Wed 23 Aug 2006
There has been much discussion this preseason about the Giants run-defense. Are they the unit that looked so porous against the Ravens in the first preseason game? Or are they the unit that held Larry Johnson, the best running back in the league, to 8 yards on 4 carries?
Letâ€™s examine the facts, starting with the fact that last year, the Giants had one of the best run defenses in the league. FootballOutsiders, using their advanced metric DVOA, ranked the Giants run-defense as second best in the league last year with a DVOA of -18.3%.
(Important Digression [Read Carefully]: I strongly recommend that all thinking football fans take the time to read FootballOutsidersâ€™ fleshed-out explanation of DVOA. But in this space, to make a long story short, the basic principle of DVOA â€“ which stands for Defensive-Adjusted Value Over Average â€” holds that all football stats should take into account situation and opponent, and therefore, looking at such primitive measures such as Yards is insufficient. For instance, a 1-yard run that yields a first down against a tough defense like the Bears is a more successful play than a 2-yard run on 3rd and 4 against a bad defense like the Texans. Without going into all the crazy math, DVOA breaks down each play of the NFL season in terms of “success points,” so that the 1-yard first-down run against the Bears is worth more than the 2-yard run on 3rd and 4 against the Texans.
DVOA, while imperfect and still the subject of constant tinkering, is the best statistical measurement there is right now. The number itself [in this case, -18.3%], stands for the percentage of success above league average. This doesnâ€™t correspond to yards or points; it is on its own scale. But to give you a sense of what the numbers mean, 30 means excellent, 0 means league average, and -30 means terrible. For defense, on the other hand, the more negative the number, the better [for intuitive reasons], so -30 is excellent, 0 is league average, and 30 is terrible. But do check out the full explanation. It also should be noted that that DVOA has a stronger correlation to a teamâ€™s record than yards, which is not surprising because it is calibrated to account for situation. And as a predictor, it has a higher correlative value than games won when it comes to predicting future wins. In other words, a 10-6 team with a DVOA of 30% is more likely to be better going forward than a 12-4 team with a DVOA of 20%.)
If youâ€™re into conventional statistics, the 2005 Giants run-D again holds up well, although not quite as spectacularly. They held their opponents to 103.5 rushing yards per game, good for 12th in the league, and 3.9 yards per carry, also 12th. That might not seem so great, but it places them in the upper 37.5% of the league, or better than 62.5% of the teams. Not bad for a supposed weakness.
It is somewhat understandable that people think the Giants run-defense is worse than it is. After all, itâ€™s hard to forget the painful image of the Panthers punishing us for 223 yards on the ground. But did that Giants defense bear any resemblance to the one that weâ€™re trotting out here this year? Not that I need to remind you, but do Kevin Lewis (who was wearing #44 for that game, for Christâ€™s sake!), a banged-up Nick Greisen, and Alonzo Jackson bear even the slightest resemblance to LaVar, Pierce, and Emmons/Short/Wilkerson? Youâ€™re going from one of the worst linebacking corps imaginable to one of the best.
(Donâ€™t believe me? Check out the FoxSports-FootballOutsiders positional rankings, which use some advanced statistical tools and some intuitive common sense to rank the Giants linebackers as the sixth best group in the league and significantly ahead of any of their NFC East rivals. Their defensive line, for that matter, ranked fifth, although that of course factors in the awesome pass-rushing abilities of Strahan, Osi and co.)
It wasnâ€™t long ago that the Giants run-defense was actually considered the strongest aspect of the team. Remember that mid-to-late-season stretch during which they held the Redskins to 38 rushing yards, the 49ers to 52, and the Vikings to 12? A little after that, before Pierce got hurt, the Giants even managed to contain the previously unconscious Shaun Alexander. Expect this yearâ€™s version of the Giants run defense to be more similar to this group than the depleted group that got trounced by Carolina.
But now, the Giants run-defense lays an egg against the Ravens in their first live contact, Colonel Tom calls them soft, and all of a sudden everyoneâ€™s sounding alarm bells. Despite the fact that the Giants run-defense was superb last year, especially when the starters were healthy, this year weâ€™re gonna have problems, the thinking went. In fact, you canâ€™t even compare last yearâ€™s run defense to this yearâ€™s, because this year weâ€™re going to have to make due withoutâ€¦. Kendrick Clancy.
Kendrick Clancy??? Losing him will single-handedly sink our run-defense???
Fellas, come on. Clancy was our best defensive tackle last year, no question â€“ he was by far the most productiveÂ member of theÂ four-man rotation of Clancy, William Joseph, Fred Robbins, and the also-departed Kenderick Allen. But doesnâ€™t the addition of LaVar Arrington, a physical enforcer at strongside linebacker, more than make up for the loss of Clancy? (And Allen, who nobody talks about.)
So the Giants run-defense should be a big strength this year, just as it was last year.
But I donâ€™t mean to imply that we should rest on our laurels, and that weâ€™re set at D-Tackle. I still think we should sign Brentson Buckner, a proven run-stuffer who will give us four starting caliber players at the D-Tackle spot â€“ Joseph, Robbins, and the impressive Barry Cofield being the others. Signing Buckner would not only make us four-deep at the position, but would also serve the purpose of lighting a fire under the asses of Joseph and Robbins, two guys who could definitely use a nudge.