SB Nation - Login for mobile commenting

Dawgs By Nature

Brian Daboll and Constraining a Defense


One of my biggest issues with Brian Daboll has been the lack of effective constraint that our offense has on defenses. In other words, Daboll doesn't punish defenses for being over-aggressive. I can't always tell what Daboll's line of thinking is but one thing is for certain; the Cleveland Browns do not constrain defenses well enough and in the end that's Head Coach Eric Mangini and Offensive Coordinator Brian Daboll's responsibility.

"Constraint plays" are usually plays like draws, screens, and play-action passes. They are designed to keep defenses honest. Take a screen, for example.  Screens work best when a defense is spending a lot of it's resources on stopping the pass; the defensive line has it's ears pinned back and they are coming after the QB hard, the safeties aren't cheating up, and linebackers are over-willing to drop into coverage or rush the passer. No defense plays that way without respecting the offense's passing game.

While our lack of success on screens, draw plays, etc. might not be attributable directly to Daboll (but instead the lack of a threatening passing game), I believe our offense also does not do enough to punish defenses from overreacting to the cues we give them.

Star-divide

Predictability and Our Plan to Constrain

One such cue we give defenses is that when we have one receiver break immediately in and all other receivers push vertical, we are often running a version of the "Drive" or "Hi/Lo" play.  Basically the concept is to have one receiver run a shallow cross and another run a square in behind him, creating a vertical stretch on anyone playing zone over the middle. One guy can't guard both these routes so a receiver has to be open:

19_medium

In a perfect world against an honest defense, there should always be someone open with this route.

But because the defense can see the first three yards of the routes and know exactly what is coming, they have time to coordinate themselves to cover the route. In other words, the route gives easy and early-developing cues for pattern-reading. Defenses see this and can fill in the blanks:

18_medium

(note: I should have highlighted it more, but Robiskie's motion in and then immediate inside break is the big giveaway for the defense, not the vertical stems of the other receivers.)

Many offenses are built around "vertical stems" to avoid this problem: defenders don't know where receivers are going until they make their break to get open. These route stems could turn in to any number of plays:

17_medium

16_medium


15_medium


14_medium

etc...


Defenses can still pattern match these routes (a trendy coach-term for this is "relating to the routes") but they can't do so as easily and they can't do so as early in the development of the play.

Now, there's nothing that says our Drive play can't be just as unpredictable. We just need to punish a defense that sees this and immediately jumps the routes.

The problem isn't the Drive play itself--many, many teams at all levels run this passing play and it works. The problem is that we run this play over and over and over but we rarely counter off of it. We rarely ever punish a defense that sees the first three steps of the route and assumes we are running this play.  That is, we don't constrain defenses well enough.

The coaching staff's strategy is to get a big gain when we do counter off of the play, which is sometimes an effective strategy.  Constraining a defense in this way, however, makes us heavily reliant upon a very small percentage of plays to generate big plays in our offense and to scare defenses away from aggression.  With so few opportunities if we don't capitalize, we're screwed.

 

Creativity and Credit Where it is Due

While I bash Daboll often, I have to give him credit where it is due.  Down 13-7 at the end of the 2nd quarter and in the red zone, Daboll broke several tendencies in a really intelligent way; first, he was aggressive in his approach down in the red zone--really going for a TD, second, he ran a play that surprised the Ravens because it countered off of our version of the Hi/Lo concept.

Just as in our normal Hi/Lo, all receivers pushed vertical except Brian Robiskie, who immediately broke in for the shallow cross. Watson, tight to the formation on the right is running the "hunt" route to complete the Hi/Lo concept. I can't really tell what Stuckey is running on the left of the formation but my guess is something out-breaking like a corner or a comeback.  I'll draw the corner below. Massaquoi, wide right, ran a post to hold the safeties deep off of Watson and to clear the right side of the formation out:

13_medium

10_medium

09_medium

Somewhere around this stage of the play Ray Lewis and Ed Reed probably know what is coming.

08_medium

07_medium

06_medium

The ravens were ready; one of the linebackers dropped to cover Watson, the DBs took away the deep routes, and another linebacker jumped Robiskie's crossing route:

05_medium

You can see from the endzone angle that Watson is covered high and low by a Safety and LB.  Massaquoi is squeezed horizontally by a safety and a CB off the screen to the right.  The LB underneath to the right near the numbers is moving to get in position to take away Robiskie:

24_medium

 

Surprise

Only this time, we weren't just running the Hi/Lo. Daboll had tagged Robiskie on almost a sort of wheel route. The linebacker thinks he knows the route but we used that aggressiveness to our advantage and Robiskie turned upfield to the endzone. Colt read the play correctly and threw to the correct route, but uncharacteristically threw an inaccurate pass way out of bounds over Robiskie's head:

21_medium

In the image above, you can see Colt already throwing the ball before Robiskie hints that he is going into the endzone, which is why I don't think this was an improvisation. 

