisions. It is rarely done in warfare and certainly not in football; so your decisions made during the week are the ones that make sense. In the final analysis, after a lot of time and thought and a lot of planning, and some practice, I will isolate myself prior to the game and put together the first 25 plays for the game. They are related to certain things.

...But whatever you have, if you have planned it and fail, you can't blame yourself for losing your poise. You can't blame yourself for panicking if you have planned these things and they fail. You may really search yourself for the kinds of decisions you made on Thursday night, but you certainly can't make the decision during the game. As a coach, one of the things you are always fighting during the game is the stress factor, breaking your will. The stress factor will affect your thinking. I have been in situations where I could not even begin to think what to do. From that point on, I knew that I had better rehearse everything.


And, too, you can add in analysis that there is no time to do during the game. You analyze the probabilities, you remove the irrational choices like going for certain field goals on certain fourth and shorts.

A few years ago I wrote about the idea that gameplanning and weekday work is advantageous both because you can be meticulous but also because you gain important self-restraint capabilities. I drew on a lecture given by Nobel Laureate in Economics, Thomas Schelling (yes I know, it is not exactly the same as a Nobel Prize as established by Alfred Nobel). To illustrate, Schelling used a story about Captain Ahab; you can read it here.

But a similar story with the self-constraints that gameplanning puts on the coach (as compared with the seat-of-the-pants approach favored by so many) is the story of Odysseus (or Ulysesses) and the Sirens: when Odysseus's boat approached the sirens -- whose sweet singing had lured many sailors to their deaths -- he first put wax in his sailors' ears to block out the music, then had his crew tie him to the mast, thus making him powerless. In the moment, when his boat went by the sirens, he was irrational, and wanted nothing more than to steer the boat to him. But his rational self had already judged this, decided against it, and denied his later, weaker self the same choice.

Although not nearly so dramatic, gameplanning and the script often works the same way (though with slightly more flexibility than one has tied to the mast). Hopefully, for Lions fans, it is this methodical, analytical approach that Jim Schwartz offers.

Moreover, this story highlights the interdependent role that head and assistant coaches must have: they must take turns as Odyssesus and the crew, tying each other up, making the plans in advance, and even, sometimes, in the heat of the moment, entirely ignored, as Odysseus was.

As noted above, gameplans should nevertheless be contingency based; they must be flexible enough to respond to what an opponent does. But of course gameplans are based on this: those who reject scripting because they are too wooden just really don't understand what scripting is. But, a well crafted gameplan can still handle these scenarios and be created in a detached setting.

Moreover, the other important factor is that some information simply cannot be processed merely in-game; answers will only be yielded by careful study through the week. You simply don't have time to crunch all the numbers, assemble data on all the fronts and schemes, nor run down the variety of contingent scenarios. But if you do that, and then combine it with what you learn during a game, you have a chance to win. And, through your study, you might see options -- like maybe what Mike Leach does, the wildcat, or some other forward thinker -- that traditional football intertia blinds you to. As I wrote a few weeks back:

[F]rom Peter Bernstein's book, Against the Gods: The Remarkable Story of Risk.

The story that I have to tell is marked all the way through by a persistent tension between those who assert that the best decisions are based on quantification and numbers, determined by the patterns of the past, and those who base their decisions on more subjective degrees of belief about the uncertain future. This is a controversy that has never been resolved.


And it will remain controversial . . . because the future will be paved by numbers and judgment, marching, somewhat awkwardly, hand in hand.


Good luck to Jim Schwartz.

Hub Fans Bid Rabbit Adieu

Sad day: writer John Updike has passed away. Not known for his sports writing, Updike wrote one of my favorite sports essays of all-time: Hub Fans Bid Kid Adieu. It's a baseball essay about Ted Williams's last game at Fenway Park, so it is technically off topic for this site, but it is both elegant and brilliant. Anyone who has ever been to a baseball game -- or ever lived in Boston -- can't help but feel the essay's immediacy and even importance. Some brief snippets below, but read it for yourself.

