Defensive scheming
Posted by Chase Stuart on Monday, November 9, 2009
I've got a theory that there aren't necessarily teams that can only play the run well or can only defend the pass; rather there are just good defenses and bad defenses (and everything in between). Certainly there are teams that have played well against just the run and not the pass, and vice versa; there's no denying that looking backwards, certain teams in certain years were great in one area of defense and terrible in the other. But looking backwards, there are teams that recover lots of fumbles and teams that faced opposing kickers who missed a high number of field goal attempts, but that doesn't mean that those things will happen again in the future.
My claim is that theoretically -- i.e., if we played 1,000,000 games so that randomness would be eliminated -- defenses should just be defenses. No good run-D, bad pass-D teams; no average run-D, excellent pass-D teams; just defenses. That (I think) is a bold claim, so I better have some good reasons behind it. What are they?
1) Defenses are like chains: they're only as strong as their weakest links (for the flip side to this argument, see Brian Burke's article that offenses are like chains). Picture an unbelievable run-D teamed with an awful pass defense. That defense isn't going to be very good, as almost every team in the league could pass on them all day long. Flip the script, and nearly every team could control the game with power football against a defense that can't stop the run. On defense, if you have a weakness, almost every opponent can exploit it.
2) What's that mean? Defensive coordinators should tailor their defensive strategies to cover their weaknesses. When we think of a strong run defense, we think of a beefy defensive line and some hard hitting, agile linebackers. Pretend you're a defensive coordinator with what appears to be the elements of a team with a great run defense and a terrible pass defense. You've got two big defensive tackles, two run stopping DEs, a run-stopping set of linebackers, a hard hitting strong safety that's a liability in coverage, and some weak cover corners. What do you do?
Here's what I would do. I'd say my guys can stop anyone in the running game and we'd still allow 42 points because our pass defense is so bad. So I'll tell my OLBs to either blitz or drop into coverage on most plays; I'll play a nickel defense pretty frequently, since I still should be able to stop the run with that package. I'll never put a safety in the box; if I want to have my SS stop the run, since that's what he's good at, I'll be sure to put him close to the line of scrimmage but I'll also take out a linebacker for another corner. I'll blitz the passer frequently to protect my weak corners, but I'll counter that by occasionally dropping my DEs into coverage since they're not going to get to the QB, anyway. I will do as much as I can to cover for my bad pass defense. On some plays I'll have just a two-man rush and have nine guys playing the pass. Most likely, I'll field six defensive backs on more plays than any team in the league, especially since my SS is mostly just another linebacker, anyway.
Do you think I've gone too far? Let's say that even though based on talent and ability I have the #1 run-D and the #32 pass-D in the league, my defense's statistics (measured in terms of effectiveness, not gross yards) say we have the #10 pass D and the #20 run D? If that ends up happening, I'll just creep back a bit. I'll stop dropping my DEs into coverage. I'll make sure I don't take out my slow but powerful MLB too often. I'll play more nickel and less dime. In other words, I'll keep tweaking my defensive lineup and scheme until I achieve balance.
The same is true if you have -- based on talent -- the best pass D and the worst run D. Let's say that you've got great cover corners, two terrific edge rushers, former college safeties at linebacker, a converted free safety at strong safety, and two agile interior linemen. You're worried, though, that your run D is going to get murdered and you're going to lose every game. So what do you do? You play 8 -- or maybe 9 -- in the box. You tell your edge rushers to hesitate for a second before selling out on the pass. You tell your corners to keep an eye on the running game. You don't go to the nickel even if your opponent brings a third receiver onto the field. And so on.
Just like with the stud run-D roster, with a stud pass-D lineup, you can keep tweaking your makeup until you achieve perfect balance.
This theory isn't perfect -- there are some minimum thresholds of ability a player must have. Dwight Freeney's never going to be able to play the run as well as other ends. A 35-year old John Lynch could only do so much in the passing game. Against the Browns, you really don't need to worry about the pass beating you. But generally, or as long as you and your opponent are not too elite/poor I think, teams can -- and will -- do whatever they can on defense to ensure perfect balance.
There are definitely exceptions. The Colts are the best example, as they've been specifically created to stop the pass. The Colts have limited resources to spend on defense, and with their great offense, are banking on needing to play good pass defense more often than good pass defense. Because of that, they've got speedy edge rushers in Freeney and Mathis, and historically, had light but fast players in the interior of the defense. They're not even trying to be good against the run.
Outside of Indy, if you've got slow cover corners, there's not much you can do. If your team is full of bad tackles, no amount of positioning or scheming will ensure success against the run. You may throw extra cornerbacks on the field to sell out against the pass, but if your GM left you shallow at cornerback, there's not much you can do during the season (in the off-season, you can request new corners). If you're hit by injuries and lose your top three defensive tackles, you're going to struggle against the run.
Are there so many exceptions that they swallow the rule? I don't think so. This is all just theory, though. What do the numbers say? Check back tomorrow.
