Writers of Pro Football Prospectus 2008

Most Recent FO Features

RiceSid07.jpg

» Sidney Rice: What Could Have Been?

Sidney Rice has retired. Is he the most random single-season DYAR leader ever? One-year wonder? Injury prone? We offer a career retrospective for the second-best wide receiver named Rice in NFL history.

14 Oct 2010

Word of Muth: Base Weak

by Ben Muth

It's a good thing Jerry Jones is keeping the roof closed at the new Cowboys Stadium, because the sky is falling in Dallas. After dropping their most recent game, the Cowboys are just 1-3. It isn't looking too likely that America's team will play at home in this year's Super Bowl, as many predicted in the preseason.

Dallas didn't ever have a lead on Tennessee Sunday, and the offensive line didn't help the team's effort early in the game. The Cowboys made some adjustments up front and played much better in the second half. Unfortunately, they had dug themselves in too deep to come back. Now Cowboys fans have to hope that the same thing won't be true for their 2010 season.

The biggest news up front from this Sunday has to be Leonard Davis getting benched in favor of Montrae Holland. Anytime the offensive line unit struggles in a game, like the Cowboys' did in the first half, something has to change. So I was anxious to see whether or not Davis really played that poorly, or whether he was simply the scapegoat for a struggling group. Davis deserved to get the hook.

When I watched the Cowboys game against the Texans, I thought Davis played well, but it seemed like he could struggle against quicker defensive tackles. Turns out, Jason Jones was that quicker defensive tackle. Jones seemed to be playing in a different gear than Davis in that first half.

The two sacks certainly stand out as low points, but even in the running game, Jones was firing off the ball and into Davis before he could get his first step in the ground and establish any kind of power. Jones was just too much for Davis to handle.

The first sack Davis gave up was a predictor of his later struggles. It was third-and-10 on the Cowboys first drive of the game. Dallas was in a man protection we'll call Base Weak. (Most teams call this protection by a number. I've heard everything from 21 to 84, but since most teams have a different number, and this is a really common protection scheme, we'll just call it Base Weak.)

The general rule for most man-based protections against a 4-3 defense is that the offensive line will take the four down defensive lineman and the Mike linebacker. The tailback will take the either the Sam (strongside) or the Will (weakside) linebacker, depending on whether the protection scheme is Weak or Strong. Since this is Base Weak, Marion Barber is responsible for the Will, making the Sam linebacker hot (hot means that, if the Sam linebacker blitzes, one of the receivers, probably the tight end, has to break his route short so the quarterback can get him the ball before he's hit).

Figure 1: Base Weak Formation

With all the schematic stuff established, we can focus mostly on Davis. Jason Jones was lined up in a three technique (outside shoulder) over Davis. Tony Brown was lined up as a shaded nose (on the center's shoulder). This meant that Davis was all on his own, because the center had to deal with a shaded nose. The center would make a call (something that starts with an L -- Larry, Lasso, Louie, Liz, Laser, Lion, even Liberace if that's your thing... and yes, people still go the wrong way) to indicate that he was working with the left guard. That means the center (Andre Gurode) and left guard (Kyle Kosier) would double team the nose tackle, but also watch the Mike (middle linebacker).

Davis should've known before that he wasn't getting any help, at least until the Andre Gurode had checked the Mike and secured the nose tackle. Only after doing both those things, could Gurode offer any help to Davis.

Davis proceeded to get beaten on very quick club-rip to the inside. In fact, Davis had been beaten so cleanly that, by the time Gurode had made sure the Mike had dropped and the nose had been secured, all he could do was bump into Davis and send him to the ground. When Jones clubbed Davis, it was so quick that he got shoulder to shoulder with Davis in one step.

Once a defensive lineman gets even with an offensive lineman's inside shoulder, it's all over. After all, not only is the defensive lineman probably (and in this case definitely) faster already, but he's going forward, and the o-lineman is moving backward. It was the kind of sudden move that a coach would see and worry about how an aging offensive lineman would be able to keep up with.

