It's a bit of a programming change for Yours Truly, it being OU Week and all. The only Keys I care about this week are the Keys to Beating the Hell Out of OU, and I think we've all got a reasonable idea of how things have been going for our major units. If you're interested, here's the short version of the regular 5 Keys piece:
QB - Kick ass.
OL - More ups than downs.
TE/Fullback - Downs and ups in near-equal measure.
Defense - You know.
Special Teams - Fix the rightmost guy on the first kickoff wave and we're in business.
I didn't have the energy to turn that into 2000 words when I am - and, I figure, all of you are - a hell of a lot more interested in how we can beat OU, get to 5-1 and regain both our mojo and a degree of control over our conference destiny.
Of course, you have to look backwards before you can look forward with any sense of clarity, so I'm breaking down our toughest moments and best plays over the course of our last two games with an eye towards prescribing a game plan to mete out the punishment OU so richly deserves.
Defense seems to be, um, kinda at the forefront of everyone's thoughts right now, so they're first up.
Five Flavors of Run Defense
In our three non-conference games, we certainly had our travails in run defense. Some were tackling-related, and many were due to simple assignment confusion against the variety of option looks that we saw (from New Mexico and Ole Miss in particular).
In the last two games, we've seen the level of competition rise and our run D results take an even further nosedive. Since both the level of opponent and style of attack in those two games is much closer to what we'll see against OU (Landry Jones ain't running any triple option on us), that's where I want to focus this analysis. In breaking down all of our snaps against OSU and West Virginia, some real patterns started to emerge. And, as your eyeballs and instincts probably told you, stunting/gaming/general confusion was the culprit in many of our opponents' biggest runs.
Is the answer just to play it straight and let our athletes carry the day?
Yes and no.
As well as I can figure, Texas' defensive calls and approaches with the front seven come in five distinct flavors. Some are sour as hell, but some are sweet indeed. Let's take a look at each type of call, with a breakdown of an exemplary play for each that illustrates the kind of results we've been getting. I'll also list a few other examples of these flavors of play, and you can check them out at:
OSU - All Defensive Plays (Courtesy of AlphaHydro)
OK - here we go.
Playing It Straight
Despite what it may feel like at times, the Texas run defense has faced more runs from a straight approach than any other. What I'm defining as ‘straight' here is basically all of our front four defenders staying in an upfield lane - no stunting, slanting or taking excessively wide rushes. If each of our DL was connected to the guys next to him by a rubber band with about four yards' worth of elasticity, it wouldn't break on these plays based on where the call is sending guys. Sometimes we're blitzing linebackers or secondary players, but usually those guys are reading the play and then acting.
Sounds like a winner, right? The problem is that some of our guys' read-and-react times aren't quite what you'd hope for. The defensive line makes a number of stops at or behind the line while playing it straight , and guys are rarely getting completely washed out or dominated one-on-one. Our linebackers have had good moments as well, but there have been many plays where the reads are coming late and our second-level guys make themselves too easy to block. Even worse, they make a wrong read and run themselves into the wrong gap - I'm distinguishing reactions here from when their path is determined by a stunting/blitzing assignment.
All in all, I counted 32 plays against the run in the OSU and WVU games where straight ahead defined our DL action. In those plays, we surrendered a total of 200 yards for a 6.3 YPC average.
Let's take a look at a typical ‘play it straight' play from the WVU game - this one went for just about our average, and it exemplifies some of our linebacker reading issues.
With 11:30 left in the fourth quarter, Texas is working hard to keep West Virginia from re-taking the lead. First and ten from our 12 yard line finds us in our standard Nickel look that night against WVU's most familiar personnel package and alignment - 3 wides with a blocker in the backfield.
As the handoff is reaching its mesh point, our DL are pretty much rushing straight ahead. Edmond is starting to react up to the play, but the center is coming free on him. Thompson is a step slower on the backside, but it looks like the LT is free to get after him as well
A half step later, it looks like the back has a solid hole. The top DT is trying to squeeze down, but Edmond is being blocked completely out of his gap. Thompson is containing on the backside, but he'd be in trouble even if there was a cutback.
And now the back is through the hole - Edmond does a pretty good job of fighting through his block late, but he's unable to get enough on Buie to slow him down.
The play ends with Buie powering almost all the way through an Adrian Phillips tackle, and it's first and goal, Mountaineers.