Fenway Park, in Boston, is a lyric little bandbox of a ballpark. Everything is painted green and seems in curiously sharp focus, like the inside of an old-fashioned peeping-type Easter egg. It was built in 1912 and rebuilt in 1934, and offers, as do most Boston artifacts, a compromise between Man's Euclidean determinations and Nature's beguiling irregularities. Its right field is one of the deepest in the American League, while its left field is the shortest; the high left-field wall, three hundred and fifteen feet from home plate along the foul line, virtually thrusts its surface at right-handed hitters. . . .
. . .
The afternoon grew so glowering that in the sixth inning the arc lights were turned on—always a wan sight in the daytime, like the burning headlights of a funeral procession. Aided by the gloom, Fisher was slicing through the Sox rookies, and Williams did not come to bat in the seventh. He was second up in the eighth. This was almost certainly his last time to come to the plate in Fenway Park, and instead of merely cheering, as we had at his three previous appearances, we stood, all of us—stood and applauded. Have you ever heard applause in a ballpark? Just applause — no calling, no whistling, just an ocean of handclaps, minute after minute, burst after burst, crowding and running together in continuous succession like the pushes of surf at the edge of the sand. It was a sombre and considered tumult. There was not a boo in it. It seemed to renew itself out of a shifting set of memories as the kid, the Marine, the veteran of feuds and failures and injuries, the friend of children, and the enduring old pro evolved down the bright tunnel of twenty-one summers toward this moment. At last, the umpire signalled for Fisher to pitch; with the other players, he had been frozen in position. Only Williams had moved during the ovation, switching his bat impatiently, ignoring everything except his cherished task. Fisher wound up, and the applause sank into a hush.

Understand that we were a crowd of rational people. We knew that a home run cannot be produced at will; the right pitch must be perfectly met and luck must ride with the ball. Three innings before, we had seen a brave effort fail. The air was soggy; the season was exhausted. Nevertheless, there will always lurk, around a corner in a pocket of our knowledge of the odds, an indefensible hope, and this was one of the times, which you now and then find in sports, when a density of expectation hangs in the air and plucks an event out of the future.

Fisher, after his unsettling wait, was wide with the first pitch. He put the second one over, and Williams swung mightily and missed. The crowd grunted, seeing that classic swing, so long and smooth and quick, exposed, naked in its failure. Fisher threw the third time, Williams swung again, and there it was. The ball climbed on a diagonal line into the vast volume of air over center field. From my angle, behind third base, the ball seemed less an object in flight than the tip of a towering, motionless construct, like the Eiffel Tower or the Tappan Zee Bridge. It was in the books while it was still in the sky. Brandt ran back to the deepest corner of the outfield grass; the ball descended beyond his reach and struck in the crotch where the bullpen met the wall, bounced chunkily, and, as far as I could see, vanished.

Miami looks to hire Mark Whipple as Offensive Coordinator

Mark Whipple is likely the new offensive coordinator for the University of Miami. That name is a blast from the past. Back in 1998 I heard of Whipple when he took UMass to a 1-AA National Title. Back then, he was a bit of a darling in football circles. He was throwing the ball around forty-times a game, and his teams were pretty wide open. So, a spread guy, you say? Not necessarily: keep in mind too that was 1998; the spread had not yet permeated all levels of college football. Although wide-open and experimental, Whipple's offense was kind of a West Coast/One-back hybrid.

Eventually, however, his success at UMass waned, and he was picked up by the Steelers to be their quarterbacks coach. And from there, Whipple's career arc has been rather perplexing: He was quarterbacks coach with the Steelers, where apparently Bill Cowher "loved" him. But Mike Tomlin was not so impressed, as he fired Whipple shortly after he arrived, citing Ben Roethlisberger's "regression."

Then he was picked up by the Philadelphia Eagles, where he coached this past season, under the vague and quite posssibly meaningless title, "offensive assistant coach." The word was that Andy Reid hired Whipple as insurance if his offensive coordinator left before this past season, but when that didn't happen, Reid kept him around anyway. Doing what though, I'm not sure. (Keep in mind that Reid is actively involved in the offense, not to mention the other assistants.) So likely Whipple was restless, and was looking for this change.

And it might be a great one for both Miami and Whipple. But the burning question for Miami fans is not whether Whipple has credentials, or whether he's a bright offensive mind. It's why has he essentially languished for several years? Is there some character or coaching issue that lurks below the surface? Who knows. In any event, on its face, the hire appears to be a good one for Shannon and the U. Notwithstanding Patrick Nix's warnings that Shannon won't allow offensive innovation, Whipple should bring a sophisticated, pro-style, wide-open but open-minded approach (and those things don't always go together, so that's a good thing). But like all else, time will tell.

Wednesday, January 21, 2009

Fire zone-blitzes

Dr. Saturday recently broke down some of the zone-blitzes Utah used against Alabama in their bowl game. It's great analysis; check it out. For a bit more on pass protection and zone-blitzes generally, check out my article here. Also, Bob Davie wrote a worthy article for ESPN about zone-blitzes a few years back. Read all that and come back here: I want to highlight a few quick aspects of the most common zone-blitzes or fire zones. I'll have to leave a more in depth discussion for another time.