This entry was posted on Monday, November 9th, 2009 at 8:11 AM and filed under General. Follow comments here with the RSS 2.0 feed. Skip to the end and leave a response. Trackbacks are closed.
For the past few years, I've often advocated that the Vikings -- who've fit the "good run D, poor pass D" label to a T -- go nickel almost full-time, especially last year when they were down to their third- and fourth-string middle linebackers.
Posted on 09-Nov-09 at 8:22 am | PermalinkThis is basically game theory, right? Find that Nash equilibrium between run D and pass D that minimizes the other team's yards per play. Couldn't the same be said of offenses, though? Since the ability to run and do play-action makes your passing game better, and a good passing game prevents the D from putting 8+ in the box and makes your rushing game better, maybe there aren't good or bad passing/rushing offenses either, just good or bad offenses in general? But why does that sound weirder to say about offenses than defenses? Is it because we keep semi-meaningful stats on offensive skill players, and therefore want to believe that passing numbers are accumulated independently of rushing ones, and vice-versa?
Posted on 09-Nov-09 at 10:02 am | PermalinkWe can see a qb bounce a football a yard in front of a receiver's feet, then see an rb on the same team break two tackles on the same play.
The weakest link doesn't do it all for me. Two great pass rushing ends will do a lot to hide a weak safety, while with two average ends, that safety will be exposed.
Posted on 09-Nov-09 at 12:08 pm | PermalinkI agree, Chase. I have been considering doing something similar--using a similar model to Brian Burke, but not splitting passing defense from run defense. They really are the same defense, and the total yards/play and interception rates (and to a lesser extent, fumbles forced rates) are what matter.
Posted on 09-Nov-09 at 12:43 pm | PermalinkThe difference with offense is real.
Since the offense gets to choose whether to run or pass, it makes more sense to build a strength than cover a weakness. The Bears can go 20 years without a good QB and still have some winning teams; the Patriots can destroy people without an elite runner.
Successful offensive imbalance gets blown to absurd proportions with some college teams. Boise or Hawaii can get a QB too good for their league & score 50 points a game; a wishbone running team can win throwing 3 passes a game.
And this year's Oklahoma team can be competitive against anyone but lose 4 games because their rushing is the worst it's been in 40 years.
Posted on 09-Nov-09 at 1:28 pm | PermalinkI was thinking about this and i think it would be easier to have a good run stopping team bad pass stopping team be a good defence than the otherway around, since stopping the run well forces the opposing offence to pass more/take more risks. As you said before, rush attempts are worth something, pass attempts arn't, thus lowering rush attempts is a good thing.
That said, i think the biggest thing that creates pass/run differentials is the opponents they play against. If you play nothing but passing teams, your run defence is going to look better than similar strengthed defensive teams and passing look worse. Second, not all players can make a huge impact against the run and the pass, so in the short run, if all your best defensive players are good against the run, there is only so much you can realign your defense to compensate for the pass (in theory then, a mark of a good defensive coach might be making the trades to round out the defense you need, instead of just amassing raw talent).
Posted on 09-Nov-09 at 2:11 pm | PermalinkLastly, i think the "there are good defenses and bad defenses" should be ammended to "there are good defenses and bad run defenses and bad pass defenses." It is nearly impossible to be a good defense and have a hole to exploit, but it is quite easy to have a bad defense that is only bad at one thing (look at the vikings defense over the years, crush the run, mediocre overall due to lack of passing defense), no where is it necissary that a bad defensive team be bad at both pass and run.
Beware of using total yardage totals in these arguments. You can allow a lot of garbage time passing yardage without being a bad pass defense. However, if you are playing D in the 4th quarter while behind by 2 TD's, you may give up more rushing yards since the other team will be trying to run time off the clock.
Posted on 09-Nov-09 at 6:36 pm | PermalinkJoseph,
Posted on 09-Nov-09 at 9:25 pm | PermalinkI get what you are saying, but wouldn't that garbage time yardage still be indicative of a weak defense? When you know what they are trying to do, yet you still can't stop it- to me, that is poor defense.
Playing 1,000,000 times wouldn't eliminate "randomness." If the outcome of the games are stochastic phenomena, then they're stochastic phenomena. If it is akin to flipping a coin, then the multiple trials would likely lead to the outcomes approaching the true probability. However, it might also be a Markovian process if the same teams were actually playing and injuries in one game carried over to the next, therefore changing initial conditions on the next trial.
Posted on 10-Nov-09 at 6:12 am | PermalinkI don't think I agree. In a laboratory where you can just create endless drives and whatnot, sure. But I think in actual games, the offense's schemes are often limited by the score and specific matchups, and some matchups happen much more often than others. It's very important to have an answer for the Colts if you're the Jaguars since 12.5% of your games are against em, but they couldn't care less about the Saints since they'll be facing em only 1.6% of the time.
Posted on 10-Nov-09 at 7:45 am | Permalink