I thought he had 2. Weak, yes, but met the definition of multiple. 
I’m too disinterested to check, but I believe he went to 2 (after chosen qb’s bailed on the game).
I’m out of this thread.
This is very shortsided even after you factor out the points allowed by offense. (eg: Pick Six)
It is an accepted fact that offenses are more likely to score points the less distance they have to go when they first get the ball. If you rank defenses based on defensive points allowed, you can immediately and significantly improve (or hurt) your defense simply by changing punters. In other words, points allowed doesn’t just measure defense. Yards allowed does.
In related news, Jason Campbell was traded to the Raiders for a 4th Round 2012 draft pick.
OK, but so what? How big a role does this play, over the course of the season? Are there stats that show that one team’s offense begins its drives ten yards further up the field than another’s, on average?
And if this is true, then punters ought to go high up in the draft, and be paid the big bucks associated with the skill positions.
It can have a rather large affect, but it isn’t just the punter. Field position is dependent on where offensive drives end and how good the return coverage is as well.
There are other significant problems to the points against as a metric for team defense. Strength of schedules aren’t equal for starters. Also, it is easier to stop an offense in the winds of the Meadowlands than in a Dome. How good one’s offense is can effect the defensive performance as well. If an offense is on the field 35 minutes a game than the defense will give up less points. Additionally, not all points scored are equal. It you have a big lead you may be willing to sacrifice points if you can cause long drives.
Football Outsiders looks at plays on a per play basis. They aren’t perfect, but are much better than just looking at points allowed.
Well, yeah. But Ellis claimed that just a switch of punters would make a significant difference in points.
The problem is, every one of these factors is equally applicable to yards. Strength of schedule will affect yards gained and given up, as well as points. Ditto park/weather factors. If a defense is on the field only 20-25 minutes a game, it’ll also give up fewer yards, not just points. And Lord knows that not all yards given up are equal, which is my underlying point here. But there’s really a pretty low limit to how many points, over the course of a season, that one team is going to give up while running out the clock with a large lead. I’d be floored to find more than one or two instances in the past three decades where a team gave up enough late-game points in that way to raise their points-allowed by as many as 3 points per game.
Well yeah, that is why you don’t use yards either. What Outsiders uses is basically is success points. What happens on average on each play vs what actually occurred. “On first down, a play is considered a success if it gains 45 percent of needed yards; on second down, a play needs to gain 60 percent of needed yards; on third or fourth down, only gaining a new first down is considered success.” More info here.
Agreed. I don’t go by yards for those reasons. I go by yards because points allowed is largely a function of field position, so it doesn’t measure defense particularly well. Yards allowed are always yards allowed regardless of field position, so they measure actual defense.
For example, let’s say the offense gets the ball inside the red zone, maybe from a turnover or a big return on a kick or punt. What is a defensive success in this situation? Most people would say holding them to a field goal is a success. That’s actually a failure if you judge defense based on points. But if you judge them based on yards, holding them to a field goal means you held that drive to under 20 total yards, which is indeed a success.
Make sense?
The team we’re discussing here, the Redskins, illustrates the underlying problem with using yards as the metric. During the recent years of low-scoring offense, the defense has had a tendency to hold quite well for most of the game, allowing few points and few yards as well. Then at the end, they give up the one or two (usually) long drives needed to lose the game. So they at least appear to give up a high ratio of points to yards, without turnovers being involved.
What you’d expect is a more spread-out distribution of drives given up - some three-and-outs, some short drives, some medium drives, and some long ones. If a defense has a tendency to giving up very short and very long drives, it’ll give up less yards, but still give up the same number of points.
You could look at yards, then find a way to measure this too. (You’d call it something like “percentage of yards given up that are in scoring drives” or some such. If someone’s already come up with a stat like that, the degree of variation across teams would tell the story about the viability of measuring a defense by yards rather than points: flat distribution = use yards; lots of variation = use points.) Or you could just look at points, and take into account points off turnovers as needed.
The whole field-position thing: again, I’d want to see some evidence that, over the course of the season, it matters - that some offenses consistently start with better field position than others, and some defenses consistently start with worse field position than others. (If someone’s already done the work here, this time it’s flat distribution = use points; lots of variation = use yards.)