Below is a diagram of one of Nick Saban's most common zone-blitzes:



This is very typical. One thing Matt did a great job of breaking down with the Utah article was in his discussion of how the Utes crossed up 'Bama's protections. But it's worth noting the constraints the zone-blitz puts on the defense: namely, the number of coverages you can viably use. Occasionally, in the NFL, some teams zone blitz with a mere Cover 2 or two-deep zone behind it. The Ravens get away with this sometimes with Ed Reed, but it is dangerous -- with all your defensive movement you don't get good jams on the receivers and you too often will let receivers run free in the deep voids.

So by far the most common coverage behind a zone-blitz is a three-deep three-under coverage. Obviously, that can leave wide open spots underneath, but that's still part of the zone-blitz philosophy. As some defensive coordinators say, with so much three-deep, it is actually a conservative approach. And it is one reason why zone-blitzes are so common on third down -- defenses get good opportunities to cross up the pass protection while forcing completions to be made underneath where guys can make a quick tackle short of the first down. This is a favorite strategy of Jim Johnson of the Philadelphia Eagles: there's nothing wrong with giving up a five yard completion on third and nine. (Further, most progressions have the quarterback read long-to-short. By taking away long and forcing short the defense gives itself another few moments to get to the quarterback before he can release the ball.)

Below is a video clip (courtesy of HueyTube) of LSU running basically the zone-blitz diagrammed above.



On the first play in the clip, notice the way the three linebackers appear to attack just before the snap, but then the left outside linebacker (lined up to the short side of the field) drops into coverage, as does the the left defensive end. The strong safety comes up to play the flat and seam areas, and the two corners and the safety drop back into a simple three-deep. And Tennessee actually gets a completion, largely because the linebacker who probably ought to have been in the middle of the field follows the tight end on a short drag route and thus exposes the middle to an in-breaking receiver. (The linebacker should have passed him off to the defensive end who had dropped out; they end up defending the same area leaving the middle wide open).

On the second play LSU is in the same zone-blitz, with three deep. This time Tennessee keeps it simple and throws an out route against the soft outside coverage. Throwing away from defenders is often a solid strategy -- not that it is always easy to identify when there's a zone-blitz on.

Next, briefly, take a look at the clip below of the New York Giants running some fire zones. (Note: the best way to watch this is to watch and rewind several times, watching different players. You can barely see the ball anyway so watch the safeties, linebackers, line, etc. each successive time you replay it.)



In the first clip against the Cowboys, you'll see the exact same coverage -- three-deep, three-short -- except they have changed who does what to further disguise things (hey, it's the NFL now). Now you have a cornerback blitz from the weakside of the formation (and short side of the field), and the three-deep look, instead of being comprised of the two corners and safety, is now in the form of what is called a "cloud" rotation: the corner and two safeties rotate over to create the three-deep coverage. The overload blitz now comes from the short side of the field rather than the wide side as with LSU, and the other linebackers rotate over to fill up the underneath zones. It's really just the same thing, just a different look. And, again, in the NFL your guys can do more. Not only can your linemen drop into coverage, but your secondary can blitz and get to the quartberback while apparently lined up over a receiver.

And, another advantage with the NFL over college comes with the techniques involved. Around the 1:30 mark of that clip is another, rather traditional fire-zone. Atlanta throws about a six or seven yard hitch for a completion. But note the look from the secondary: the cornerback is lined up over the receiver as if he is in press man, but instead he bails -- after the snap -- into the deep-third, thus giving up the underneath completion. Few college or high schools will put cornerbacks responsible for deep thirds of the field up on the line of scrimmage so close to a receiver. Again, it's the NFL. Here, there is little fault you can blame on the offense: kudos to the quarterback and receiver for identifying the coverage despite the disguise.

Finally, motion is still a good way to reveal whether a defense is in man or zone, particularly when the motion changes the strength of the formation. (Though teams are certainly much better at disgusing their coverages even with motion or shifting.) If the defender runs across the field with the receiver, it is probably man. If they sort of "bump over," it's probably zone. And if it is man, well, then maybe you're getting better matchups.

More to come on blitzing, pass protection, and zone-blitzes throughout the offseason.

Smart Notes - Jan 21, 2009

1. I have updated my articles about the "shakes" or "three-vertical," "all-curl," and "levels" pass plays with game film video.

All-Curl


Levels


2. Does defense win championships? ESPN chimes in, but Phil Birnbaum feels that their analysis was rather weak.

3. Pro-Football Reference Blog expands on an idea I brought out a few years ago: namely, that yards per pass is the best measure of the effectiveness of a team's ability to throw. PFR Blog expands on this by analyzing Super Bowl teams and using measures to add yards for touchdowns and subtract for interceptions.

