The strangest thing about that stat is that the other four defenses on that list vary from good to average by DVOA: Chiefs are 11th, Panthers are 10th, Jaguars are 7th and Chargers are 16th. The Pats, meanwhile, are 32nd, and are as far from 31st and 31 is from 20th. We're a DVOA trainwreck.
As damning as that stat seems, YPC allowed on first down doesn't seem to correlate all that strongly with overall defensive effectiveness at this point in the season.
I wonder if you would take away the big busted plays (and resulting touchdowns) on third down how much our DVOA would improve. Just to make clear I am not saying we should just ignore that those happen but I am just curious how effected the DVOA is by things that I consider easier fixable (eg. communicaton) than others (eg. DL run issues).
If we can get close to 16-20 (ie. average) on defense when removing the big plays then I wouldnt get too concerned with all those advanced metrics unless you dont believe the coaching staff can fix the communication issues.
I believe the major problem with DVOA and the other Football Outsiders derived statistics is that they are proprietary and opaque. While you can get a general sense of what they are attempting to measure, it is hard to evaluate how well their metric captures it without being able to evaluate the nuts and bolts of that metric.
I found a Field Gulls blog post that attempts to describe DVOA, yet the final post contained many corrections and comments from football outsiders (FO). I think the most telling statement from FO is the final one:
Just what the heck is DVOA and why should I care?
Also, let's be honest, one of the issues with DVOA is that it is proprietary. I have to calculate it, nobody else has the equations and baselines. It's not the kind of thing that's easy to add up from a box score even if I made all that stuff public. I hope that helps clarify some of the stuff you wrote. Honestly, trying to make this stuff easy to understand is one of the hardest parts of my job. The fact that you didn't quite get everything right, and you're a hardcore FO follower, is just an example of how we still need to do more to make this stuff clear to everyone.
You can read FO's description of the stat here:
FOOTBALL OUTSIDERS: Innovative Statistics, Intelligent Analysis | Methods To Our Madness
Now, this is not meant to discredit the work that FO is doing here. I believe that they are actually attempting to create a summary statistic for defensive efficiency. I do not have a similar amount of trust in ESPN's QBR statistic. So DVOA might be capturing something, it is just hard to judge actually what it is. Furthermore, I feel like DVOA is not much of a conversation starter, given it's opacity. I prefer non-proprietary football statistics:
NFL Football Stats - NFL Team Opponent Red Zone Scoring Percentage (TD only) on TeamRankings.com
For example, consider
opponent red-zone touchdown percentage. The importance of this statistic is obvious. Preventing your opponent from scoring a TD in the red zone is a 4 point play at least, and sometimes a 7. It removes some of the "garbage time" or "game management yards" biases that exist when teams play softer coverages and allow the other team to move the ball slowly. While it might be in your interest allow the team to slowly move down the field while you are ahead, you rarely want to let them get the TD.
Yet the even better part about a statistic like
opponent red-zone touchdown percentage is you can immediately consider the context and limitations of that statistic. A team that keeps giving up 21+ yard touchdown plays will not be penalized on this stat. This stat will not provide insight as to a team's ability to force a 3 & out to give their offense a chance to win the game, or a team's ability to prevent them from getting in field goal range at the end of a half. It is a good measure for one facet of defense play, and does not attempt to do more than that.
