Smart Football has moved!

Please check out the new site, smartfootball.com. All future updates will be made there.

Tuesday, March 31, 2009

Speculations on play-calling on first, second, and third and ten

From Advanced NFL Stats:

All the numbers that follow are from all 10-yards-to-go scrimmage plays in the first 3 quarters of regular season games from 2000-2007. The only other limitation was that the game score was within 10 points. I wanted to exclude situations when teams exercised an abundance of either risk or caution.

Note the percentage of play types called on 1st, 2nd, and 3rd downs (with 10 yards to go). There is a fairly even split between run and pass calls on 1st and 2nd downs. On 3rd and 10, the a pass is far more expected.

% of Play Types by Down, 10 Yds To Go

Type - 1st - 2nd - 3rd - Total
Pass - 47.2 - 52.7 - 91.1 - 49.6
Run - 52.8 - 47.3 - 8.9 - 50.4

Although 91.1% isn't 100%, it's close to where the anchor point on the lower right side of the game theory graph--almost the pure pass vs. pass defense strategy combination. Now let's look at the average outcomes for these situations.

Yds Per Attempt by Down, 10 Yds To Go

Type - 1st - 2nd - 3rd - Total
Pass - 7.0 - 6.3 - 6.5 - 6.9
Run - 4.2 - 4.4 - 6.9 - 4.3
Total - 5.5 - 5.4 - 6.5 - 5.6


When passing is most predictable, it yields half a yard less than on first down, when it is less expected. Conversely, running is most successful when it is least expected.

At this point, I should point out that passing on 3rd and 10 yields slightly more yards than on 2nd and 10, which isn't completely what we'd expect. This is almost certainly because defenses will allow short complete passes on 3rd down in exchange for being relatively assured to be able to stop the gain short of 10 yards. This is part of the problem posed by the fact that yards does not equal utility. We'll have to dig a little deeper. The next table lists interception rate by down.

Interception % by Down, 10 Yds To Go

Int Rate
1st - 2nd - 3rd - Total
2.6 - 2.9 - 3.5 - 2.7

Now we see more what we'd expect--a slight increase from 1st to 2nd down, then a large jump on 3rd down, in accordance with the associated increases in passing predictability. The next table lists adjusted yards per attempt, which is YPA with a -45 yd adjustment for every interception thrown. Adj YPA, however, still exhibits the same problem as plain YPA. It underestimates the drop off from 1st to 3rd down in passing effectiveness because defenses will allow gains, as long as they're not more than 9 yards.

Adj Yds Per Attempt by Down, 10 Yds To Go

Type - 1st - 2nd - 3rd - Total
Pass - 5.9 - 5.0 - 4.9 - 5.6
Run - 4.2 - 4.4 - 6.9 - 4.3


So what we can say is, the reduction in passing effectiveness due to predictability is likely at least 1 full adjusted yard per attempt. The drop from 1st down to 2nd was 0.9 yards, so the true reduction in effectiveness from 1st to 3rd down may be far larger.

Except that there's a problem with this analysis. There's a bias in the data. Which teams are more likely to face a lot of 2nd and 10s and 3rd and 10s? The ones that stink at passing. So the 2nd and 3rd down numbers are lower than would be representative of the league as a whole. In other words, poor passing teams 'get more votes' in the analysis.


All this is intended to tee up a game-theory analysis for finding some kind of ballpark run/pass equilibrium. Do read the whole thing.

But a few brief thoughts:

  • The adjusted final numbers intrigue me, particularly second down as compared to first. (As Brian notes, third down is tougher to break down since it's really a binary question of conversion versus failure.) But I'm struck that on second down the yards per pass attempt drops by nearly a full yard while the yards per run goes up only .2: why does the defense get so much better on second down? Is the data skewed to losers? Is play-calling worse on second down?

  • In that vein, I wonder if the old conventional wisdom about "getting back half on second and ten" works against the offense. On first down the passing plays are likely to involve play-action as well as quick or intermediate passes -- coaches can use their full asrsenal; maybe on second coaches are too concerned with screens and quicks -- trying to just get half -- that they give up too much in the way of expected points?

  • But on the other hand, what if they get this 5.0 yards per pass attempt on second and ten with more certainty and less variance than the 5.9 on first down. If so, then possibly the offense is in better position to convert third down than they would be even with a greater expected play value that carried more variance. Could cut either way; football is complicated.

Hopefully Brian can shed some light as his series develops. I look forward to it.

Monday, March 30, 2009

More on the Run & Shoot "Choice" Concept

Mike Drake of Longmont High School (CO) sent me the below video of the "choice route." Note that I have updated my original article on the choice with these clips. But here are the clips below with brief descriptions. See the full article for more of an explanation.

First, here is how Drake has his kids run the "choice." It is the same concept I described, though he has added a "safety divide" or climb type route by the tight-end to help stretch the safety; it turns it into something like four-verticals from trips, though with the choice.



First, here is the play against Cover 3 with a zone-blitz. The blocking isn't quite there but the QB moves in the pocket and finds the seam-reader (slot receiver). The free-safety here jumps the tight-end on the climb/divide route, which makes the slot wide open. It is actually the backside cornerback who makes the tackle in the end zone, though he is too late.



And the video:



Against Cover 2, where the safeties drop very deep, you can see in the video below where the seam-reader works off the underneath defender and under the safety into the void for a nice completion.



And finally, against Cover 1 man they get a matchup they like and throw the go to the singled up choice receiver (though they don't complete it). The quarterback does a good job getting the ball there before the single-deep safety can get over. On this play they also keep two extra guys in to protect, so it kind of becomes a three-man route.

Sunday, March 29, 2009

Should college athletes be paid?

The New York Times had a recent roundtable on the topic. A few interesting points worth highlighting and responding to. Mostly professors chime in, and the specific topic is March Madness, but the debate highlights the tension between old University ideals and the realities of college athletics.


First, Murray Sperber, Indiana University professor:

One of the great myths about March Madness is that it earns huge sums of money for participating schools. Yes, CBS pays billions of dollars — over many years — to televise the games but only a very small amount of that money trickles down to the Division I schools eligible to play in the tournament.