Monday, January 19, 2009

The rise and fall of the spread via Purdue's Curtis Painter

Dr. Saturday recently wrote about his guiltiest pleasures of this past college football season. One of them was Purdue Quarterback Curtis Painter's rather miserable season, despite all the preseason hype from so-called experts like Mel Kiper.

Curtis Painter's Implosion. I do feel for Joe Tiller, a perfectly decent coach who brought the spread to the Big Ten when it was still considered a novelty, went to a Rose Bowl and whose tenure in West Lafayette should be remembered as an unambiguous success. But one of the banes of my existence in preseason was the unfathomable hype for Painter, led by Mel Kiper, who anointed Painter the top senior quarterback prospect in the country despite his wretched mark (0-14 from 2005-07) as a starter against BCS conference teams that finished with a winning record. Painter subsequently tossed one touchdown to six interceptions during the Boilermakers' 0-4 Big Ten start -- during which Purdue scored 6, 3 and 6 points, respectively, against Penn State, Ohio State and Minnesota -- and was benched just in time to watch a redshirt freshman who began the season at running back light up Michigan for a season-high 48 points in November. In short, I was right and Mel Kiper was wrong ... so wrong, in fact, he'd dropped Painter to No. 2 on his list of the top senior quarterbacks by December. Way to eat crow.


For starters, I totally agree about Kiper. Aside from the fact that almost no one knows how to properly evaluate quarterbacks (scroll down for a discussion of Malcolm Gladwell's "quarterback problem"), it is well known in football circles that Kiper is just a fan -- he possess no uncanny scouting skills. (But who really does?) There's nothing wrong with that, but all he brings to his "rankings" and assessments is exactly the same thing you or I would after watching a lot of games on TV and checking the stats. That's it. No more, and no less. Kudos to him for doing what he does, but that's all it is.

But I think Curtis Painter's woes (great against weak teams, mediocre to poor against good ones) can be partially explained as a data point in a larger story. This story gets back to my discussion of the rise of the terrible spread team, and even my earlier post about whether the spread has reached its apex as far as helping the little guy beat the big goliaths with lesser talent.

Painter never started under the original Purdue offensive scheme architects, namely Joe Tiller plus Jim Chaney. Chaney left to go to the NFL (now he's with the University of Tennessee), and in stepped Ed Zaunbrecher, who had followed great success coaching offense at Marshall with some success doing the same at Florida under the Zooker and less at Illinois. When Zaunbrecher got there Tiller had already decided to move in some new directions with the offense. But, while there were differences the problems wound up being many of the same things, because of the talent and overarching philosophy.

In many ways, under Zaunbrecher I liked a lot what Purdue was doing. If I had to compare their offense to anyone else's in terms of structure and schematics it likely would have been the New England Patriots under Belichick/McDaniel -- one-back sets with a tight-end, shotgun, and lots of base, simple 5-step concepts like the snag, all-curl, three-verticals, four-verticals, underneath option routes, and smash. This was slightly different than the original Tiller model with Chaney when Drew Brees and Kyle Orton had been there, which was more no-back and more three-step drops.

Brees-Tiller-Chaney clips:



Painter-Tiller-Zaunbrecher



(Compare the quick drops and completely spread sets that Brees tended to operate from with the longer developing plays used with Painter. Some of this is styles--Painter probably had a stronger arm than Brees, and Brees was a quick decisionmaker with a quick release. And some of the evolution was necessary. But it's worth pointing out the slightly different styles.)

In any event, the Tiller-Chaney-Brees model of four and five wides and three-step drops began to turn somewhat stagnant against the big boys; it's not a phenomenon entirely unique to Painter. Kyle Orton began the 2004 season as a Heisman contender and then Purdue rattled off loss after loss and failed to generate enough offense. And the reasons were simple: by then, if you spread out Wisconsin, Michigan, or Ohio State, they had guys who matched up with all your receivers, and if you had any advantage at all they could still put a floater or robber defender to bracket him and take him away.

But, despite the changes with Zaunbrecher, the exact same pattern emerged, except almost even more brutally. From about 2002-2004, with the Chaney short-passing model, Purdue would manage a number of completions, all of them for very, very short yardage, no run after the catch, and would hope to break a play or two. You saw a lot of that with Zaunbrecher, but mixed in were a lot of very difficult to complete downfield passes to guys who were not open. The week before, against Syracuse or even Minnesota, they'd look like the Patriots. Against Penn State or Ohio State, they looked like Syracuse. Though Zaunbrecher was more willing to stretch the field, against these top teams they could not shake anyone free. Plus, this exposed the quarterback to pressure and the line to certain protection issues, something that had not been as much of an issue with the previous quick-release approach.

