I thought that myself when I read that article that talked about a Data Scientist(tm)
The Punt and the Pendulum
In the aftermath of the frightening NIU game, I promised a guest post from Jon Chait about the defense and its decline. This was delayed by a series of performances appeared to be at least acceptable--17 points versus Notre Dame, 23 versus Wisconsin, 23 versus Minnesota--coupled with the spectacular implosion of the offense, which made criticism of the other side of the ball seem silly in comparison.
Recent events, though, reveal a pattern that do merit said guest post, and Jon has kindly provided it...
With the season finished, we can now see that the basic arc of the defense was cast at the end of 2004. Carr decided that the problem on defense was our propensity to give up big plays, and so we designed our defense to stop them. So our base defense became a three-deep zone.
For those of you who don't follow these things, a three-deep zone is just about the most conservative defense you can run. Most zone defenses have two deep safeties. The three-deep, obviously, has three. It makes it hard for the opposition to complete a long pass, and it assures that you have lots of containment to prevent long runs. If you're watching on TV, you can usually recognize it by seeing the strong safety move toward the line of scrimmage before the play, and the cornerbacks retreat.
The downside is that, compared with a two-deep zone, you have one-fewer defender against shorter passes and running plays. Few teams at the college level use a three-deep zone as a base defense. It's just too easy for the offense to stick to underneath patterns and play 11-on-8. It's especially weak on the perimeter, where the deep retreating corners allow outside running and short outside passes.
The defense we saw against Northern Illinois is the defense Carr and Herrmann planned to use throughout the season. It's a defense that basically let our opponent move the ball at will.
Now, the result against NIU was such a disaster that we had to use change-up defenses. You could see this when we were tied, and especially when we were behind. After the first drive against ND, we played far more aggressively, and enjoyed phenomenal success, holding a potent offense to just 10 points and 168 yards after the opening drive.
But the coaches kept it in their mind that the key was not to allow big plays, and they kept trying to run the three-deep zone. We used it mostly or exclusively on the first drive of almost every game, and almost always got torched. (Sometimes we were bailed out by opponent mistakes, like dropped passes, at MSU, or fumbles, at Northwestern.)
Only after it failed would we start mixing up our defensive calls. And, of course, when we got a lead at the end of the game we'd go back to soft zones.
Now, you might say that the defense improved over last year. I'm not so sure. It's true that our points per game yield was down. But a lot of this improvement came not because we got more stops, but because we made our opponents take longer to score, and thus limited our own offensive opportunities.
I haven't tried to gauge the entire season this way, but compare the 2004 and 2005 Ohio State games. In 2004, the defense allowed OSU to score 30 points (plus seven on a punt return.) In 2005, the defense gave up 25 points. But in 2004, the Michigan offense had 15 possessions, compared with only 10 in 2005 (of which one came with only 24 seconds left, and hardly counts as a possession.) The 2004 defense forced Ohio State to punt six times, the 2005 defense only twice. Is this an improvement? I don't think so.
That's why the fact that Carr talked after the OSU game about the fact that we gave up too many big plays is so discouraging. The obvious problem is that we gave up too many long drives, and had too few sacks, QB hurries, and three-and-outs. As long as big plays is the metric they use to gauge the defense, we're going to be mediocre.