Last night's tilt will leave us with two indelible images: DeAngelo Hall sprinting towards the end zone with no time left on the first half clock and the final play, in which Alex Barron is draped on Brian Orakpo like a drunken bride during a money dance.
It's the latter of these images that I'm betting haunted the Cowboys' offensive staff in the week leading up to the game. The nightmare we'll all remember is precisely the nightmare that Jason Garrett and his staff were game-planning to avoid. We all knew, and most of us were nervous about, the fact that the Cowboys were going into the game with two backup offensive linemen, each with a game, the story went, featuring a specific set of weaknesses: Montrae Holland has lead feet and can't handle quickness; Alex Barron has "left tackle feet," but is a ballerina at the point of attack. As it turns out, the Cowboys' staff weren't too concerned about Holland; they did very little out of the ordinary in this game to help him out. Barron, however, was a different story. A review of the tape shows a gameplan designed almost wholly to help Barron--and, by extension, to keep Tony Romo alive to fight another day.
As they do every week during the regular season, Garrett and his staff dipped into the Cowboys' massive playbook and extracted a collection of plays. A great majority of these were selected a) to help Barron or b) because they didn't require him to make a key block at the point of attack or c) they developed too quickly for him to impact them negatively. Lets take a brief look at some of these strategies after the break.
Formations: For the better part of the first three quarters, the Cowboys stayed out of more wide-open, obvious passing formations (three wide, empty backfireld, shotgun). As they usually do, they employed a lot of two-tight end sets. With a couple of exceptions, they only went to more open formations, specifically shotgun, on third and long situations. And when they did, they lined up with Witten and either Choice or Barber on either side of Romo. On a couple of occasions, Witten went out into a pattern; on several, however, he stayed in as a pass blocker, often on Barron's side. On multiple occasions, the Cowboys lined up in a "12" set (one RB and two TEs), with both Witten and Bennett lined up outside of Barron. Often, Bennett stayed in to help out the RT; at the very least, their presence forced oncoming rushers to take wide angles, thus helping Barron - who is often either too fast or too slow off of the snap - reach them in time. On a couple of occasions, including the first play of the game, they went to an unbalanced line, with Barron outside of LT Doug Free, and Bennett outside of RG Leonard Davis. What this did, in effect, was make Barron the tight end and Bennett the RT. Think about this for a second: on some level they trust Bennett more than Barron as a blocking tackle.
Short Drops: Until the score forced the Cowboys out of their game plan, the vast majority of Romo's dropbacks were of the one- or three-step variety. As a result, there were a lot of slants, short hitches and WR screens called. Garrett called plays requiring five-step drops only often enough to keep the defense from coming so far up as to render Dallas' game plan impossible (and it worked, as the 'Skins had to be nervous about the Cowboys' deep speed). As the game announcers made clear, this plan was designed to avoid pressure on Romo: no outside rusher, even if unblocked, can get to a QB taking a one-step drop. What IS required, is that the interior of the line remain stout. In calling these kinds of plays, the Cowboys' brass showed faith in Holland's ability to anchor and keep the front of the pocket clean. The conclusion to be drawn, then, is that they were most concerned with Barron's ability (or inability) to keep Orakpo and Andre Carter off of Romo.
Run Left: The Cowboys' running game was painfully one-sided; they ran predominately left all night, away from Barron's side. This makes sense, given that Barron is notoriously weak at the point of attack. When they did run right, which happened exactly four times in the game, they usually gave him help; on one occasion, on their fourth drive of the first half, Felix Jones took a handoff and started to head off RG, but a blitzing linebacker filled the hole. He might have taken it outside, but Barron's man was getting penetration upfield, so number 28 made like Barry Sanders, stopping on a dime, then cutting left for a short gain. From that point on, they gave Barron help. After the 'Boys had clearly established that they were a left-handed running team, they had to break tendency. On the short second-half touchdown drive, they ran two consecutive plays over RT, one of which was a new package that I hope to see more in future: a handoff to Choice, who then became an option quarterback. On both plays, #82 was positioned outside RT and helped Barron seal the edge; on neither play did Barron seem to get any upfield push.
As I hope this admittedly sketchy analysis makes clear, the offensive game plan was designed primarily to help out Barron. What is most disappointing is that he is supposed to be a good "foot athlete." Yet when, by necessity, Dallas opened up the offense at the end of the first half and then again in the fourth quarter, he responded by getting beat badly on the pass rush and committing key holding penalties. In the past 12 hours, Garrett has been the target of a great deal of Cowboy fandom's collective anger, especially for his conservative game plan. But had he developed a more wide-open plan, I'd wager, we would have seen even more of Barron getting beat. The result? More holding penalties, or more pressure on Romo - and more big plays. We all remember the key turnovers from the Denver and Green Bay games that were caused by pressure. So does Garrett. It was clear that keeping Orakpo and Carter (and various defensive backs; Redskins DC Jim Haslett dialed up a smorgasboard of blitzes) off of Romo was priority one so, given his o-line personnel, Garrett had no choice but to batten down the hatches and play smallball.
Romo emerged relatively unscathed; in that respect, the game plan was a success. Here's the problem with smallball, however: it requires extended drives of 10+ plays. As O.C.C.'s highly illuminating three-part series on "drive killers" will attest (part 2 is here; part 3 here), Dallas' is a big play offense largely because it has to be. The simple fact is that this is a mistake-prone bunch, even with Flozell no longer in the fold. Smallball requires perfect, consistent execution. If a team is going to go down the field in 5-7 yard chunks, a penalty or a sack becomes an almost certain drive-killer. A look at last night's drive charts substantiates this. On one hand, the Cowboys were impressive: they sustained drives of 7, 10, 7, 13, 5 (end of first half), 5, 6 (touchdown), 10 and 14 plays. The downside to long drives, however, is that they increase the opportunity for mistakes.
Indeed, drive-killers are exactly what plagued the 'Boys last night. In the first half, a promising second drive was curtailed by a holding call on Dez Bryant and then capped off by a missed FG; the next series was derailed by a sack of Romo; the next time the Cowboys got the ball--in the shadow or their end zone--they amassed four first downs, the last of which, a 21-yard pass to Austin, placed them at the Washington 34. A delay of game penalty put them in a first and long from which they couldn't recover. In the second half, a catchable pass, which would have put the Cowboys deep in enemy territory, clanked off Roy WIlliams' hands. After their TD drive, once they had opened up the offense, two drives of substance, sandwiched around a three-and-out, were killed by Barron holding calls, including on the game's final play. It's no accident that the only Dallas TD drive of the night was also one of their shortest: six plays and 34 yards. They scored largely because the short field didn't present them as many opportunities for a drive killer.
So, if you are Jason Garrett, which poison do you drink? Do you throw caution to the wind and risk losing your franchise QB for the duration, or do you protect him, playing smallball, knowing that your squad doesn't have the intestinal fortitude to win that way? He clearly opted for the latter, and it didn't work. But before you pillory him, consider what might have happened had he opened up the playbook: more holding penalties, more second- and third-and-longs; higher potential for turnovers; a FedEx feeding frenzy; and a potential blowout. All I'm sayin' is, it could have been worse.
And, Marc Columbo, wherever you are: pleeeease come back to your '09 pre-injury form. The season's riding on it, buddy.