In my 2006 article, I wrote this about where the spread was headed:

The offense has arguably become the opposite of an equalizer, it has become an amplifier: if you are talented you can really rack up the points because no one can cover Vince Young, Ted Ginn or the like one-on-one, but if you're not, you just get sacked and no one gets open.


So -- and I recognize that there were other issues at work like play-calling and Painter's at-times erratic decision-making -- but to me Purdue and Curtis Painter became an object lesson for the effect of the spread. When they played out of conference opponents or Big 10 bottom dwellers, they lit them up: their offense worked perfectly to create matchups and generate plays that garnered chunks of yardage at a time. But against the big boys, they got manned up, pressed, jammed, and blitzed into oblivion.

And maybe even Purdue's new head coach, Danny Hope, who coached with Tiller back in the Brees days and this past season, has noticed this. He chose not to retain Zaunbrecher and has instead hired Gary Nord from Florida Atlantic, who spent the last two decades as Howard Schnellenberger's offensive coordinator. If anything, his offense is NFL-esque, but almost a throwback to the early 90s of the Cowboys under Aikman and 49ers with Steve Young. Maybe it will be a success, or maybe it won't. But the days of Purdue being spread-only are likely over.

Tiller? Definitely a success at Purdue. Maybe his biggest fault was that his idea was so good it got copied and assimilated too quickly.

Tempora mutantur nos et mutamur in illis.

Thursday, January 15, 2009

Deep Crossing Route - Larry Fitzgerald style

Crossing routes are a part of every passing team's offense. But we mostly hear on TV about the shallow cross, where a receiver drags across the field at about six yards or less. Good route. But when teams want a big play, they increasingly turn to the deep crossing route, a route that seems -- at least based on the media coverage -- far less well understood than its little brother.

There was a fantastic example of the deep cross this past weekend in the NFL playoff game between the Arizona Cardinals and the Carolina Panthers: Kurt Warner hit Larry Fitzgerald dragging across the field for a touchdown late in the first half.

So this is a play worth examining. Moreover, it is an Airraid staple, and Bill Walsh himself put this route and related concepts to great use.

In this article I will to first discuss the route itself. Then, using a little video from the Cardinals game (hence the Larry Fitzgerald reference) I want to show how the Cardinals and many other teams combine the deep crossing route within an overall route structure or pass concept. And finally I will briefly discuss how the route is used in the West Coast or Airraid offenses, within a similar but slightly different route structure. I have video of Texas Tech to further demonstrate.

The Route

The route is simple to describe but takes a lot of practice to perfect. The receiver must have a medium to tight split from the tackle (i.e. in closer to the line rather than split out too wide) and he begins his route inside as if he is running maybe an inside curl. He of course climbs to a depth of around 10-12 yards before breaking across the field, to a point roughly 17-22 yards from the sideline. Against man, the receiver will keep running. Against zone, he will find wherever the open grass is. Sometimes that involves just continuing to run to the open spot near the sidelines; other times the open grass is found by settling in a window between defenders.




Here's where technique comes into play. While, as with any crossing route, the receiver has a lot of freedom, there are some specific guideposts for anyone running it. The biggest is he wants to go under Will (or Sam) and over Mike. What?

To explain: the receiver wants to slip underneath or inside the outside linebacker and over the middle linebacker. (Mike = Middle linebacker; Sam = strongside; Will = weakside.) There are a few reasons for this. One is that, if done correctly, the receiver should have a clean release while still distorting the zones as he passes through them. Second, there are specific distortions he's looking for.

The outside linebacker is typically responsible for some kind of outside or hook (intermediate) zone; going inside him is the best way to get away from that guy, but if the linebacker tries to cover the crosser in any way, he will have to chase him rather than wall him off.

Going over the middle linebacker is consistent with general route-running strategy, which is that the easiest way to lose a guy is to go so he can't see you. If the receiver cuts in front of him the defender can see exactly where he's going and break on the ball and the player. By going behind him, the receiver not only stretches the linebacker deep, but the receiver also is more likely to lose him. You'll see in the video below that Fitzgerald goes over the middle linebacker, number 52 for the Panthers. (Of course, especially in the NFL, it's not easy to see what coverage they are in. NFL announcers misidentify coverages all the time; good coaches in the booth say they can't always even identify what they are doing. This is one reason why routes like this are good: they are versatile, by flooding zones and breaking away from man, and generally giving the receiver freedom. I think on the play below the Panthers were in some kind of man coverage with some defenders, including the middle linebacker, acting as floaters.)