04_medium

The LB may have also briefly thought about Mike Bell, standing near the 10 yard line. All it takes is .1 seconds of hesitation on his part and Robiskie is wide open:

12_medium

 

03_medium

Look at all of the space that Massaquoi's post cleared for Robiskie, there's no one in that third of the endzone:

25_medium

You can barely see the ball but if Robiskie is 6'3" it was a good 12-15 feet in the air when it went out of bounds. You can see it above the referee's head and between to guys on the sideline, 5 yards out of bounds and still about 6 feet in the air:

01_medium

Credit Daboll, he used the ravens' aggressiveness against them, he used what was effectively a constraint play in an aggressive attempt to score. My only complaint is that I wish Daboll would do this sort of thing more. As I mentioned above, because we don't run this sort of play very often, we NEED it to work when we do run it.  We needed a TD on this play and we didn't get it, which is pretty much how this game went for us.

3 recs  |  53 comments

Comments

Good write up Rufio. I’ve learned something new for the day.

I remember seeing this one from the stadium but had no idea I was looking at a constraint play. Great post and as always thanks for taking the time to break this down for all of us that are lower on the football IQ spectrum.

Wow, Rufio. Really great, insightful post here. Without completely removing the mask, can you provide any info on your background as it pertains to your robust understanding of football tactics and strategy? Grazie . . .

I think he was birthed of football gods and had both an offensive and defensive playbook in the womb…

I don’t coach anywhere, I just read a lot of books, blogs, playbooks, coaching clinic materials, etc. written by coaches and I watch Browns games several times to try to understand what we do specifically (when I get the chance). I’ve been fortunate enough to talk with a few coaches/former coaches and some of the CU football team when they’ll talk with me.

I’m not out there on the front lines or anything, I just understand some of this stuff.

I’m not out there on the front lines or anything,

though you could be.. I try to read whatever I can on playcalling but some of the stuff you see is very impressive. I am just curious how long it took you to break down this play.

(also, could you post a link to some of these blogs you speak of?)

this is a play he should be able to hit as he gains expierence. I can remember BQ hitting some of these, and one in particular that a RB dropped in the endzone….forget who that RB was, though….maybe last year or 2 ago.

He will be able to hit these and he is already doing a pretty good job. Just did not have a good day, though.

Very nice write-up rufio – very enlightening. I’ve been looking foward to seeing this (you mentioned in the recap thread you were going to do this). This is also a great breakdown of that particular play.

you can see Colt already throwing the ball before Robiskie hints that he is going into the endzone, which is why I don’t think this was an improvisation

Very interesting – especially to me because I was very curious about that play and the next day I watched it several times (nfl rewind) because I wanted to understand what happened on the play. I mentioned in one of my comments in the recap thread that the more times I watched the play and payed attention to Colt’s movement it appeared to me he had started tracking Robiskie early in the play’s development – which I think is consistent with your thinking that it was a designed route.

NFL rewind is an amazing thing. So much clearer than my DVD-R.

If I am Daboll, I would tell Colt to probably read a few keys on his dropback (getting a feel for what kind of defense they are in, especially looking at the safeties) and then to look for Watson and Massaquoi over the middle. If Colt pays attention to them, they can really suck the DBs inside and then the play they are trying to hit (Robi) should be wide open. He might instruct Colt to pump fake at the shallow or the dumpoff to incite more of a bite by the LB as well.

Wherever Colt was looking, he did a good job of manipulating the defense, we couldn’t have gotten Robiskie any more open.

But have we factored in on how slow that Robiskie ran the route? To me it looked like Colt had given up on Robiskie and threw it out of bounds. It really, really did not open up to very late.

It is supposed to be thrown before Robi actually gets open and it was thrown right on time. It takes a long time to run the shallow cross and even longer to run the shallow-to-wheel that he was tagged on.

It was a slow developing play, which I didn’t like because of the lack of time at the end of the half. But Colt said in his press conference that they knew baltimore would rush 3 and drop 8 in the redzone because that’s what they do, so I don’t mind the play call versus a 3 man rush.

Really nice explanation, Rufio; you never fail to impress. I’m curious, how much of this do you pick up the first time you see the play (live)? Do visions of X’s, O’s and arrows superimpose themselves on the screen?