The money is less a bonanza for colleges and universities than a lottery. To get a ticket, the N.C.A.A. requires every Division I school to have many teams in many sports and many athletes on scholarship — and almost all of these teams are in money-losing sports. Indeed, last year in the Chronicle of Higher Education, the N.C.A.A. conceded that almost every athletic department in the United States, including those at schools participating in March Madness, generates red ink.

While the N.C.A.A. tournament generates billions of dollars, even the schools and conferences that do well in the games lose a lot of money.

But according to standard university accounting methods, an athletic department cannot end the year in deficit. Thus universities frequently mop up the red ink by taking money from other sources, especially their general operating funds. This is money that could go to student loans, faculty grants and other worthy academic enterprises.

In the last decade, the N.C.A.A. has raised the cost of its lottery tickets. As a result, a number of schools like Georgetown University, with outstanding basketball teams but no Division I football squads, were coming close to breaking even in their athletic department finances until the N.C.A.A. stepped in and required the school to field an expensive, money-losing football program.

So while March Madness generates billions of dollars, athletic departments lose a lot of money. Even the schools and conferences that do well in the tournament and receive seven-figure payouts are forced to put every dime of that money toward their athletic department expenses and deficits — and that’s why the academic departments see very little, if any, of that money.

The whole system stinks and cries out for a much-needed reform of the N.C.A.A. and its requirements for participating in Division I. Such reform should be No. 1 on the agenda, far ahead of the allocation of money from March Madness.


This is accurate: the argument of "when have 60,000 people shown up to see a chemistry exam?" is entirely beside the point. Athletic departments are self-contained businesses, really. That said, there are academic benefits to good athletics: almost all schools report a surge in admissions after successful seasons, and more admissions means that selection committees can be more selective and thus bring in better students. In any event, this deserves more in depth treatment, but I personally would be fine with paying players, as there are problems with the current model.

That said, I don't agree with agree with this rather naive view by professor William Dowling of Rutgers:

The various proposals for “reforming” college sports — by paying a stipend to the athletes who provide television profits, say, or diverting some of the money to academic purposes — show just how oblivious we’ve become to the damage commercialized Division IA athletics has done to American universities.

It’s not the money, it’s the silent triumph of consumerist ideology over academic and intellectual values in higher education.

The real issues have nothing to do with the millions generated by the N.C.A.A. tournament or holiday bowl games. They have to do with the silent triumph of consumerist ideology — in particular, a T.V.-revenue-driven behemoth with tremendous power to shape social consciousness — over academic and intellectual values in higher education.

You couldn’t ask for a better example than “March Madness,” the media spectacle that turns American universities into marketing vehicles for advertisers like Coca Cola and General Motors. For four weeks every spring, lower-level professional athletes wearing college jerseys are seen running back and forth on the television screen between the commercials.

Meanwhile, sportswriters and T.V. commentators maintain the solemn pretense that these are college students, young people who came to university to study Wittgenstein and learn medieval history and master the intricacies of R.N.A. replication. The N.C.A.A. grinds out public relations hype about “Academic All Americans” and “academic progress ratings.” And the public, as though mesmerized, never raises an eyebrow.

We shouldn’t be worrying about exploited athletes — few really are. Nor should we be worried about steering T.V. money to academics. Real colleges and universities — New York University, say, or Harvard, or the University of Chicago — have ways of paying academic costs without prostituting themselves to commercialized athletics. The solution is to end the prostitution itself.


Okay fine, I agree that it stretches credulity to contend that Derrick Rose or Michael Beasley (one and done freshman phenoms who went to the NBA after just a year in college) went to their schools to learn academics, especially during frantic March Madness time, where games take place in some far off city on Thursdays and Fridays (class?).

Yet notice Dowling's argument: to him, the athletes aren't exploited (despite being, at age 18, promised the potential of millions in professional sports but inevitably, with few exceptions, failing to get there while in many cases squandering the chance for a degree); nor the fans; but instead the University itself -- that holy place of learning -- that is corrupted by this "consumerist ideology" brought in by what he might describe as "the athletic element."

But this presupposes that Universities are powerless about this, yet most University presidents are academics, not former athletes. And while the "60,000 people don't come to see a chemistry exam" argument is bogus, boosters and sponsors do go to see football games and often donate to the University writ large.

Yet the whole argument just seems weird to me: it's not like anyone -- including announcers - really sits and gushes about Rose (or Michael Crabtree, Brian Orakpo, or even Tim Tebow) how amazing it is that they handle all these football duties on top of studying Wittgenstein. Instead, we all know it is a strange and precarious relationship that athletics has to academics. But strange does not equal corrupt.

Nor are plaintive invocations of the evils of "consumerist ideology" going to tell us anything: assuming this "ideology" is evil and it "triumphs," is it athletics that is the cause or a symptom? Finally, when did this utopia of the perfect and serene academic setting exist before athletics helped undermine it? I'm sure in Dowling's mind that, if not for athletics, Universities would look like Raphael's school of Athens, some kind of modern Lyceum. (And, no doubt, in this utopian academic setting it would be the professors and the brightest students exalte, rather than athletes and highly paid coaches.)

But this sharp dichotomy is false, and this utopia never existed. Is that reality optimal? I don't know. But even Plato, in his training regimen for his philosopher kings, recommended years of intense physical training beginning at the age of 18, and it was only out of those that excelled at athletics who ought to be then chosen to embark on ten years of math training, another five doing dialectics, and fifteen more managing the polis. (This emphasis on physical tools has long rankled the modern academic who sees sports and thinks only of barbarism.)

Yes, Universities are often put in an odd position serving the dual purposes as places of higher learning and something like minor league sports teams. And athletic departments often turn into little fiefdoms to few people's gain. But railing against the entire sports industry and our "consumerist ideology" strikes me as unproductive. I'm satisfied for now to just commend schools like Vanderbilt, who have consolidated their athletic departments into the rest of the school, and to hope other schools follow suit. Otherwise, it strikes me that people like Dowling would only be happy with eliminating college sports (save for a club cricket team or something) and replace it all with true professional minor leagues. I do not think it would have any positive effect on "society" in beating back the supposedly evil "consumerist ideology" that Dowling sees engulfing us. And I especially don't think it would return The University to some gilded age that exists only in Dowling's mind.