As a final note, if the middle linebacker refuses to let the receiver go over the top, there are two responses: one, the crosser can break underneath because the Mike is either out of position or is actually playing the deep middle ("Tampa-Two," which is not really two-deep at all), and, second, the coach should call an inside curl by that player the next time to take advantage of this super-deep drop by the middle linebacker.

Deep Cross with a Double Post


Below is how the Cardinals ran the play, with the split-end (here, Larry Fitzgerald), running the deep crossing route underneath two receivers on the playside running post routes.



In the clip below, both the runningback and H-back stay in to block, but my guess is that their responsibility is to "check-release," i.e. look for blitzers and release if no one comes. See here for more on check releasing and pass protection in general. Below is a clip of the Cardinals running the route from a few angles. And do forgive my nascent internet video editing skills. There's a slight pause in the middle but the clip keeps going.



And below is a diagram of how the Colts use the play off of a play-action look for their stretch run play. (They like this as play-action off the stretch run to the weakside, with the tight-end staying in to the backside.)



As you can see from the diagrams, the first thing that the overall route structure does is pull the secondary deep to open it up for the deep cross. Nothing too exciting there. Also, the underneath tight-end, H-back, and runningbacks pass protecting and leaking into the route (along with play-action) helps suck up the intermediate defenders to create that window for the cross. The play is a type of vertical stretch.

But more specifically, let's look at the double-post. Actually, the outside receiver (here, "Z") will run a "skinny post" or "glance" route -- he will burst off the line and at around 10 yards will stick his outside foot and angle just slightly inside the corner, and the ball will not be chucked down the field as a bomb but instead quickly into the seam at around 15-18 yards as a kind of deep slant. But, if he gets press man coverage with two deep safeties, he will typically convert his route to a "go" to get a stretch with the nearby safety between him and the post runner.

The inside receiver running the post has a relatively simple route. If the middle of the field is open (i.e. Cover two-deep) he just runs right down the middle of the field. If there is a deep middle safety (Cover 3, Cover 1) he will try to cross that defender's face as he heads for the opposite hashmark. This puts the safety in a bind. If he comes up, however, that receiver can still just head for the end zone.

Below is how that concept looks against Cover 2:



Below is the double-post concept against Cover 3:





Airraid Deep Cross

Below is how the Norm Chow and the old BYU guys and the Airraid types run this route.



Most of the routes here are self-explanatory, but a note on the route by the halfback (left back). It is an option route: he pushes to five yards, tries to "step on the toes" of the defender over him. If it is man and the defender is trying to guard him, he will slant in against outside leverage by the defender, or break out against inside leverage (usually out). If it is a zone and defenders try to bracket him, he will find the hole and just sit in it. Now, onto Norm Chow's reads in brief form:

Quarterback: Five-step drop; hitch-up in the pocket only if you need to.

Read: Eye the half-back (left runningback). Read half-back #1 in your progression, then the Y (deep crosser) #2.

QB and receiver must make eye contact vs. man. Against zone, receiver finds open grass. Only time quarterback will throw to the deep-crosser is if the Will linebacker and Mike linebacker (middle and outside linebackers) squeeze the half-back -- i.e. take the HB's option route away.

The Airraid guys read this basically the same, except they use a more pure progression: (1) Split-end on the "Go," (2) Y on the deep cross, (3) runningback on the option (sometimes just runs a basic flat route), (4) flanker on the square-in/dig route, and (5) other runningback in the flat as an outlet.

Otherwise the Cardinals and Airraid versions are the same: it is still a vertical stretch play; you still try to get the defense to defend deep and have the cross come underneath. The big difference is that, with the Cardinals version, you get a ton of frontside pressure with the two posts and the crosser. But if the weakside defenders -- safety, linebackers -- cheat and collapse down, you really need a different play.

With the Airraid version you have less frontside pressure, but on the backside you get a great combination between the deep-crosser and the backside deep in route. If the linebackers or safeties collapse on him as he heads up the field and breaks across, the backside flanker often comes quite open in the hole in the zone. Below is some video of Texas Tech running this route. Sorry if it is difficult to see, but that's often the nature of game cut-ups.(Hat tip: Hueytube)



Conclusion

It's a good play.

Monday, January 12, 2009

Airraid offense information, reading material, and passing concepts

This post is intended as a resource dump for links and items related to the "Airraid Offense," the pass-first offense devised by Hal Mumme and Mike Leach as a derivative of the old Lavell Edwards/Norm Chow/BYU offense. Other notable Airraid acolytes include coaches Chris Hatcher and Tony Franklin. And other notable coaches, including Sonny Dykes, Art Briles, and Mark Mangino, have all coached with Leach or Mumme and incorporate their concepts to varying degrees. At core, this article gives me something to link to every time I use the term "Airraid."