I think I maybe got excited when I saw it live (we finally countered off of the Hi/Lo!) but honestly you can’t see a whole lot of the field from the normal TV angle so I didn’t really know what was going on. I think I understood what we did when they had the endzone angle replay but I always have to watch the tape to make sure because there is a lot going on.

So much of the NFL is small tweaks—tagging this shallow to turn upfield for example—little changes to “normal” passing concepts that it’s really hard to keep track of everything live.

If you can get a first down with a run, that´s you´re first option, every play.
It´s only when they overcompensate that you throw the ball, if you´re doing it right, with an intimidating tandem at fullback and halfback, that will be every play.
Yesterday I read a Sid Gilmann quote: “The problem with the pass, is that two of three things can go wrong.”, quite likely to mean interception or incompletion. Why push the option, where two of three things can go wrong, over the option where only one of two things can go wrong, run success or run failure?

This one I could decipher….I think you and Sid are cut from the same cloth :)

Let me push the odds some more. With a quarterback that can take off, the fullback dive option, the tailback with the fullback blocking for him, and maybe a pitch option with the flanker, it´s a strong 4 of 8 probability. Is 4 of 8 stronger than 1 of 2?

This is how a little knowledge can be a bad thing. The point that 3 things can happen implies that each of those things has a 33% probability of occurring. If you buy a lottery ticket, 2 things can happen a)win b)lose. You can play this game with a lot of things. As often as not, teams set up the run with passing. There is nothing magical about either running or passing. Forcing defenses to respect both equally, keeping them honest, is what Rufio’s example is about.
Any fool can figure out that when the defense overcompensates to run you throw and vice versa.

You missed what I implied. I was talking about improving the odds. I´m saying that the old catch line “establish run, to pass”, is overly simplistic. I thought my delineation was sound logically, but lockjaw brain tactics are still evident here.

You gave it away before I had a chance to reply….devil.

I´m trying to understand what you were saying.
Probability is a mathematical subject with it´s own calculations, I had some of that in school. Odds are not haphazard if seen within a mathematical exclusiveness. 33.333….percent is a solid fact. But we are dealing with reality, it is not calculable but only guesstematable, meaning 1 third chance of incompletion is subject to similarly incalculable factors, such as QB armstrength, WR dexterity, windchillfactor, fanencouragement…ect., ect., ect…In other words, if math is barred, you can optimize performance and change the odds, the historic odds later billed stats.
What you pronounce here is the age old problem of past, present and future, versus the human weakness to predict accurately, usually achieved by emphasizing anything other than the present. The strength of a statistical mind set, is that there is a before and an after, both strongly contestable by yourself and the adversary. The problem with this is that you are allowing yourself to mutate your behavior, thought established in written or spoken form is also a deed. Your mathematical preponderances are making you drift further and further away from reality. The quotients become palpable in our computable actionisms.
I think what you are saying is: 33.3% + 33.3% + 33.3% can never equal 100, because the chance of failure at each, pushes the general odds towards failure. F.e. reduced incidence for each occurance states 12% + 22% + 31% does not equal 100%, but in xyz would have to look like this: 12% + 22% 31% = 21% + 9% + 3%, so in the ends the entire conundrum is reduced to 12% + 22% + 31% = 55% = 21% + 9% + 3% can never equal 100%. By delineating that there is an attainable odd or several you are distracting from your true belief, that it is safer to calculate end failure, than end success. Input, Output and Endput, so to speak. Humans must be defeated by their tools and science. I can´t touch this without smudging it. In the example, you have calculated that the odds adjusted by realistic propensities, gives any of the odds an overall possiblity of 55%, rather than the 100% it warrants in conclusion of implementation. In essence your claim is that due to the pluralistic tendency in trying to decipher the odds, perfection becomes less attainable. You´re basically questioning the sanity of the whole process, given mathematical scrutinability versus human struggle.
I´ll take x(-x) + y+(-y) + z+(z) = x+y+z (-x-y-z) over 00+0=0, in favor of a game even taking place.

Also interesting: If the quarterbacks propensity equals 100% if he throws it, does each receivers propensity of 33% towards catch, drop and intercepable, equal that the receivers have more potential than the quarterback? As in potentiating their control of the odds towards the gameplan versus the quarterbacks simplistic 100% attempt. Are they superior because of the modern game´s intense mathematical scrutinability? Jay Cutler anyone?

The solution is for the quarterback to throw the ball at the receiver, not to him. It must not be easier for the receivers to implement failure than the quarterback. The trend is for the QB to not be accountable, yet ultimately responsible, and for the WR to be instantaneously responsible, yet not accountable in the end, due to their almost exponential potentiability towards failure. Their greater odds at failure, strangely enough, makes them ultimately less liable. Old fashioned American suckerdom holds it that QB=leader=responsible no matter what=my salary versus your cap space=WR success/failure odds control versus his lack of cap space. It´s liability ping pong.