Reason #4080 to love Mike Leach

Bye bye Easy Ed?

Edward Britton, Texas Tech’s most experienced split end, was demoted before spring practice and further raised the ire of Tech coach Mike Leach this week. After Friday’s practice in 30-degree weather and a few snow flurries, Britton was sent to study classwork [at midfield on Texas Tech's practice field]. . . .

“Ed didn’t like showing up and studying at places I felt like he needed to and like the academic people asked him to, so he can go study out there on the 50-yard line,” Leach said. “We’ll take baby steps, and if he does good studying out there, we’ll decide if we’re going to actually let him practice.”

Asked if Britton was on thin ice, Leach said, “I’d say that’s accurate. All guys that don’t study are on thin ice, as far as I’m concerned.”

Leach said Britton was to remain out in the cold, studying for at least an hour and a half after his teammates had gone in for the day.

“If somehow he fails to do that, then that’ll be the last we ever hear of Easy Ed,” Leach said.


Full article
(h/t Double-T Nation)

Saturday, March 28, 2009

Assorted links

1. How would Malcolm Gladwell describe the 2008 Texas Longhorns' season?

2. How much is Jay Cutler, the commodity, worth? (I like Cutler more than Cassel, but it's too early to tell.)

3. I want to gradulate corch urban meyers and percy harvey.

4. Travis Henry: "A near perfect example of an infinite discount rate."

5. Is Jim Schwartz thinking lineman or quarterback for the Lions' pick? (Based on what I've written previously about picking quarterbacks and on Jim Schwartz, I'd be surprised if he went with a quarterback, but who knows. More on this topic later.)

6. Want to work for Chip Kelly at Oregon? Fill out the application. (And give yourself awhile to do it. And make sure you have experience coaching offense at the college level. And only need two hours of sleep per night.)

Friday, March 27, 2009

Texas Tech - First-Year QB Comparisons

From Double-T Nation:

For the first time in two seasons, Texas Tech has a brand quarterback and although we're accustomed and grateful for what Graham Harrell did, it's time to look a bit forward and wonder what Taylor Potts might bring to the Red Raiders as a first year starter under Mike Leach's system.

Going into Graham's second year, I asked how much he could improve from one year to the other, but this time I thought that it might be a good idea to take a look at Symons, Cumbie, Hodges and Harrell's first (and sometimes only) year. The nice part about this is that it's a nice mix of players. It's not just one type of quarterback, which means that perhaps there's actually something to gain from looking at what we can expect.

. . .

Playing It Safe

What's the one thing that jumps out at Harrell's 2006 season? For me it's the fact that he had over 50 attempts for every interception. Contrast that with the touchdowns per attempt? Now, contrast that with Symons and what does that tell you? For me, it tells me that Symons was a guy that was going to take chances, while in Harrell's first year, he was dead set on playing it safe, evidenced by the lowest yards per attempt of any of the four, although he only beat out Cumbie in that category by one-one-hundredth of a point. There's got to be some middle ground here, and taking a look at Cumbie's 2004 season, his touchdown to attempt ration is far and away better than his partners in crime. Statistically, he's really not much better than his fellow quarterbacks and lost in all of this, sometimes is that Cumbie was just damned good at putting the ball in the endzone.

My Favorite QB Stat

I've probably beaten everyone over the head about yards per attempt and it's a really bad habit, but if you'll indulge me here, I'll try to make this quick. In the Air Raid offense, there may not be a more telling statistic about the success of a quarterback than yards per attempt. Every offense is better when the team is moving the ball vertically, rather than horizontally. That's probably one of the real misconceptions about Leach's offense, is that the intent may be to make it a dink-and-dump offense, but I think this is more than likely a product of the quarterback rather than the offense itself. Exhibits "A" and "B" are Symons and Hodges. Granted, the Air Raid is not as vertical as many other offenses, but taking last year as an example, Texas Tech ranked 20th in the nation at 8.11 yards per attempt. The offense bogs and becomes not as effective if the pass is going sideline to sideline.


While I completely agree that yards per pass attempt is the most valuable passing statistic, I also think it can be adjusted slightly to better capture the issue Seth is looking at here. Specifically, you can factor in interceptions using a simple rule of thumb. This is relevant here particularly for the raw numbers between Graham Harrell's first season, in 2006, and B.J. Symons's first and only season, in 2003.

- With raw numbers, Symons threw for 5336 yards on 666 attempts, for a yards per attempt of 8.01. He also threw 21 interceptions.

- For Harrell in 2006, he threw for 4555 yards on 617 attempts, along with only 11 interceptions.

What the stat guys are doing now is subtracting 45 yards for every INT thrown: they've crunched the numbers, and this is about what it takes away from you in terms of field position, scoring probability, etc.

If you did that for Harrell in 2006 (multiplying 11 times 45 yards and subtracting that from his raw passing yards) you get him 4060 adjusted total yards. Compare that with Symons' 21 INTs, which brings his total down to 4391. This makes their adjusted yards per attempt stats now 6.58 (Harrell) and 6.59 (Symons) -- nearly the same, though by different roads. Interesting, no?

The other X factor is QB sacks/runs. College stats make this hard of course: in the NFL, sacks are counted against passing yards and thus factored into yards per attempt. For Texas Tech QBs I think the safest thing is to just count the rushing attempts and yards all as part of the adjusted yards per attempt. (If this was Oregon or Tebow at Florida it'd be very difficult to do this without completely going back to the raw data and recreating the "sacks" and "yards lost by sack" statistics.)

Harrell's rushing stats in 2006 were 32 rushing attempts for -66 yards. Throwing that with the above adjusted numbers makes his new adjusted-adjusted total yards 3994, his total adjusted-adjusted attempts 649, and his adjusted-adjusted yards per pass attempt 6.15.

For Symons, in 2003 he rushed 74 times for 140 yards. Adding this to his passing attempts/yards we get 740 attempts and 4531 adjusted-adjusted yards. (I know that this number, unlike Harrell's, is actually positive, but I think it defensible to add it all back in because few Tech QB runs -- other than sneaks -- are called run plays.) So the adjusted yards per attempt is 6.12.