Below are the major Airraid/BYU concepts combined with Norm Chow's reads for each. Note that this more closely hews to the original BYU version than the Airraid version, which has slight differences. If you can't figure out the differences after reading all of the above, then heaven help you. (Thanks to Bruce Eien for some of the diagrams.)

61 Y OPTION




5 step drop. Eye Y and throw it to him unless taken away from the outside by S/S (then hit Z), OR inside by ILB (then hit FB). Don’t throw option route vs. man until receiver makes eye contact with you. Vs. zone – can put it in seam. Vs. zone – no hitch step. Vs. man – MAY need hitch step.

62 MESH



5 step drop. Take a peek at F/S – if he’s up hit Z on post. Otherwise watch X-Y mesh occur – somebody will pop open – let him have ball. Vs. zone – throw to Fullback.

63 DIG



5 step drop and hitch (7 steps permissible). Read F/S: X = #1; Z = #2; Y OR HB = #3.

64 OUT



5 step drop. Key best located Safety on 1st step. Vs. 3 deep look at F/S – if he goes weak – go strong (Z = #1 to FB = #2 off S/S); if he goes straight back or strong – go weak (X = #1 to HB = #2 off Will LB). Vs. 5 under man – Y is your only choice. Vs. 5 under zone – X & Z will fade.

65 FLOOD ("Y-Sail")



5 step drop and hitch. Read the S/S. Peek at Z #1; Y = #2; FB = #3. As you eyeball #2 & see color (F/S flash to Y) go to post to X. Vs. 2 deep zone go to Z = #1 to Y = #2 off S/S.

66 ALL CURL



5 step drop and hitch. On your first step read Mike LB (MLB or first LB inside Will in 3-4). If Mike goes straight back or strong – go weak (X = #1; HB = #2). If Mike goes weak – go strong (Y = #1; Z = #2; FB = #3). This is an inside-out progression. NOT GOOD vs. 2 deep 5 under. (See my article on this route here.)

67 CORNER/POST/CORNER ("Shakes")



5 step drop and hitch. Read receiver (WR) rather than defender (Corner). Vs. 2 deep go from Y = #1 to Z = #2. Vs. 3 deep read same as “64” pass (Will LB) for X = #1 or HB = #2. Equally good vs Cover 2 regardless if man OR zone under. (See my article on this route here.)

68 SMASH



5 step drop and hitch. Vs. 2 deep look HB = #1; FB = #2 (shoot); Z = #3. Vs. 3 deep – stretch long to short to either side. Vs. man – go to WR’s on “returns”.

69 Y-CROSS/H-Option



5 step drop - hitch up only if you need to. Eye HB: HB = #1; Y = #2. QB & receiver MUST make eye contact vs. man. Vs. zone – receiver finds seam (takes it a little wider vs. 5 under). Only time you go to Y is if Will LB and Mike LB squeeze HB. If Will comes & F/S moves over on HB – HB is “HOT” and will turn flat quick and run away from F/S. Otherwise HB runs at his man to reinforce his position before making his break.

Video clips

Below is an assortment of video clips of the offense. Nothing too technical. You should be able to recognize the concepts nonetheless.







Saturday, January 10, 2009

The case for Tim Tebow going pro and foregoing his senior season

This is rank speculation built on a foundation of conjecture, but, the more I think about it, the more I think that there is a decent case to be made for Tebow to go pro now rather than come back for his senior year.



First, I have no idea what kind of pro player Tebow will make. He's obviously one of the best and most accomplished college players in recent history. But what would he gain exactly by coming back, comparing to what he might gain (or lose) by leaving early?

By coming back he could win the Heisman (he has one), a National Championship (he has two), or he could just "enjoy the college experience." Though I imagine if Tebow says that he'll have something different in mind than Matt Leinart did when he offered that reason. (Leinart took a single music class his senior year, but presumably kept an otherwise busy schedule.) Tebow also might be able to improve his draft stock by improving his footwork and decision-making skills.

But why not just work on those things in the pros? It's unlikely that Tebow would be able to suddenly vault into being a first or even second rounder next year, particularly since Florida is not exactly going to change the gameplan. And if he wants to work on things like his footwork, reads, and learning the contours of an NFL offense, why not do it in the NFL? Many of the current starting quarterbacks all had significant apprentice time on a roster, and I think we all expect Tebow to work rather diligently at improving once he is there. And he does have some physical tools. I mean, he could hire some quarterback gurus and tutors for the offseason (he would be doing that anyway), but once in the NFL he would be able to work for a couple of years on nothing but the things that might make him a successful NFL QB.