Of course, Davis wasn't the only one struggle in the first half. His partner on the right side of the line, Marc Colombo, also had his troubles against the Titans. Just like Davis, Colombo also gave up two sacks in the first half. But Colombo wasn't benched. I think there are two main reasons that Colombo was allowed to stay in the game. First, he was more effective at run blocking than Davis was. And second, and probably more importantly, Alex Barron is his backup. (That's the last shot I take at Barron until he plays in another game. I need to stop relying on him as a crutch.) It's also important to note that I think Colombo's sacks were generally technique lapses as opposed to fast-twitch-muscle shortages.

The first sack was just a case of stepping in the bucket right as he punched. A lot of times when you're trying to punch with your hands and kick slide at the same time, you can step underneath yourself and shorten the hoop to the quarterback. That's what Colombo did on the first sack, and that's what got him beat.

The second sack was just the result of a really nice spin move by Jason Babin. Colombo was on his heels a little on the play, but sometimes you just get beat by a good move. After all, Jason Babin gets paid a lot of money to rush the passer. When he gets a tackle one on one, he needs to win occasionally or else he won't be in the league very long.

With all this going on, I'm sure it was clear to Jason Garrett and the rest of the offensive staff that they needed to make adjustments at halftime. That's exactly what they did. They didn't pull out the win, but they mounted a comeback and might have pulled it out if it weren't for a couple of turnovers. There were more adjustments than just benching Davis, which turned out to be pretty minor considering he returned when Holland was injured later in the game. The major adjustments had more to do with scheme than personnel.

The first thing the Cowboys changed that I saw was keeping the backs in to block longer. When the Cowboys would run their core protections, Base Strong/Weak or 2/3 Jet, the running back would stay in and clean up any defender leaking through the line rather than releasing if his linebacker dropped into coverage. This seems so simple, and it is, but you would be surprised how many offensive coordinators stubbornly refuse to lose their check-down routes in order to bolster protection. For some reason, they would rather risk giving up a sack in order to keep a safety net that rarely results in more than four yards.

When you consider that it's pretty rare for multiple offensive linemen to get beaten cleanly on a single play, having one extra blocker makes a big difference. The Cowboys decided it was necessary to have the extra help from the backfield, and the difference was noticeable. With hindsight being what it is (20-20), one wonders if the Cowboys could have made this adjustment early in the first half. All it really involves is telling your running backs to help anyone who needs it instead of releasing on short dump offs.

The other adjustment was more of a schematic overhaul. The Cowboys started going with a lot more full-slide protections in the second half than they have shown all year. A full-slide protection is great for offensive linemen because it gives them just one gap responsibility. If someone makes a great move inside, it doesn't matter because you have help there. There are three main reasons why teams don't run full slide protections more often.

1. If you do it too much, the defense can see it coming and overload the edge that the line is sliding away from.
2. You can end up with a skill player, like a tight end or running back, on a defensive end.
3. To make it schematically sound, you have to keep two skill players in to block rather than one.

Jason Garrett and the staff were willing to put up with those pitfalls because the scheme kept their franchise quarterback upright.

Figure 2: I-Formation Slide

There were two different kinds of full-slide protections that the Cowboys relied on in the second half. The first one was a quick play-action pass out of the I formation. The Cowboys offensive line would slide to its left, with each man responsible for the gap to his left. The Dallas running backs would run an Iso (short for Isolation) play fake. Third tight end Scott Chandler (playing fullback) would then throw a cut block at the defensive end to get his hands down in case Tony Romo needed to throw quickly. Marion Barber or Felix Jones would be responsible for any remaining rusher coming off the right side.

It's a great protection when used sparingly, as the Cowboys demonstrated in the second half. But if they continue to run it, defenses will prepare for it, and they'll start finding ways to put really good pass rushers on the tailback. That is not a good matchup for the Cowboys.

The second run-slide protection, which the Cowboys used more often, was for longer developing pass plays. Dallas would usually run it out of a single-back formation. As with all other slide protections, the entire line would slide one way (let's say left since that was called far more often on Sunday).