So Harrell actually beats out Symons in adjusted-adjusted yards per attempt, 6.15 to 6.12, though that's basically too close to make a call. I think it reinforces Seth's point that Leach has gotten it done with QBs of vastly different styles, especially considering these two guys were (probably) the best of that run by Leach where each first-year QB excelled that Harrell broke by starting more than one season.

In any event, the real point of this is to show how you might compare apples to oranges for any system or QB, with a guy like Symons who was acting as more of a gunslinger and Harrell who -- within the confines of Leach's wide-open offense -- was operating slightly more conservatively.

(I don't have exact cites but credit must be due to Advanced NFL Stats and the Pro Football Reference Blog, both of whom have undertaken similar analyses.)

Texas Tech run game cutups

For a team that throws it as much as Texas Tech does with Mike Leach, I get a surprising number of questions about the Airraid run game. The reason, I suppose, is that if you throw the ball a lot you need a good complementary run game that will take advantage of the defense when it overreacts to your passing game but also doesn't require too much practice time.

This topic deserves a fuller discussion later, but I was sparked when I saw these clips of Texas Tech's spring football. Video below (hat tip Tortilla Report via Double-T Nation):



Okay, you might be saying, I see some guys running around, but what does it mean? Again, this topic deserves a fuller treatment, but here's some diagrams and quick explanations of Leach and Texas Tech's main run plays.

Base



Base is essentially a "man" blocking run play that has each lineman block the man over them, and if uncovered, they head up to the linebackers. (The "fold" technique comes into play where there is a sort of "shaded" nose -- a defensive tackle who would be too difficult to "reach" for the guard -- so they can make a "fold" call at the line.) The play is easy to teach because it uses largely the same scheme as their main pass protection (big on big; back on backer) but uses drive blocking. Finally, this play is often mistaken for a "draw" -- it just looks like one when run from the shotgun.

Lead



"Lead" is your basic "isolation" play: all the linemen block "man on" or "down" and the lead blocker bursts into the hole and blocks the first man that shows; the ball carrier then cuts off the lead back's block. Tech uses this a lot when they get into any two-back set (whether from gun or those rare times under center). From shotgun this too looks a lot like a "lead draw," but it is really just one of the oldest plays in football run from Tech's funky wide line splits and shotgun.

Stretch



The "stretch" has increasingly been a weapon for Leach over the past few years. A big reason is that Leach is now fully committed to the wide line splits, so at some point in the game the defensive ends tend to stop lining up so far outside the offensive tackles and instead line up heads up or inside them, thus giving the offensive guy an easy "reach" block to hook the defender inside. As a result the runningback has an easy spring to the outside.

Conclusion

Leach's run game is not complicated and no one will confuse Tech with Paul Johnson's flexbone option teams, but they have had decentbalance (depending how you define it) over the years and the run plays are some of the most tried and true schemes around. He just uses them from his spread sets, and only when the defense is completely stretched out.

Thursday, March 26, 2009

Run and Shoot Series Part 3 - The "Choice" Concept

[This is Part 3 of a multi-part series on a "Simple Approach to the Run and Shoot." In one sense I mean "simplified," but the series is, more than anything else, intended to both diagnose and explicate some of the fundamental concepts behind the shoot as well as discuss how I might marry them with some passing modern ideas, all in an effort to just understand passing offense generally. You can see the full series here. Also check out Parts 1 and 2.]

The previous posts have cleared away much of the heavy lifting: we know about the basic principles behind the offense, and we know about that most important of routes, the seam read.

But the route that maybe most exemplifies the offense's variable, adapting approach to attack defenses, is probably the Choice route. And it is a concept that is used by many, many teams -- in one form or another -- across the football spectrum.

It is another trips play, and is intended to be used as a counter to the Go when the defense overplays to the three-receiver side. The "choice" in the route belongs to the singled up backside receiver -- often the "X" receiver. The idea is that you put your stud there and make the defense wrong every time, until they overshift to that side, thus opening up the three receiver side for easy plays or big ones. The base form of the route is shown below.



To the three receiver side, it is simple and familiar to what was done with the Go: the outside receiver runs a "streak read" (burst on a vertical route, but if you can't beat the defender, break down at 14-15 yards and come back down towards the line of scrimmage); the middle slot runs the seam read (attack the near safety and then have a multiple way go depending on the coverage -- explained in depth here); and the inside slot runs a five-yard drag route (explode to five yards, then drumroll the feet and head across the field; may settle down in a hole in the zone once on the opposite side of the field).

The single-side "choice" receiver, typically, has three options: run an out at 10-12 yards (or a comeback at 15); run a glance or skinny post (cut on the seventh step at a skinny angle, never crossing the near hashmark); or run a vertical go route. How does he know what to run? And how does the quarterback know what he's running?

The Choice Route Itself

So how do you handle the choice route itself? The R&S guys themselves -- Mouse Davis, Jones, Jenkins, and all the rest -- typically taught this as a true "read" route: it was all done on the fly. The QB and receiver simply had to be on the same page, and they were confident that they could complete this pass whenever they wanted.

To run this route I don't think you have to commit to this; admittedly it takes a great deal of practice time and young quarterbacks and receivers have plenty else to work on. But, in brief, the R&S guys did this about how you'd expect:

  • The base route in the choice concept is the out -- the receiver wants to sell that he is going deep and then to break to the sideline. If done correctly, it is difficult to defend this route. For the speed out, the receiver would burst upfield for six steps and then roll on his seventh and eighth steps to the sideline, driving his outside elbow to turn his body around. He then would flatten to the sideline and expect the ball at about twelve yards. (The comeback works how you'd expect -- sell deep then break down at fourteen to fifteen yards to the sideline to a depth of about twelve.)




  • But that's not the only option. If the receiver go to the top of his route and the defender pressed him, he could either run a skinny post (glance) or go route. If the defender sat on the out or generally tried to play any outside leverage, the receiver would break for the skinny post on his seventh step.




  • If the defender pressed and took away the post with inside leverage then (and only then) would the receiver continue to streak up the sideline.



