Skip to main content

Running the Numbers: Cowboys vs. Giants Film Study

When the Cowboys win, breaking down the film is usually a whole lot of fun. When the team loses, not so much. And when the 'Boys go down like they did yesterday, but they still run 83 offensive plays, well, that makes for a pretty long and agonizing film study session. Nonetheless, here are my thoughts on the Cowboys' 29-24 heartbreaker in Week 8:

  • In the season opener, the Giants defense disguised their intentions, – meaning they either blitzed after not showing it or showed a blitz and then sat back in coverage – on only 7.1 percent of the Cowboys' snaps. In Week 8, they did it on 14.5 percent of snaps, often showing a blitz but backing out. Dallas passed on all 14 plays, with Tony Romo getting sacked on one of them. He completed nine of the 13 passes he attempted for 71 yards (5.46 YPA), no touchdowns, and one interception. I think the Giants made a serious effort to try to confuse Romo with their alignments, and it appeared to work.
  • One of the plays on which the Giants may have baited Romo was his first interception. On a first-and-10 at their own 45-yard line, the Cowboys lined up with "12" personnel, which is one running back, two tight ends and two receivers. Jason Garrett called for a play-action pass. By my count, the Cowboys had run 20 play-action passes on the season up until that point, and nine of them (45 percent) resulted in the same post pattern to Dez Bryant. The Giants showed blitz on the play but backed out. I'm not sure whether or not it confused Romo, but either way, the Giants seemed to know what was coming and safety Stevie Brown jumped the post to Bryant for the interception.
  • For the record, I labeled that first interception as a play-action pass because Romo pretended to hand the ball off. If you recall, however, there wasn't a running back in the area, so it looked like a busted play. I think that Garrett called for a play-action pass but, because the Giants showed blitz, running back Phillip Tanner abandoned the run-fake action and jumped outside to pick up what he thought would be a free rusher. Thus, while not a play-action pass in the technical sense, I labeled it as such because that's what was called.
  • Dallas ran five play-action passes on the day. You may remember the Cowboys were thriving on play-action looks prior to Week 8, with Romo completing 16 of his 20 attempts for 250 yards and a touchdown. On Sunday, it was a different story. In addition to his first interception, Romo's second pick, the deep ball to Miles Austin, also came on a play-action pass. The quarterback's other three play-action looks were all on the Giants' 1-yard line. One resulted in a touchdown pass and the other, the bootleg, actually resulted in a touchdown run.
  • The Cowboys attempted one screen on the day, and it fell incomplete. On the season, Dallas has run only 10 screens, six to running backs and four to receivers. The 'Boys have totaled 33 yards on the screens to the backs and two total yards on the screens to Bryant and Kevin Ogletree. I predict we'll see more screens to the running backs in coming weeks, as the short passes can be used as an extension of the struggling running attack.
  • Coming into the game, the Cowboys had attempted 20 passes that traveled at least 20 yards in the air. Romo completed 10 of them for 253 yards, three touchdowns and one interception. On Sunday, Romo attempted 11 passes of 20 or more yards, mostly out of necessity. His first two interceptions to Bryant and Austin came on deep looks, but Romo also had completions of 53, 40, and 25 yards on deep passes. On the day, the quarterback totaled 10.72 YPA on deep passes, although the numbers are a bit deceiving. The Cowboys were two Bryant fingers away from a game-winning touchdown, and that pass traveled 46 yards from the line-of-scrimmage. Plus, a few of Romo's deep passes were low-percentage tosses late in the game that were basically thrown away.
  • While the risk of a sack or interception is admittedly greater on deep passes, the potential downside is probably worth it in most situations, particularly when Bryant sees single coverage. Yes, Bryant has struggled with drops this season, but he's simply too talented to ignore when defenses don't roll a safety over top of him. Romo's deep passes in Week 8 exemplified the high risk/high reward nature of the calls, but let's not forget this was just a single game. The Cowboys have excelled on deep passes for years, and such a large sample of data is superior to that from one game. Had Giants cornerback Corey Webster not made a great play on the deep interception intended for Austin, for example, we might be looking at the game and the calls for deep passes from a completely different angle.
  • It's a shame the majority of the focus will be on why the Cowboys came up empty as opposed to Jason Witten's record-breaking day. We all know about his 18 catches, but how about this for a stat: I tracked Witten as staying in to block on seven passes on the day, and on those plays, Romo was 1-for-7 for 23 yards and two interceptions. When Witten was in a route, Romo's YPA shot from to 3.29 to 7.53. I think there are times to keep Witten in, particularly when the Cowboys want to take shots downfield, but the tight end's Week 8 performance was really remarkable.
  • Garrett called 15 designed runs, but only three increased the points the offense could be expected to score on that respective drive. I think that lack of success is why we saw the offense pass the ball on three straight plays with 1 yard to go for a first down on their second-to-last drive. Still, even if Garrett didn't trust Felix Jones to hang onto the football, I think the offense could have benefited from a Phillip Tanner run in that situation. Since 2008, NFL teams have converted a first down or touchdown on 73.4 percent of their running plays that have come with 1 yard to go for a first down, compared to only 26.6 percent of their passes in the same situation.
This article has been reproduced in a new format and may be missing content or contain faulty links. Please use the Contact Us link in our site footer to report an issue.

Related Content