Figure 3: Single-Back Slide

The tight end, Jason Witten or Martellus Bennett, would be responsible for the defensive end by himself. This is a better matchup than you would think, as long as your tight end can block -- and Witten can. Defensive ends spend so much time working on the timing of offensive tackles. When a tight end blocks them, they can lose their pass-rush rhythm. The running back is then responsible for any blitzers coming off the right side. These protections aren't revolutionary by any means, but they are a great change up, and a nice way to help a struggling offensive line.

There was one last adjustment that the Cowboys made in the second half -- they took the training wheels off of Doug Free. The first couple of times I've seen Dallas play, it seemed that they were protecting Free. There a lot of ways you can protect an offensive tackle in the passing game, and the Cowboys seemed to use them all. They would almost always slide to him, they would keep a tight end in, and they would chip. Rarely would Free be on an island. But in the second half of this game, the Cowboys gave some of the protection to Colombo and left Free on his own. I think Free really stepped up.

It's unfortunate that his one major mistake, a missed cut block that led to a pick, wound up being so costly. Other than the miscue, I thought Free was great all game, but especially in the second half when his team really needed him. Combine that with his solid run blocking, and it's very possible that the Cowboys have found themselves a legitimate NFL left tackle.

Posted by: Ben Muth on 14 Oct 2010

43 comments, Last at 16 Oct 2010, 12:12pm by 57_Varieties

Comments

1
by AnonymousA (not verified) :: Thu, 10/14/2010 - 11:16am

As good as always.

3
by Theo :: Thu, 10/14/2010 - 11:33am

Always a good read.
But I'm not familiar with the cowboys offensive line and had to figure out the position Leonard Davis plays (right guard).
Also, I find it funny that the QB in your diagrams takes 11 yard shotgun snaps and linebackers play 10 yards off.

Ben, what do you mean with "With hindsight being what it is (20-20),... "

6
by Andrew Potter :: Thu, 10/14/2010 - 11:52am

20/20 vision is considered ideal vision. It refers to eye tests, and literally means that from a distance of 20 feet, you see what you ought to be able to see at that distance (meaning you aren't visually impaired).

The phrase "hindsight is 20/20 vision" roughly means that when looking back at an event, everybody has good vision ie. can see clearly even if the picture wasn't clear at the time. It's usually used idiomatically to say that it's easy to draw conclusions after an event that weren't obvious during it. It's expressed as a caution against hindsight bias.

Ben's saying that looking back, the change in protections seems so obvious that you wonder why it wasn't made earlier. That's easy to say after the event (hindsight giving a clear picture of events), but it may not have been so obvious at the time.

15
by Peebens (not verified) :: Thu, 10/14/2010 - 1:48pm

Or if you're Walt Harris "hindisight is 50/50"

2
by chemical burn :: Thu, 10/14/2010 - 11:32am

My main complaint about these columns thus far, is that you have a tendency to write "other than X devastating mistake, they really played well!" I think you've written some variation on it every week. At a certain point, aren't those very devastating mistakes the actual measure of a bad game?

7
by ABW (not verified) :: Thu, 10/14/2010 - 11:55am

It would be fun to read about an offensive line that just dominates all day long, and how they did it.

11
by chemical burn :: Thu, 10/14/2010 - 12:57pm

I don't even mean that - I mean, he keeps labeling things "good" that don't deserve that label. And as has been pointed out, other notable commentators have labeled "bad" or "unacceptable." I'm interested in reading about the line play, but I'm dubious of the grading on a curve that seems to pop up constantly...

21
by Ben Muth :: Thu, 10/14/2010 - 4:28pm

I can definately see that criticism. After all, Doug Free's missed cut block turned out to be the biggest play in the game. The thing is it's real easy to point out mistakes by offensive linemen, and that's what most offensive line analysis comes down to. Like on most sacks it's pretty clear that the offensive lineman screwed up and most commentators are quick to jump on them for it. I want to point out some positives and explain how someone could get beat for a sack and still have good day. All that being said I probably do need to weight the big mistakes more, even though Free played good and had less total mistakes than Colombo, Free's mistake had a bigger negative impact. I'll try to work on that.