  • Below is a clip from Mike Drake from Longmont High (CO), where they throw the choice route backside against Cover 1 (man-to-man with one deep safety). (They keep some extra protection in on this play so it is a three-man route, but it's the same concept except without the drag.)



  • Against Cover Two this worked a bit different, because the rolled up corner would play a type of press coverage, but with outside leverage (trying to force the receiver into the safety) and would release the receiver up to the safety. As a result neither the out or glance are good looks, but the streak is good, because it is either open on the sideline before the safety can get over or he has opened up the middle. This wasn't too difficult of a read because the receiver would quickly realize that the corner wasn't playing him anymore, so his read became the near safety.




  • But, as observed above, the quarterback and receiver both had to read all this on the fly -- not always easy, and it certainly requires a lot of practice. As a result, what most teams do (and I recommend) is to handle all of these "choices" as a pre-snap adjustment between the quarterback and receiver by hand-signal.

    The rules basically work the same as before, though since it is predetermined you can expand the options to include corner routes or anything else. Specifically (thanks to Ted Seay for these):

    If the corner plays inside leverage, run a speed out at 10-12 yards;

    if the corner presses, run the go/fade route; and

    if the corner plays outside leverage, run the skinny post/glance route.


    Voila. But how do they communicate this? Well, it ain't rocket surgery, but it can be done either verbally or by hand signal. One way is to use very specific ones, just with the quarterback making his signal behind his back. The other is to do something as follows:

    The quarterback, before the snap, will hold his hands in the usual ready position, but with subtle variations:

    Outside hand slightly higher than inside hand = Speed out

    Inside hand slightly higher = post glance (in breaking route)

    Hands slightly higher than normal = go/fade


    There's tons of other variations. If you (for some reason) keep the choice between one of two routes, then the "signal" can just be eye contact between the quarterback and receiver pre-snap. Finally, another twist is to let the receiver make the decision, and signal the route by switching his feet in his stance and then back, or by where he places his hands -- the list goes on.

    The obvious downside, however, with doing it all pre-snap is that the defense can sometimes fake you out: the corner can play way off and then come up at the snap to play a press technique, or vice versa. But those kinds of fears vary depending what level you're at. If you're in the pros, where they do all that stuff on every play and Ed Reed will play the deep half of the field while lining up in the guard-tackle gap faking a blitz, then you ought to be able to teach people to read on the fly. If you're in high school (or college really too), then the pre-snap stuff should be more than sufficient.

    So that's the Choice. It's a great route, especially if you put your best receiver there, as most teams do. Yet, if you're too much of a wuss to do even the pre-snap decisions, you can always "lock" the route and just signal it in from the sideline. That's permissible too, because you still get a one-on-one with great backside capabilities. Let's turn to the backside now.

    Backside and variations



    As we can see, the backside has lots of options. The two most important are the drag and the seam-read.

    First, if the weakside linebacker or flat defender tries to widen out to stop either the speed out or glance, then the drag route should come wide open in that voided territory. See below.



    Similarly, in the above diagram, if both the choice route and the drag are taken away, the quarterback will look to the free safety -- he is probably cheating too far to the single receiver side and therefore the backside seam should be open. As an example, see the below clip from Mike Drake again, this time against Cover 3. The defense brings the ever popular "Magic Blitz" or overload zone-blitz with three-deep behind it. They don't block it quite right, but the quarterback moves his feet. (They also put their TE on a kind of "climb" route to help draw the free-safety, which is exactly what he does below.)



    To better clarify, here is a diagram of what the receivers were doing and the free-safety's movement where his taking the tight-end opened up the seam reader.



    Against a cover two, the calculus changes slightly but the basic progression and read is the same. See below, and then watch the clip again just paying attention (as best you can) to the free-safety.



    To stop both the go route by the single receiver and the drag by the slot, the linebackers, corner, and safety have to overreact to the single-receiver side. As a result, the quarterback should be able to work the deep hash safety to the three-receiver side, who has a two-on-one with the seam-read (now running a post) and the backside streak. If the middle linebacker tries to retreat to take away the post (common with the so-called "Tampa Two" defense) the quarterback still has both the drag runner who has settled into an open spot and the running back that he can drop the ball to underneath for a catch and run. (Keep throwing those check-downs until the defense comes up for them; that's when you gash them for the big play.)

    Again, another clip from Mike Drake. Here it is against Cover-2: the deep hash safety plays very deep, however, so the seam-reader throttles it down a bit into the void for a nice completion. See below.



    So that's the basic framework. Really, you can just teach the quarterback to read: (1) choice, (2) drag, (3) seam-read, (4) backside streak-read, and (5) (outlet) the runningback on a "leak-out" route.

    Below are a few variations to the backside. The two most common just switch assignments. In the first, we switch the two-slots so you can get a "rub" for the drag receiver's man.



    In the next, the two outside receivers switch assignments. I will discuss the true run and shoot "switch route" in a later part in this series, but this illustrates the basic gist. Again, all you're doing is changing assignments.



    The next is a slight variant on the traditional run and shoot formation because it uses the "bunch formation," where the three receivers tighten their splits so that they line up no more than one-yard apart.



    This set lets the receivers get more rubs against man coverage -- you can see from the image that now the outside receiver runs the drag and any defender playing man to man on him will have a difficult time covering. This is a great response to teams who think the way to play you on choice is to go to man coverage. And if they stay in zone, well you have all the good zone-stretches I outlined above.

    The final variant is the most different, but also the one most increasingly popular: "levels." I have described previously how the Indianapolis Colts use this concept, but June Jones has really used this route ever since he got to Hawaii. (Note they will do all the same switching of receiver assignments I outlined above in it.)



    Here you get both a rub and a high/low type stretch on the inside defenders, typically linebackers. Jones has liked this because it is easy for the quarterback to read, he really just must progression sequentially from the single-receiver and scan across the field. It's not a perfect visual, but below is a version of "levels" with from trips #1 running the deep-in and #2 and #3 running the quick ins.



    And, although not quite the same, below is an example of the Green Bay Packers running a version of levels to the three receiver side:




    Conclusion

    And that's the choice concept. It highlights much of what drives the run and shoot: a well-designed route intended to set up a receiver with many options, combined with a great basic combination with equally as many options.