This isn't a horrible play, but it isn't a good one - in fact, it's pretty exemplary of our ‘play it straight' results against the run. Our LBs catch blockers on the wrong side of the LOS all too often, and while Edmond has OK instincts and good physicality Kendall Thompson just isn't there yet in either department and DeMarco Cobbs is still on Oceanic Flight 815. This approach has yielded some good:
OSU 1Q, 1:03 - 0 yards: Play it straight, Edmond has an easy read and looks like a linebacker - drops Randle at the LOS.
OSU 3Q, 8:14 - (-4) yards: Play it straight, Whaley murders his man and nails the RB.
And some bad:
OSU 2Q, 7:46 - 26 yards: Play it straight, Santos reacts late and wrong-shoulders the block to leave a huge hole.
OSU 4Q, 9:52 - 20 yards: Play it straight with a 2-man blitz - they go off tackle and Adrian Phillips loses badly against the lead blocker.
WVU 4Q, 4:51 - 11 yards: Play it straight - LBs just sit still and catch blocks - Malcom nearly gets Buie at the LOS but it's up to M Thompson to clean up.
Ultimately, plain vanilla is so-so at best for us in the run game. What's the best way to spice things up?
A Different Slant
A slanting DL is hardly the most exotic thing in the world, but it can be an effective way to get penetration for your DL and mess up pulls and combo blocks for the OL. You can get caught helping out the O-line if there's something like an Inside Zone call on and the opposing linemen just shove you the way you want to go and the back cuts behind it, but a LB or two scraping behind this kind of action can help mitigate that risk.
I counted fourteen run plays where Texas was slanting - five against OSU, and nine against WVU. On those fourteen plays, we surrendered a total of ELEVEN yards. That's right, kids - that's a verrrrry spicy 0.8 YPA average.
Let's take a look at how one of these plays unfolded against WVU.
With 8:40 to go in the second quarter, we've got our 4-DL Dime package in against WVU's 4-wide look.
As the handoff nears its mesh point, three of our four DL are aggressively slanting towards the field side. Malcom Brown (the topmost DT) is working completely across the face of his man, while the other DT is piling into a double team and Okafor is working to get his outside shoulder free. Just as important, Steve Edmond is flying downhill towards the back side of this traffic jam.
It's decision time for Buie, and that mess doesn't look too inviting. He might have been able to bounce it outside of Okafor, but he's already cutting back - right at #33.
BANG. Edmond smacks him two yards deep in the backfield. The slant did two terrific things for Edmond. It tied up the blockers, and it gave him one clean lane and read that he could attack with his natural linebacking instincts - no dithering here. When Edmond can play in pure downhill mode he's still not perfect, but he's pretty damn good.
Check out some other sweet slanting action at:
OSU 2Q, 8:11 - 0 yards: All slant, Edmond has an easy read to drop Randle for no gain.
WVU 1Q, 4:43 - 1 yard: 3 DL slant, Wilson is wider at the bottom but Kendall Thompson is moving at the snap and fills (for him) aggressively. Vaccaro cleans up for a 1 yard gain.
WVU 3Q, 5:08 - 1 yard: DL slants together and Malcom Brown makes as good a play as you'll see a true freshman DL make.
It doesn't always work...
OSU 2Q, 14:53 - 6 yards: All slant on QB keep, Edmond lets himself get cut by the lead blocker.
WVU 4Q, 5:25 - 6 yards: DL slants - this time the double team catches Brown and drives him back into Edmond's pursuit while Jeffcoat gets caught inside.
...but it's created more negative plays than any other approach, and those are the two longest run plays we saw against OSU and WVU when slanting. I want to see a whooooole lot more of this, and given that we nearly doubled our slant output from OSU to WVU I'm hoping that Manny is of the same mindset.
Icy Hot Stuntaz
And now we come to the stunt - bête noir of Longhorn defensive observers. We primarily run tackle-end (T-E) stunts - occasionally we'll run a T-T stunt and sometimes we run two T-E stunts at the same time, but our most common call is a single T-E stunt. Now obviously stunting is not a bad thing in and of itself or it would have died out not too long after the first defense tried it. But one of the stunt's most pernicious side effects is its tendency to leave a big hole in the center of your DL - particularly if you don't have our other defensive linemen working towards the middle to close it. It's this effect that creates pure hell on under-experienced linebackers who can be left looking at a 5-yard wide chasm while trying to figure out where the hell to be. Of course, that chasm isn't empty - sometimes it's filled with one or even two offensive linemen who have been freed up to fly/lumber/waddle downfield and lock up the LBs, and sometimes there's a hard-charging backfield blocker to boot.