22
by Dean :: Thu, 10/14/2010 - 4:42pm

The general perception seems to be to me that if an OL makes an acceptable or better block on every play, except for one where he gets beaten for a sack, that OL had a bad day. The perception seems to be, "it's not fair, but that's how it is."

One of the things that makes FO great is that they are willing to debunk conventional wisdom (or occasionally even verify it).

I realize that I've made some pretty broad generalizations above, but do you think that it's realistic to say that the guy described above had a bad game?

24
by BlueStarDude :: Thu, 10/14/2010 - 6:42pm

Ben, It's awesome that you're so open to improving your already great column, but I don't know that you need to change what you're doing that much. It fits right in with the FO ethos. It's not helpful to judge an offensive linemen on the consequences of his "failed" blocks.

If the pass that Ball tips because of Free's failed block doesn't get intercepted it doesn't change the quality of the block or, more importantly, the predictive value of knowing that he fails in his pass pro on, say, one out of every thirty dropbacks in contrast to another LT who fails on one out of every forty dropbacks or another who fails on one out of every twenty.

The consequence of one missed block may have a major impact on the game, but it's irrelevant to the overall quality of the lineman's play.

29
by Will Allen :: Fri, 10/15/2010 - 2:00am

Life just ain't fair for the offensive line. If just one offensive lineman blows his assignment on each pass, that offense is in for a miserable day, and a likely loss. In contrast, if just one defensive lineman succeeds in executing his assignment on each pass, that defense is likely in for a decent to good day. The terms "good" or "bad" thus become very relative.

31
by BlueStarDude :: Fri, 10/15/2010 - 10:46am

Even worse, a team can be setback greatly by poor TE and FB/RB blocking but all that will be talked about will be how bad the offensive line is. A great example was the Giants-Patriots Superbowl: both lines had their share of mistakes, the real difference was in how much better New York's TEs and backs blocked in both the run and pass games that day.

38
by tuluse :: Fri, 10/15/2010 - 1:31pm

Tuck was destroying the Pat's interior line too. So Brady couldn't step up, otherwise his tackles and tight-ends might have looked a lot better on the edge. While Eli's guards and centers performed much better.

34
by Pat (filler) (not verified) :: Fri, 10/15/2010 - 11:42am

It's more than that, too - if Free's failed block happens when they're up 14 points, people say "well, he was okay except for..." because it didn't cost them the game.

There needs to be a general term for this: something like "leverage bias." Every sport has this - people see high-leverage failures and put their importance much higher than the exact same failures in low-leverage situations.

That being said - it is important to make sure that you're not giving too much of a pass to guys, because it is a team game and you can find other people to blame on most plays. Heck, if nothing else, you can usually blame the coaches. But, in this case, analyzing Free, what you really need to ask is "does an average NFL left tackle have a mistake like that once a game?" My guess is "yes."

40
by chemical burn :: Fri, 10/15/2010 - 4:26pm

Hey - awesome, thanks for the response. Really, I hope I didn't sound too harsh! I just feel like as much as it is conventional wisdom for o-lineman's entire performance to get dismissed after a single game-changing mistake, there really is something to the fact that line-play is MORE game-changing than conventional wisdom even understands. That is to say, one badly blown block at the wrong time matters almost as much as a critical interception or a fumble deep in the opponent's territory... Anyhoo, great stuff, hope I don't come across as a swine with pearls in front of him...

4
by chemical burn :: Thu, 10/14/2010 - 11:34am

My other complaint is that since you picked 2 NFC East teams, I have to read about divisional rivals all the time...

5
by andrew :: Thu, 10/14/2010 - 11:44am

Will be interesting to see how these adjustments play out vs a vikings defensive line that absolutely destroyed the dallas o-line in the playoffs last year.

Jared Allen hasn't been the same since he cut his mullet, but he has to be hoping for more production vs. this line than he got vs the Jets (Ferguson), Dolphins (Long), or Saints...