    As a final note, the run and shoot is a four wide-receiver offense. That is how it was designed, and how it is run when one commits to it fully. If you don't use four wides -- for example, by substituting in a tight-end -- many of the purists would say you aren't a run and shoot team anymore. I will leave that debate for a later day. I just want to point out that it wouldn't be too difficult to imagine the Choice where the drag runner is actually a tight-end instead of a slot receiver. Indeed, many pro teams would agree with you.

    Wednesday, March 25, 2009

    Tebow's new QB coach: Scott Loeffler

    USA Today:

    When it came time for Urban Meyer, head coach of the defending national champion Florida Gators, to offer some advice to his new quarterbacks coach on how to handle Heisman Trophy winning quarterback Tim Tebow, the message was pretty simple.

    "Coach Meyer said just don't screw him up," Scot Loeffler said.




    All well and good. But we know the real reason that Meyer brought in Loeffler: to help Tebow become more "NFL Ready." And Loeffler has some serious bona fides, including his years doing wonderful work with quarterbacks at the University of Michigan (like Chad Henne, Tom Brady, and Brian Griese), though his encore was spent this past season with the Detroit Lions just trying to teach his quarterbacks the simple things, like, you know, not running out of the back of the end zone for a safety to seal a loss.

    As I have previously written, Tebow needed to put "becoming a better pro prospect" at the head of his "To Do" list. And many -- including Tebow in the USA Today piece -- cited Loeffler's hiring as a reason he came back. But I'm not so sure that this is the perfect marriage. I mean, I think for Florida Gator fans even a slightly confused or tinkered with Tebow is better than the next guy, but what about for Tebow?

    Loeffler can definitely help in a few major areas: footwork (avoiding the famed and slightly ridiculous "spread quarterback" label); reading coverages; accuracy; and the ability to make a wider variety of throws than Florida asks him to. But again, as I said before, I'm not so sure that he wouldn't get all this and more actually in the pros. All he turns down is a set amount of money (but not a "fixed" amount because an increase in Tebow's draft stock is purely speculative), aside from of course, more college glory and the "college experience," which did not serve people like Matt Leinert all that well and Tebow has achieved all that is humanly possible in college.

    And I still think it ignores a big drawback for Tebow: all the hits he takes. He essentially takes hits like a runningback does, and the studies are voluminous showing that runners deteriorate, not so much over the course of a season, but at some point in their careers.

    So the real parties in interest, Florida and Urban Meyer, now must try to satisfy two masters: win games but also make Tebow more of a "pro-guy." Do you try to call fewer runs for him? Do you emphasize different points of the game? Do you take some of the (limited) time Loeffler gets to spend with him and dedicate that to pro-style stuff rather than the immediate gameplan?

    Don't get me wrong, Tebow becoming a better dropback passer can only help Florida, and Loeffler is a very good coach and is a great addition no matter what. But for Florida's sake I hope they don't fixate on this too much, and for Tebow's sake I hope he isn't expecting more focus on his pro prospects than might be appropriate. Time will tell.

    (Hat tip: Dr Saturday)

    What might the University of Tennessee's new offense look like?


    Among the reasons that Lane Kiffin was hired at Tennessee -- other than to stir up various controversies and to publicly go after both Steve Spurrier and Urban Meyer of course -- was to revitalize a stagnant Volunteer offense. And, other than his stint in Oakland (where offense goes to die, just ask Randy Moss) Kiffin sports some some fairly impressive offensive credentials, i.e. his years at Southern Cal first under Norm Chow and later as co-offensive coordinator with Steve Sarkisian. To aid him in bringing potency to the offense is the Vols' new OC, Jim Chaney, who is best known as the offensive whiz who brought basketball-on-grass to Purdue (along with Joe Tiller and Drew Brees). Chaney is most recently of the St. Louis Rams with Scott Linehan, but, much like Kiffin's time in Oakland, the less said about that the better.

    So what will the UT offense look like in 2009? Hard to say, but it is likely to be a blend of the USC offense and what Jim Chaney did in college and the pros.

    Since Pete Carroll and Norm Chow put together the USC offense and it took off (there is some minor controversy about who should receive most of the credit), Southern Cal's offense has been built around a few basic features:

    1. It is pro-style in the sense of formation and personnel: They use a tight-end, they keep the quarterback under center most of the time, and use a variety of formations.

    2. The running game is based around zone blocking, which focuses on double-teams at the point of attack and gives the runningback freedom to hit it playside or cutback; wherever the crease is. This kind of running works well from one-back sets and multiple formations, since it doesn't require (though it can use) a lead-blocker and the rules for the linemen stay the same regardless of whether there are two tight-ends or four receivers in the game.

    3. The passing game is a steady dose of simple dropbacks and quick, three-step passes, but with plenty of play-action is thrown into the mix for the purpose of striking for big plays. Think Indianapolis Colts in terms of play-selection, though with more quick, three-step passes, like the "spacing" concept:



    Also, you can get a flavor of the old USC offense by watching the below highlight video of Palmer:



    What Chaney has historically done is actually quite similar: He too has long utilized one-back sets, five-step and three-step passes, and the run-game is all zone blocking based. So it's a good fit, which is one of the things that Kiffin had long made clear: he wanted to find someone to call the plays but run his system.

    The biggest differences between Chaney's system and Kiffin's -- at least as highlighted during Chaney's time at Purdue -- was Purdue's total commitment to the spread, including the shotgun and lots of five-wide sets, and their receivers' heavy dose of "option routes," which give the receiver the freedom to cut in or out (or curl up) depending how the defense plays them. That said, Norm Chow had these routes in his arsenal too, so it isn't like Kiffin is unfamiliar with them. Diagram of Chow's version below (hat tip Bruce Eien).



    Compare the Drew Brees highlights with the Carson Palmer ones from above.



    Bottom line

    The upshot is that it's still too early to tell -- and I will have to wait until closer to the season to give the Vols' offense a fuller analysis -- but don't expect an Urban Meyer or Rich Rodriguez style spread offense, but neither should you expect the old West Coast Offense either. The formations will likely be basic one-back ones with a mixture of three-, four- and five-wide receivers, but with the ability to "get big" with tight-ends and fullbacks when the situation requires. In other words, they will be multiple. Below is a clip of Jim Chaney answering tentative questions about the offense.