I think it’s much more simple than that. You can blame the defense for every yard allowed, while you can’t blame the defense for every point allowed. The question then becomes a matter of finding out what percentage of points aren’t the defense’s fault. But why bother, when we already know that every yard the opposing offense gains is the fault of the defense?
Because a lot of those yards don’t matter. But almost all the points do.
I agree that it’s much simpler, just in a different way.
Because some defensive philosophies result in more opposing yardage gained despite being equally effective overall - and some teams naturally give up more yardage than others as a result of their offensive philosophies.
Take the Colts and Eagles - both employ quick-strike offenses*, and this results in their defenses being on the field more than, say, the Panthers’. Both teams’ undeniably effective defenses might be ranked higher every year if their offenses held the ball longer.
*I know the Eagles are still ostensibly a West Coast passing team, but they’ve relied increasingly on the deep ball in recent years, as Andy Reid finally figured out where McNabb’s strength is.
Cute! Not well thought out, but cute!
There are those of us who opine that McNabb doesn’t have what it takes to win the big one. Doesn’t stop us from rooting!
Differences of opinion intensify interest in sport and in various rooting interests. The most rabid fans are often the most opinionated, and are often toughest on players. New York is a good example. It sure has stifled their interest and championships!
Woot!
A lot of those points aren’t the fault of the defense. All of the yards are.
It’s a lot easier to separate out the points that aren’t the fault of the defense, than the yards gained on the defense that never amounted to shit.
Look, the Redskins were 10th in yards allowed last year, but 18th in points allowed. You’re saying the yards are really a better gauge of the defense’s quality, and the difference between 10th and 18th must really be the fault of what the offense stuck the defense with in the way of turnovers and field position.
So far, I’ve seen no evidence presented that the differences in field position between teams in general amount to anything nontrivial over the course of the season.
I’ll cheerfully acknowledge that there are serious differentials between teams in turnovers. But the Redskins’ offense gave up 28 turnovers in 2009; league average was 27.3.
So I don’t see evidence that the Redskins were the 10th-best defense, but happened to get stuck with ~24 extra points allowed due to their crummy offense. I just see the defense of a team that was, overall, 18th in points allowed, with no reason so far to blame the offense for the differential.
IOW, I’m still seeing no reason to not evaluate this defense by points allowed. They’re the 18th-best defense, not the 10th.
Unless you’re talking about grabage time, this bespeaks a fundamental ignorance of the game of football. Football is a game of field position. While it might seem worthless to gain 6 yards on 3rd and 10, that extra field position can play a huge role in the momentum of the game.
To the point at hand, yes, the Redskins were the 10th best defense last year, not the 18th. I’ve explained why I (and the NFL) feel that yardage is more important than points. I’ve also linked to FootballOutsiders rankings, who while I personally don’t agree with much of their methodology, they also rank the the Redskins 10th. Exactly in line with yards allowed. I have the NFL and FO agreeing with my position. Do you have anyone agreeing with yours?
This is more of that fundamental misunderstanding. If you really think that field position is basically irrelevant, I don’t think we’ll ever be able to find common ground.
Here’s another disconnect. Offense and Defense aren’t the sum total of football. Why can’t some of the blame be on their special teams? Bad kick/punt coverage can really hurt your defense.
Also, have you even bothered to do this? A quick example from the most cursory inspection shows that Terrell Thomas of the NY Giants returned a Jason Campbell interception for a touchdown. That’s 7 points right there where the defense never even stepped onto the field, yet you downgrade the defense because of it.
Judging a defense based on points allowed is just plain silly.
Have we established that McNabb is a Redskin yet? I’m a little unclear.
You’re both right, and therefore both wrong. It’s already been said, but using DVOA addresses (as far as I can tell) all of your respective objections.
If you don’t want to use that, however, maybe drive stats? It doesn’t account for opponent quality, field position, or game situation, but it’s still better than using either raw yards or points. Washington’s Defense was 14th in yards allowed per drive, and 17th in points per drive. They were 12th in Drive Success Rate, which measures the percentage of down series that result in a first down or touchdown.
I will note that Washington’s defense had the 7th worst average field position to start drives, which, as **Ellis **points out, really will drive up their points allowed through no fault of the defense.
I don’t see how he’s even a little bit right. Also, DVOA doesn’t address any of my objections in that I have no objections about using yards to measure defense.