9
by The Powers That Be :: Thu, 10/14/2010 - 12:34pm

The Vikings primarily destroyed the right side of the line in that game, and primarily Colombo, who was clearly not fully recovered from injury. Free couldn't replace Colombo because he had to play LT when Flozell went down. Colombo's been better so far this year; next week will be a big test. But I (along with most Cowboys fans) am primarily concerned about Bigg.

8
by iwatt :: Thu, 10/14/2010 - 11:58am

Simply a great and clarifying read.

10
by Aaron Schatz :: Thu, 10/14/2010 - 12:41pm

Fixed above to note that fullback was tight end Scott Chandler, not Dan Gronkowski.

12
by Little D (not verified) :: Thu, 10/14/2010 - 1:15pm

I'm not trying to be a jerk, but his name is Chris Gronkowski.

25
by The Ninjalectual :: Thu, 10/14/2010 - 6:49pm

How many Gronkowskis are there????

26
by Andrew Potter :: Thu, 10/14/2010 - 6:56pm

Three in the NFL - Chris (Dallas), Dan (Denver), and Rob (New England).

30
by nat :: Fri, 10/15/2010 - 8:38am

Gronkowski might be a good topic for a Cover-3 article.

42
by AudacityOfHoops :: Fri, 10/15/2010 - 7:46pm

And to save everyone else the 5 seconds it takes to google it - yes, they are brothers.

13
by Karl Cuba :: Thu, 10/14/2010 - 1:37pm

Love these articles but what does 'stepping in the bucket' mean? Is it like the slap-stick routine when someone steps in a bucket and if it is how does this translate to an NFL lineman?

17
by Eddo :: Thu, 10/14/2010 - 1:57pm

In baseball, the term "stepping into the bucket" means a batter striding towards the closest base (third for righties, first for lefties) instead of directly forward when swinging. It's generally considered bad form. I imagine it means something similar, here.

14
by Little D (not verified) :: Thu, 10/14/2010 - 1:38pm

Thanks for the great stuff, Ben.

I noticed Davis had the same issues last year. A couple of games that come to mind immediately were against the Falcons and Seahawks. He struggled against Jonathan Babineaux and Jamaal Anderson (when he was lined up at defensive tackle), and against Brandon Mebane in the other game. Teams had success slanting their linemen, and Davis along with Flozell Adams seemed to have the biggest issues.

Davis also doesn't strike me as a particularly smart player, although I'm not sure if this really applies to the Titans game. He seems late to recognize stunts, and with those solid stone feet of his it's a certainty that he'll be late to react anyway. Teams had a lot of success blitzing the 'A' gap between him and Gurode, although I haven't seen it nearly as much through the first four games. To be fair, Gurode isn't the sharpest knife in the drawer either, so not all of those free blitzers ripping Tony's head off are necessarily Davis' fault.

16
by mBoutte (not verified) :: Thu, 10/14/2010 - 1:49pm

...Davis is the guy to go after with an undertackle...he continues to lose quickness and is becoming a larger liability to the offense as a whole...teams can isolate him by scheme...he should be fighting for his job next year...

18
by starzero :: Thu, 10/14/2010 - 2:20pm

this column serves its purpose: it vastly improves my appreciation for offensive line play.

--
hail damage

19
by Drew DC (not verified) :: Thu, 10/14/2010 - 2:34pm

Word of Muth has quickly become a contender for this season's MVC (Most Valuable Column) award. I've never seen offensive line play explained so cogently -- this is the kind of close analysis that TV just cannot do justice to. Muth's writing is good enough to stand on its own, but the diagrams have been a definite plus the past few weeks.

On a technical, nitpicky note, you would save bandwidth at no loss of quality by storing the diagrams in PNG format instead of JPEG. PNG is typically better suited than JPEG for images that have blocky graphics and simple color schemes. For instance, by converting Figure 3 to a 256-color PNG (at the maximum compression setting), I reduced the image's file size by nearly one-third. (Hard-core website optimizers use utilities such as PNGOUT to squeeze those PNGs down even further, but that's probably overkill for this case.)