    If you want more specifics on the dropback game, check out this post from Trojan Football Analysis on Norm Chow's passing offense. (And see my Airraid post that includes routes and reads from Norm Chow, whose offense Mike Leach's Airraid is a steroid-infused and mutated version.)

    And for run game specifics, might as well get them from the horse's mouth: Alex Gibbs, run-game guru of the Denver Broncos and Atlanta Falcons, who taught Chow, Carroll, Sarkisian, Kiffin, et al. how to properly run the zone run game.



    UPDATE: In response to an email, I thought I'd mention that Jim Chaney helped orchestrate the the single worst whipping of a Nick Saban defense I've ever seen (and I have watched this tape a bunch of times): Purdue's 52 to 28 victory of Saban's then #7 ranked Michigan State Spartans. Michigan State had no answer for Brees: Purdue took a 28-6 lead on four first-half Brees TD passes, and for the game he was 40 of 57 (!) for 509 yards and 5 touchdowns. As Saban said, they were "humbled." As I said, I can't think of any other game where a Saban defense just got destroyed in that way, and it was Chaney there calling the plays.

    The bad news? The next season (when Purdue went to the Rose Bowl), Michigan State beat Purdue 30-10. Saban's pretty good at making adjustments.

    UPDATE 2: The good Senator over at Get the Picture chimes in with good thoughts. He accurately notes that many observers noticed that playcalling under Kiffin and Sarkisian at USC was more erratic than Norm Chow's "surgical precision." Then again, it's supposed to be Chaney doing the actual play-calling, so as long as Kiffin's scheme is sound, all should be well, right? We'll see.

    Tuesday, March 24, 2009

    Getting Vertical with Dan Gonzalez

    I asked Dan Gonzalez, author of the new book "Concept Passing: Teaching the Modern Passing Game, to discuss his take on the famous four verticals" route concept. I've discussed it only briefly, and even then as part of discussing other concepts.

    So this is a great opportunity to look at the play in depth; it is a foundational route in several respects. After Dan's lucid explanation of the route I have some video and a brief description of how the Airraid guys run this route. But I'll begin with a brief overview before we dive into the details.

    First, the concept is a great introduction for coaches, quarterbacks, and all players (as well as fans) to how to "stretch" or break down coverage. At core, the route involves four guys running "vertically" -- hence the name. They split the field four ways, and as a result typical "two-deep" (Cover two, Tampa Two) or one- or three-deep coverages (Cover three, Cover one man, certain zone blitzes) should not be able to defend the route.

    Although when you talk about a "vertical" or "streak route," most people think of a deep bomb down the sidelines, the four-verticals concept really attacks the safeties; the outside routes will be thrown at times based on matchup, but usually you're trying to make the free safety wrong.




    Above the the typical route setup. It carries some dynamic aspects, most usually the use of a "bender" receiver -- an inside receiver who will "bend" to a post route when the middle of the field is "open" (i.e. two deep safeties on the hashes, rather than a single one down the middle), and will stay on his seam route up the hashmark when there is a deep middle safety, as with Cover Three. Most college and even pro teams run it this way; it's a good play. I've discussed this "bender" or "divide" route in more depth previously.

    But Dan's take is interesting because it goes a step further, which reflects his version's run and shoot origins. The "bender" receiver actually is given a far more dynamic "seam read," which gives him a wider variety of options to get open against the deep coverage, and even to come underneath the safeties if they play too deep. Also, the outside receivers are given "steak reads," where they are not locked into just going deep; they can stop and come back to the ball if the defender plays them deep.

    As an example of this, on Texas Tech's famous final drive against Texas this past season, the Red Raiders called four-verticals on nearly almost single play of the drive (there was a bubble screen on at least one play). They never hit the seam routes and never threw a deep bomb, but they hit the outside receivers on "fade stop" routes, including on Michael Crabtree's famous game winner. (Tech's technique tells the receiver to go deep, but if he can't get deep -- and the defender is in press coverage with his back to the quarterback -- the QB will rifle the ball to the back of the defender's head and let the receiver stop and make a play. Video of the final drive and final play below.)



    Finally, Dan's approach is also good because he shows how the play can be readily adapted to different formations while keeping it simple for the players (and coaches). As a side note, when people talk about being "pro-style," this is what they are talking about, or at least what they should be. Versatile, attacking passing concepts that let's the offense do a variety of things to a defense. In my view, unless you're an option team (and maybe even if you are), this is how your should be running your four-verticals play.

    Dan Gonzalez's four verticals, in his words:

    [Ed Note. This part is all Dan.] The "four verticals" pass play has become increasingly popular at all levels of football. What we have done that’s a little different is conceptualize its teaching. This gives us more flexibility and allows the time we invest in it in practice to carry over to other parts of the offense.

    Let’s look at the basic setup:

    1. We have one guy on a locked seam (Usually the "Y" receiver; vertical in 2-2 or crossing the hash in 3-1) who "owns" the hashmark AWAY from the seam reader, who we always consider part of the "frontside."

    2. The seam reader is taught to take the highest angle through the defense (which eliminates the confusion of Cov two, three, four, etc.). His two caveats are: he must cross the face of any man defender [ed. break inside on a post or in-breaking route] and he must look to the quarterback as he passes by the linebacker (thus, versus a blitz, he looks as soon as he comes off the ball).

    3. The outside wide receivers run streak reads: first, win deep if you can, but break down if you cannot. They will make this decision at 10 yards, while running full speed. If the defense maintains deep leverage, the receiver must slam on the breaks and come back down the stem of his route (i.e. rather than an out or an in or curl) towards the line of scrimmage.

