A couple of months ago, during a conversation with a college defensive coordinator, an idea picqued my interest. We were talking about stats and the general perception differences between coaches and fans/analysts, and he said something that stuck with me. While we spend most of our time talking about defensive alignments -- 4-3, 3-4, 4-2-5, etc. -- the first thing most coaches will look at when they begin film study on a given defense is different. They're not concerned as much about alignment as whether a given defense is playing zone or man coverage. So much of both defensive personality and offensive scouting is based on zone or man, and it's something we don't spend nearly as much time on.
To be more specific, we definitely don't spend a lot of time thinking about the statistical effect of playing zone or man defense.
Our conversation reminded me of one of my favorite aspects of Ken Pomeroy's work in college basketball: the Defensive Fingerprint.
Defensive Fingerprint attempts to objectively identify the style of a team’s defense. Inputs into the system are the departure from the D-1 norm of the following defensive characteristics…
- assist percentage (triple weight, higher means a more likely zone team)
- 3-point attempt percentage (triple weight, higher means a more likely zone team)
- free throw attempt percentage (double weight, higher means a more likely man team)
- turnover percentage (single weight, higher means a more likely man team)
- defensive rebounding percentage (variable weight depending on offensive rebounding percentage, higher means a more likely man team)
All of these things are thrown together producing a number which will forever remain a secret. The number is then used to book the defense into one of four categories in an attempt to describe what type of defense the team employs…
- Mostly man: The team probably plays man defense on the vast majority of its possessions.
- Inconclusive: The data does not support identifying defensive tendencies.
- Some zone: It’s likely that this team plays a significant number of possessions in a zone defense.
- Mostly zone: It’s likely that this team plays most of its possessions in a zone defense.
I've always loved the personality stats, so to speak -- the measures that tell us about tendencies more than quality. We need the quality measures, obviously, but whereas that's what most people think of as "stats," that's just scratching the surface of what we can learn about a team through numbers. Run-pass ratios are great for revealing offensive tendencies, and the further we get down the road of charting data (2013 Charting Project data is still being processed, by the way), we can start learning more about how given styles of offense utilize the run or the pass.
Defensively, though, we're still in the frontier stages. Take a look at what I use for my offseason previews.
The run-pass ratios are good for telling us what opponents felt was ripe for attack. "Bend-Don't-Break" is a look at a team's success rates compared to its explosiveness measures. "Need for Blitzes" compares a team's base pass rush (on standard downs) to its ability to bring the heat on passing downs. "Go After the Ball" is a general havoc stat that I'm tweaking for the 2014 previews. And Covariance tells us if a defense was more likely to play its best against good or bad opponents. That's ... something. But it only scratches the surface when it comes to real tendencies. What are defenses choosing to do?
While acknowledging that different zones have different levels of aggressiveness (as with basketball) and that certain defenses incorporate aspects of man and zone (as with basketball), let's say that we want to create a way for determining a defense's man or zone tendencies. What should we include?
I thought I had a pretty good starting point, but it was kind of a dead end. I thought that if you look at a team's PD-to-INC ratio -- the percentage of a defense's incomplete passes that were defensed (either intercepted or broken up), that would seem like a pretty good first step. Teams that play man defense might have the tendency to get more hands on the ball, perhaps with the expense of allowing a few more big plays. But pure PD-to-INC ratio doesn't really seem to work. For one thing, the year-to-year correlations here are rather weak -- the correlation between a team's 2012 and 2013 PD-to-INC ratios was 0.27. For another, one of the most notorious man defenses in football last year, Michigan State's, had one of the lowest averages.
2013 PD-to-INC ratio
1. Tulane (47%)
2. Tulsa (45%)
3. Vanderbilt (44%)
4. BYU (42%)
5. Ohio (41%)
6. Houston (41%)
7. Louisiana Tech (41%)
8. East Carolina (41%)
9. Alabama (39%)
10. Boise State (39%)
111. Michigan State (27%)
I still think there's something that can be gauged from looking at a team's ability to get its hands on passes (in a way that is independent of pure quality assessments), but I haven't quite figured it out yet.
Perhaps the bend-don't-break idea gets folded into this. A bend-don't-break defense is by stereotype likely to play its corners pretty far off of the ball, keep its players facing the line of scrimmage, and swarm to the ball. That describes a zone pretty well.
Perhaps the "Need for Blitzes" idea gets folded into this. A zone is often a safer defense in most hands, and offenses will often attempt to exploit it with curls and short "camp out in the holes of the zone" passes instead of waiting to see if a receiver can find a hole downfield with a corner route. Meanwhile, a man defense almost requires a solid pass rush to protect defenders downfield with their backs turned.
Anyway, just spitballing. We can obviously look at statistical tendencies of different types of defense ... but first, we need to identify defenses. Which ones are Mostly Man? Which are Mostly Zone? And for that I'll need a little bit of help. It's time to crowdsource a bit.
Here's a spreadsheet with each FBS team from 2013 on it. If you have knowledge of a defense's tendencies -- either because you're a fan of the team, a fan of its opponent, or just a knowledgeable fan of college football -- mark it down. (And if you have a link to support your finding/selection, please include it. Not required. I'm also including a Comment column if somebody disagrees with a selection.) If we come up with a relatively lengthy, accurate list, we can start to look at stats and make generalizations. This could be a pretty cool resource.