Another factor is his mediocre draft grade. Although at first blush this seems like a disadvantage, my guess is that that whoever actually does draft him will do so because they see potential in him. They probably too will know a thing or two about developing quarterbacks. (Maybe Urban's good buddy Bill Belichick will take a flyer on the kid? Worked for Matt Cassel.) So he might land in a great situation where he'd be able to work on his game and develop.

The biggest possible disadvantage I'd see for Tebow is that, if he is only projected to be a fourth through sixth rounder this year, by leaving he might lose an opportunity to leap into the second or third round next year where he'd make significantly more money. But something tells me that is not the sort of thing driving him. And in any event, that's a speculative reason; his draft grade could also remain the same.

Finally, one factor plays a part with Tebow that normally wouldn't with other quarterbacks, but often does with runningbacks: hits and shelf life. NFL runningbacks have a limited shelf-life, before their skills and ability quickly diminish. The common wisdom is that this is a factor of the hits and punishment their bodies take. Tebow, unlike most future pro quarterbacks, takes many hits like a runningback; indeed, he often delivers them. It's a crude measure, but compare the total carries by Tebow to Texas Tech's quarterback, Graham Harrell, over the last two seasons. (In college "carries" includes sacks.) In 2007 and 2008, Harrell had 79 total "carries." Tebow? 386. Adding another 200 carries, sacks, and other hits might just start to wear on his body. And we all know NFL guys will be trying to hit him hard and often when he gets there, so body preservation might be reason alone.

So there it is. I'd bet that he comes back, but it's worth considering some reasons why he might not. The biggest is to get in the right situation with the right coaches. I mean, this afternoon we have Jake Delhomme playing against Kurt Warner, two guys who made their careers by beginning as backups and developing once they were in the league (or in the Arena league or NFL Europe). Leinart? He's on the bench.

Update: Well, Tebow is coming back.

Smart Notes - January 10, 2009

1. Air Yards. I want to highlight one of my favorite new stats: Air yards, as developed by Advanced NFL Stats:


If football were a brand new invention, and we had to decide how to credit the various amounts of yards gained to various players, how would we do it? If I said, "There's this kind of play called a pass, in which a thrower passes the ball to a another player who then runs with it as far as he can. I say we credit all the yards run by the receiver to the thrower." You'd say I was nuts.

I'd say, "Well, it takes a special kind of talent for a passer to get a lot of yardage after the catch (YAC). I won't be able to prove it, in fact, I won't have any evidence for that statement at all, but I still think our primary measure of a passer should include all those yards." I'd be laughed at.

Here are the QBs from 2007 who led the league in percent of their passing yardage as YAC: Croyle, Testaverde, Greise, Harrington, Favre, McCown, Losman, and Lemon. The 2006 list includes Brunell, Carr, Favre, (Rob) Johnson, and (Alex) Smith. There's isn't a single guy on that list who we can call a legitimate starter.

The 2008 season's list of leaders in %YAC include Cassel, O'Sullivan, Campbell, Favre (again), Losman, and Wallace. But Matt Cassel is good, right? Maybe not. Keep in mind how good the team around him was. He was handed the keys to a Ferrari. If a QB racks up his passing yards with YAC, he's either throwing lots of short check-downs and screens, or he has spectacular receivers--or both. Neither is necessarily an indication of a particularly skilled passer.

If we throw away all the YAC and look underneath, what do we have left? I call it Air Yards (AY). It's the distance forward of the line of scrimmage a pass travels. Although it's not a perfect measure of a passer, I think it makes a lot more sense than crediting Donovan McNabb with 71 yards and a touchdown for a 1-yard screen pass to Brian Westbrook.


I agree with this. It's just one measure, and of course accuracy does help lead to more YAC and even screen passes take some athleticism and dexterity to pull off, but the analysis here indicates that we should not give QBs too much credit.

The other reason I like it is because it heightens accountability for yards after catch yards for receivers. If receiver X gets a lot of YAC and receiver Y doesn't, it might be a function of the offense they play in or the types of routes they run, but you nevertheless get a way to compare receivers and put the burden on them: either he's getting yards after the catch or he isn't, and the team should play a guy who gets them. Now, part of Brian's analysis here seems to indicate that it is QBs who drive this as much as anything else -- if a QB can't complete it downfield he'll inflate YAC all around with dump-offs. But, as I said, it's a fascinating statistic as a way to unpack yards per attempt, which is still the best overall metric of a passing offense.


2. Assorted Links


3. Favicon. Smart Football now has a "favicon." We're doing it real big over here.