    4. The back drags toward the seam reader.

    With twin receivers to both sides:



    From three by one (trips):



    Quarterback: The QB starts his eyes on the locked seam, timed up with a quick five-step drop (from shotgun, a quick three-steps). The quarterback will throwwill turn it loose with no hitch step. If he has to hitch, his eyes must move from the locked seam to the other slot, the "seam reader" (this allows the pattern to develop). [Ed. Note: This means he takes a full drop but must be ready to throw the moment his back foot hits; you see a lot of QB's "hitch up" or step up in the pocket, which is fine, but the first read must be thrown as soon as the back foot hits. It is what is known as a rhythm throw, and the ability to do this often separates quarterbacks. Also, it's not just an arm strength thing, it's a discipline, footwork, rhythm, and timing thing, though arm strength can sometimes be a factor.]

    The quarterback must be ready and have a window with which to deliver the ball to the seam reader. If he can't, then its not open. If he hitches again he has to move his eyes to the back.

    The backside streak read comes into play if the free safety's pre-snap location puts him in a spot which tells the quarterback that the locked seam will be there (this is referred to as "displacement"). If the locked seam isn't there when this free-safety displacement says it should be (the backside linebacker must be taking it away), the QB hitches and throws the 1 on 1 streak read.

    The big thing about the way we teach the four verticals concept is that when we get to these other looks (diagrammed below), we introduce absolutely no new learning for the QB or the key players:

    From two backs:



    From no backs:



    With “Switched” releases from three-by-one or two-by-one sets, respectively:





    Using an outside receiver as a drag runner:



    Using the tailback as a vertical threat:



    Variations on a theme (other approaches to four-verticals)

    [Back to Chris] So that is Dan's introductory breakdown of the four-verticals concept. As noted above, there is more about this concept to be found in his book, along with lots of other material. I recommend it.

    Below is video of some Airraid teams -- Texas Tech (Mike Leach) and New Mexico State (Hal Mumme) -- running their version of the four-verticals route. It is similar, with these differences:

  • They use the same "landmarks" approach with the outside receivers up the numbers (about 4-5 yards from the sideline) and the slots just up the outside of the hashmarks. (In high school the hashmarks are wider so the receivers just head up them; in the NFL they are narrower than college so the landmark is further outside the hashmarks.)


  • The outside receivers use a similar technique, except as observed above they use more of a "fade-stop" approach than Gonzalez's true route conversion to comeback to the football. If the receiver cannot beat his defender deep and that defender has his head turned away from the quarterback, the QB has the option of drilling the ball at the back of the DB's head, thus letting the receiver make a play.


  • The inside receivers don't have a "bend" or true open/closed read. Instead, the Airraid guys just have them try to get deep, but if they get to 12-14 yards or so and cannot, they will break down and effectively run curl routes at about 14-16 yards. In this way they get their variety and can attack different coverages.


  • The runningback runs a true "option" or dodge route: he pushes to a depth of five yards and tries to "step on the linebacker's toes" (break any cushion). If the linebacker plays inside leverage, he breaks out; if he plays outside, the RB breaks in. If it is zone he will settle down in the hole underneath. As you can see from the clips, the RB often comes wide open when the defense retreats too much for the four-verticals.

    Otherwise the QB's read is basically the same: he reads inside to out from the free-safety and looks for the best one-on-one matchup (Michael Crabtree). The cut-ups are below, thanks to Court Allam of Olathe South High School (Olathe, KS).



    Finally, below is some video of of the "verticals" concept (though with some variance to both what Dan does and the Airraid guys does) courtesy of Bruce Eien, coach of Brethren Christian High School.



    Wrap-up

    The four-verticals is a staple of nearly every modern passing game, be it spread, pro-style, or option-based. But even though it is common doesn't mean it can't be better. Dan's approach should give food for thought to anyone who has just been sending four guys down the field (or watching their team do it) without the minor coaching points that make the play really go.
  • Monday, March 23, 2009

    Matt Stafford and the Wonderlic

    So, the guy in the photo to the right did better on the infamous Wonderlic intelligence test (given to potential NFL draftees) than you probably would have. (He scored a 38.)(If you want to try a sample, do so here.)

    For comparison, Mark Sanchez and Josh Freeman scored 28 and 27, respectively. Stafford fell just shy of QB Wonderlic mavens Eli Manning (39) and Alex Smith (40). (Vince Young reportedly got a 6.) At wide receiver, Michael Crabtree only got a 15, Percy Harvin got a 12, and Darrius Heyward-Bey got a 14, while Jeremy Maclin snagged a 25. (Pat McInally, punter, wide receiver, and Harvard graduate is the only player confirmed to have scored a perfect 50.)

    Apparently this puts Stafford in good company. According to Wonderlic, he has a high enough IQ for any number of professions besides pro quarterback:

    Chemist: 31
    Programmer: 29
    Newswriter: 26
    Sales: 24
    Bank teller: 22
    Clerical Worker: 21
    Security Guard: 17
    Warehouse: 15


    But what does it all mean? Does the Wonderlic actually help predict who will be a good NFL quarterback? I previously discussed this question here. And, to see a visual representation of the average scores for each position in the NFL, see the chart below (hat tip Ben Fry):



    So, while Stafford's 38 really jumps out, Crabtree's 15 doesn't seem all that bad. The interesting thing is what was noted in the original post: the closer you are to the ball, the higher your score.

    But the question remains whether the Wonderlic is relevant at all. Indeed, Hall of Fame QBs Dan Marino and Terry Bradshaw did quite weak, both scoring a mere 15. Here's what I said previously:

    Yet two anomalies do not disprove the notion. Charlie Wonderlic believes that "What the score does is help match training methods with a player's ability, [like the ability to understand] a playbook. [O]n the field, the higher the IQ, the greater the ability to understand and handle contingencies and make sound decisions on the fly."

    I don't have a firm answer either way. I think there's nothing wrong with giving the test, and different teams appear to put varying degrees of emphasis on the test results. Some only care if a player scores extremely low or extremely high, while others take the test quite seriously. The image above clearly indicates it matters more or less depending on the position. Advanced NFL Stats has previously discussed studies that attempted to chart out QB performance as a function of their Wonderlic results. My guess is that the Wonderlic is a weak predictor in the same sense as the 40-yard dash, shuttle run, and the bench press: If you chart out performance with those combine statistics, although you will see a positive trend, it will be full of noise, will not give you predictions with high certainty, and counterexamples - like Dan Marino - will be abundant.