Thanks to the Bears, the Offensive line may now market their failure potential versus QB health, to get their liability ping pong paddle.

Great write-up on this play. I went back to the play on my DVR and saw exactly what you were talking about. This is the sort of thing that I really like. Thanks a lot.

I havn’t heard the term “constrain” used in this way before but I think it’s critical that an offense keeps the defense honest.

It’s what I love about the Joe Gibbs redskins offense (mid 80s). There’s the big RB in the backfield threatening with the counter trey, with quality wideouts threatening long TDs if you start loading up to stop the run.

The success with a string of average quarterbacks shows just how that system can make a good offense look great.

great writeup. interesting play, though at the time it felt a little slow to develop. i do agree our problem on offense seems to be aggressiveness. the D always seems to know what is coming.

Very nice post. I am not a guru like you to be able to articulate my offensive ideas. This is what I was getting at with the pumpfakes and passes to the flat. This exact play, If McCoy pulls it down instead of tossing it skyward with a quick pumpfake, he buys Bell a yard or two. That’s exactly what I was getting at on the third down post. It looks as if he could have easily gotten 8 yards there maybe more. What do you think?

Sorry missed the reply to burntorangeandbrown. Either way, deep or shallow, I think like you said, the pumpfake freezes the D a bit. This makes me sooo long for our team to stay intact and continue to grow an entire repertoire.

Bell probably gets a good gain there, but Robiskie was open in the endzone. I like the read and the decision, it just needed to be a better throw.

Yep, that was the design…I was more referring to 3rd n mid to long from the “third down issues” post. Thats exactly how we get a 1st down or a touchdown (If we get the ball there). I am confident this kid can do it.

purposeful

light

Sorry, here I go not making sense again….damn Maker’s Mark n being off tomorrow. I was encouraged by Rufio’s interpretation’s of Daboll’s game plans as I looked back. I wanted to throw Daboll under the bus but I’m thinking…why not give it ALL another try.

Thanks for the detail and effort on the post Rufio. I really believe that our WRs are so pathetic that they cannot stretch the field, and they tie Daboll´s hands. They just cannot demand a cushion or respect. On Colt´s pick, there was a quick shot from behind MoMass that captured the play perfectly. MoMass did not run a route, did not run full speed or show any burst, and then did not fight the DB for the ball. It was really a piss poor play and effort.

Where did my comment on this go?

Anyway, I think you are way too harsh in saying Massaquoi gave “piss poor effort” and that our WRs are “pathetic”.

On the deep INTs, I think Massaquoi was running an option route where he is going to a skinny post, fade, or deep comeback depending on the coverage and the DB. He beat his guy deep and shouldn’t have been “bursting” because Colt had already thrown the ball and he would have been bursting right past the ball. He doesn’t “run a route” because that wasn’t what was called/what happened after the snap to give him his read.

He could have fought for the ball better, but a lot of the blame is on Colt there. You can’t just throw up jump balls when Ed Reed is the deep safety, your WR might win some of those but it’s a very risky throw especially considering the matchup. Colt’s throws needed to be more in front of Massaquoi and lower, thrown with more velocity. They weren’t.

Colt has to demand the cushion and respect as well. If they can press us and our WRs beat them deep but we can’t complete the passes, they will keep pressing us.

The WRs demanded a lot of respect vertically on the play I described above, all the vertical receivers drew double teams leaving Robiskie one on one with a LB and Robiskie torched him.

I know the ravens’ secondary outside of Ed Reed is really not all that good, and I know we could use AJ Green on this roster. But for this game at least, our WRs were getting good separation and Colt just didn’t make enough plays, he said so himself. He’ll learn from this and it’s just one game, but he didn’t have a very good game.

I was there and I have to concur. The receivers were getting open, especially Robiskie. The picks were bad throws by Colt where the ball seemed to float in the air forever and were thrown way too short. Though I do agree with Realmccoy that I’d like to see Chainsaw fighting harder than he did to break up the play.

No doubt, that is an area where Chainsaw can improve. He lets the ball get into his frame too often instead of plucking it out of the air. This habit can lead to not going up to “become the DB.”

I think sometimes Daboll ties his own hands just fine.

This is great! Thank you rufio.

As always, very nice writeup Rufio.

Wow. Excellent post. I’ve seen this sort of concept (constraint plays) done with formations and personnel packages, but I’ve never seen anyone break down how to use route stems to confuse the defense. Great work, rufio.

You must Login with your SB Nation account and be a member of Dawgs By Nature to post a comment.