27
by tuluse :: Thu, 10/14/2010 - 7:17pm

A GIF should get close to that compression too.

20
by dpease :: Thu, 10/14/2010 - 3:16pm

That's great that Jason Garrett figured out what adjustments to make at the halftime break, but hasn't it been obvious for at least two seasons that the pass protection problem for Dallas was in the hole between center and right guard? How about starting with a scheme that protects that and then make an adjustment when the defense finally comes up with a counter? Dallas should be playing with a lead at that point instead of always playing from behind. Also, considering Tennessee was rotating in a fresh d-line often, Dallas should have attempted some sort of no-huddle offense.

I get that Leonard Davis is a problem, especially one on one, but I've seen too many games this year where Marion Barber has basically made it 11 vs 10. I won't guess as to whether he's confused, not looking for contact, or more interested in making a play with the ball, but I've seen him run the table for allowing sacks and pressure - letting a pass rusher through, making a weak attempt at blocking a pass rusher, going out into a pointless route when the rush is closing in. Other teams figure out max protection with nothing close to the talent Dallas has at wide receiver.

23
by speedegg :: Thu, 10/14/2010 - 5:34pm

I second the nomination for MCV (Most Valuable Column), this is great stuff!

28
by franky (not verified) :: Fri, 10/15/2010 - 1:14am

Moooooth ..... awesome read. Also to plug I'm Stanford '83. Yes, I sat in the sunny side watching Elway play back in the day. Glad to read your awesome insights, good to see your intelligent and thoughtful analysis didn't get dinged playing for the awe-inspiring Buddy Teevens and Wlat. Please keep it up and Go Card!

32
by kin (not verified) :: Fri, 10/15/2010 - 11:21am

great article!

33
by JoRo :: Fri, 10/15/2010 - 11:38am

As others already said, good job Ben, glad to have some insight into O-Line play, easily the most ignored aspect of the game overall.

35
by qed :: Fri, 10/15/2010 - 12:14pm

I'd really like to see how some of the protection schemes change against different defensive looks, i.e. how do the line responsibilities change in a Base Weak set if the SS threatens to blitz, or if two DBs line up in the A gaps.

36
by Matt Bowyer :: Fri, 10/15/2010 - 12:27pm

I wish this column existed during the Chiefs' offensive dominance from 2001-2003. I loved watching Will Shields out in front of Priest Holmes laying waste to defenders. I'd love to know more about what made them so effective, beyond having two Hall of Fame talents on the offensive line.

41
by chemical burn :: Fri, 10/15/2010 - 4:29pm

Oh man, second that. That was one team where it was more fun to watch the o-line than the skill position players!

37
by cardroo :: Fri, 10/15/2010 - 12:33pm

Great column again. Would love to get your thoughts/analysis on two items:

Watching the game, the commentators kept talking about all the spin moves being a big reason that the Titans were able to get to Romo. It would be great to get a breakdown on how, as an offensive lineman, you are supposed to deal with the various moves of a pass rusher (spin moves, bull rush, rip moves, speed moves, etc.).

Second, from watching the Cowboys/Titans game this week and the Bears/Giants game from earlier, it seems that defenses can get on a roll with sacks and they come in bunches. Why does this happen? Does momentum play that much of a role or is it that it takes time to adjust a protection scheme to deal with what the defense is doing? What is it that allows a team to put up 6-8 sacks in such a short period of time?

Looking forward to your next column.

39
by nat :: Fri, 10/15/2010 - 2:36pm

A related idea would be to focus on the opposing defense for one of these lines for a game - as a change of pace. Same kind of article, but instead of "what the OL did, what worked, what didn't, and how they adjusted" do the same thing from a defensive perspective. I'd love to get an offensive lineman's take on what the defense was trying to do, and what was likely to give the OL problems and why.

43
by 57_Varieties (not verified) :: Sat, 10/16/2010 - 12:12pm

Nice work, I enjoyed it. Would love to see something similar about the Steelers o line. One thought: could you remind us who's playing what position, either in the text or diagram?