It's this effect that's killing the defense, and we're not just creating it with stunts - a number of times, we're sending our DTs on wide enough rushes to each side that it's actually creating a BIGGER hole than our stunts since there's not even the possibility of a looping DE closing things down. Remember those elastic bands between the DL I asked you to envision in the ‘Playing it Straight' section? When one of those bands breaks, we tend to be in trouble.
I counted nine total plays against OSU and WVU (fewer than I would have guessed) where we voluntarily created this kind of a gap in our interior run defense with no design for filling it. Those plays surrendered a total of 112 yards, for a ghastly 12.4 YPA average that probably made it seem like it was happening far more often.
Not that you probably need this point belabored for you, but let's take a closer look at why some of these plays have been such a defensive detriment.
Early in the first quarter we're aligned in our basic Nickel look, but we've got Dime personnel - that's Adrian Phillips backing the line along with Edmond. We've got six box defenders against five blockers, though Phillips is obviously A) not someone you really want taking on a guard in the hole and B) in pretty significant run/pass conflict at the snap since that's Tavon F. Austin on the inside of that trips formation to the bottom. It's early in the game and WVU's run-heavy M.O. hasn't really come to light yet, but that's the kind of situation where you'd like your DL to give your second-level guys some protection, yes?
In the immortal words of Lana Kane from Archer - Nooooope! We're (I guess) selling out to get after the pass - Jeffcoat and one DT are running a T/E stunt at the bottom that leaves their half of the formation wide open, and the DT at the top (Whaley) is taking a strong move to the outside as well. We've basically parted the sea by choice here - there's a four yard gap between our innermost DL's, with OL's walled in place to make it wider. Edmond is on the move to make sure the back can't take the thing cleanly around the end (which I'm assuming is his assignment since Okafor went hard into the B gap rather that containing on the outside). This would be a good time to have a big, physical player closing hard from the backside to get into that massive hole, yes? But right now that player is Phillips - big, physical play at the LOS is not exaaaactly his métier. What's gonna happen?
Lo and behold - the back is cutting upfield into a still-gaping hole! Phillips is no closer to the LOS than he was, and how he's got the guard on his ass to boot. Jeffcoat is closing as fast as he can and Whaley is trying to work back from the top, but neither is in position to win here.
At last, first contact with the back is made - a mere 8 yards past the line of scrimmage. This certainly wasn't WVU's most damaging run of the night, but it's very illustrative of the pressure Diaz is putting on his second-level guys when he gets game-happy with his DL. You're either asking a safety - not your run-support safety, by the way - to fill aggressively into that A gap despite being in initial run/pass conflict with Tavon Austin in the slot to his side, or you're asking the improbable for Jeffcoat to be cleaning up two gaps over on his stunt.
Sometimes we get away with this:
WVU 1Q, 4:03 - (-3) yards: T-T stunt leaves some gaps, but Malcom Brown flashes NFL athleticism and drops the back before it matters.
But usually not:
OSU 1Q, 7:00?? (no play clock visible) - 8 yards: DE goes wide, KThompson reads instead of filling and gets dumped.
OSU 4Q, 5:12 - 15 yards: Really wide rush by Jeffcoat in our Dime Prowler look - we don't fill that void fast enough and Vaccaro can't hold up against the lead block.
WVU 4Q, 15:00 - 23 yards: We have a stunt on as Austin is running a fly sweep - no one gets into the gap and he cuts it up for 23. You HAVE to have a motion call that calls off any stunts on fly sweep motion - just an absolute awful thing to have on against this kind of play.
Basically, shelve this shit until Edmond is a senior then see what it looks like in non-con play. And be ready to shelve it again.
Evel Knievel's Guide to Safer Stunts
On the heels of all that, what if I was to tell you that stunting is sometimes...not so bad? That hole in the interior of the defense is a major vulnerability, sure - but it's a lot worse if the LBs are standing flat-footed behind it waiting to passively receive a blocker. Fortunately, that's not all we've had them doing. On a number of plays where we've stunted or taken wide DE rushes, we've managed to pair this action with a second-level player aggressively blitzing the gap at the snap. It can be a bit of a judgment call to tell whether it's a designed blitz or simply a quick read/reaction, but once you've watched enough repetitions of our LB's tortured reaction processes it's pretty easy to tell when an instant downhill attack is by design (HINT - it's instant).
I counted thirteen total plays where a stunt or wide DT rush was paired with a called blitz into the gap. Back to the rubber band image again - on these plays, a second-level player was in the gap quick enough to catch both ends of the broken band before they flew completely apart. Those thirteen plays surrendered a total of...52 yards? For a more-than-acceptable 4.0 yards per carry average? Yes, indeed - Manny may stunt too much and may be guilty of under-instructing his LBs and using them as blitz monkeys, but when these two tactics are at least intelligently paired the results aren't too bad.
Here's an example of this approach working well:
With 6:45 left in the third quarter, it's our Nickel package facing off with WVU's 3-wide look yet again.
At the snap, we've got a stunt developing at the bottom between Jeffcoat and the DT, with the other DT taking a wide outside rush. This sort of thing has been baaaaad ju-ju for us, but check out the second level - that's Kendall Thompson right on the uppermost hashmark, blitzing hard into the gap.
A half second later, Buie is in trouble. The guard is walling off Jeffcoat, but the center who was watching that DT fly outside him was too slow to react to Thompson's blitz. With the F-back stepping out towards what was normally a trap block of the DE, there's no one to pick up a hard-charging Thompson.
That's what we like to see - a play in the backfield thanks to Thompson getting to play to his strengths and fly upfield. I'm not saying this to bag on an inexperienced sophomore, but I will promise you this - if Thompson had been simply reading this play, he'd be at the 49 yard line getting blocked by the center right now with Buie angling towards a quickly-developing hole between Okafor and the DT up top. But since we shot him aggressively into the gap in a smart pairing with our stunt action, the play went our way.
A few more examples of this approach working out:
OSU 3Q, 9:48 - (-1) yards: T-E stunt up top but Edmond blitzes into the middle gap and drops Randle for -1.
WVU 1Q, 14:53 - 3 yards: Double T-E stunts, but we have Edmond filling outside off the bat and Thompson coming into the middle at the snap - Des Jackson forces the cutback and it's cleaned up for three yards.
WVU 2Q, 13:58 - 1 yard: We split the DTs on their rush but Edmond blitzes the gap and trips up Buie for a short gain.
And a couple of times when it didn't:
2Q, 2:34 - 18 yards: Wide rush and we blitz to fill the gap, but Turner follows Edmond into the wrong gap and leaves a big hole
4Q, 2:53- 22 yards: We part the seas on the DT rush and Edmond blitzes into the hole, but the lead blocker gets him and Thompson is being blocked six yards up the field.
Manny has been nearly schizophrenic with his calls this season. Some of these calls act like (or seem to recognize) that our LBs are barely capable of making reads and reacting with the blend of decisiveness, speed and physicality that the position requires and therefore send them on blitzes of wildly variable soundness. Other calls blithely present them with five yard-wide holes and free blockers that would frequently defeat a Senior All-American. Which approach would YOU rather see on Saturday?
The Unsound and the Fury
A quick word about a few plays that didn't land in any of the above categories - we had four plays that stood out for sheer mindlessness and certainly involved at least one botch rather than the active intent of 11 players that their coordinator. They were:
- The 69-yard first quarter run by Randle when we followed motion and were left in an unsound front (5 players, 6 gaps) that wasn't addressed by either a 2-gap approach or a filling secondary player
- The 50-yard jaunt by Walsh when TWO men followed motion and left us in a similarly unsound front - this one was even worse since the back served as a lead blocker and didn't find anyone to hit until 20 yards downfield
- The partial-birth abortion against WVU that saw us blitz BOTH linebackers into the same backside gap, leaving Buie with the easiest playside run he'll ever see.
- The embarrassment of an 8-yard QB sneak in the 4th quarter when the defense wasn't even lined up - it was particularly galling that this play went off after the first-half tragedy where a sideline timeout robbed us of a sack and more or less swung the game.
I don't have any more desire to illustrate these plays than you do to relive them, but suffice to say they illustrate the importance of not baffling your own defense.
So, what does all this mean when it comes to beating the hell out of OU? I'll have some specific defensive play thoughts drawn up on Friday, but for right now I'd advance some simple principles:
#1 - Play the DL as an integrated unit to protect the linebackers AT ALL TIMES.
#2 - Exotic pressure packages are not the foundation of a defense - they are a luxury afforded to you by high-level play in the back end. We do not have that level of play, so we do not have that luxury.
#3 - With three or four of our high-caliber DL slanting aggressively into gaps at the same time, one or two of them is going to break through and wreck a play pretty damned often.
#4 - If you're going to break a rubber band, make damn sure someone is in the gap in time to catch both ends.
#5 - Spreading your DL with a lead blocker in the backfield is asking for trouble, since he can pick up a blitzer and still leave the back with a 2-way go in a big hole.
Regardless of specific personnel and alignment, on downs that aren't short yardage/goal line or 3rd and 10+ I'd like to see a mix of roughly:
20% Blitz-Supported Stunts
and that's all against Oklahoma.
What do you want to see?