CB performance in 2007

[Mediator12]

I just got raw numbers from KC Joyner on how the CB's performed in coverage from the NFL last year and it was not encouraging. Couple of Points need to be addressed though.

CB performance is extremely tied to the front seven getting pressure on the QB and forcing the QB to hurry his throws and not progress his reads and be able to dump off balls. So we need to start with DEN's ability to do that last year. Here is the scoop on that:

1. Dumervil was the only DL starter in the top 64 in the league in sacks. Surprisingly, Tim Crowder did have 4.5 sacks last year in limited duty. However, 3 of those four were termed COVERAGE sacks. Coverage sack: A sack that occurs in the pocket more that three seconds after the snap.

2. DEN was 28th in the league in forcing bad decisions from QB's over the course of the season. That infers, but does not conclusively prove, that the pass rush was really poor overall, and absolutely dismal without Dumervil.

3. DEN had No LB's in the top 100 in Sacks.

4. Together, it means the CB's performance was extremely influenced by the lack of pressure from the front seven.


Denver had 3 CB's ranked in the top 94 in attempts. The first CB was second in the league in Yards per attempt (YPA). The second however was tied for 54th in YPA. And the third was tied for 73rd. This looks pretty good overall considering they all performed above their pass rush support, until you look at who those players actually were.

The top CB in YPA was Karl Paymah with an outstanding 4.7 YPA with 32 attempts and a league leading SYPA of 2.0. SYPA (Success percentage X Yards Per Attempt): A metric that combines both success percentage and YPA in an effort to measure a playerʹs overall effectiveness. His success rate in coverage was 56.3% for 4th in the league. For all the crap people dish on Paymah, he led the league in SYPA last year.

The 54th best CB in YPA was Champ Bailey who dropped almost 3 YPA from last year with 7.8 on 60 attempts. His success rate was the worst of his career with only 31.7 %. His SYPA was 5.3 also the worst of his career.

The 73rd ranked CB was Dre Bly with an 8.8 YPA on 79 attempts. His success rate was better than Champs though with 41.8%. His SYPA was the same at 5.3.

The good news is that Hamzah was the second run safety in the NFL in direct coverage with 2.1 YPA on 13 attempts. His success rate was an outstanding 76.9%! His SYPA was 1.61.

John Lynch was perfect in Deep coverage assist AGAIN for the third straight year yielding NO completions in deep help. Deep assist coverage: Coverage where a safety is responsible for helping another defender cover a pass downfield. So, for all the people who say he has lost a step, it did not matter in deep coverage and never really has. It just means they did not use him much in direct coverage, which has never been his strength.

Overall, this just shows that the pass defense was not as good overall as it should be. It was hidden by the fact that the run defense was so poor overall.

I do not think that anyone questions the talent of Bailey or Bly. It was just that they were left out to dry in that scheme and were abused having to cover longer than their contemporaries. What is surprising is that Paymah was so effective behind that same front seven. By contrast, PIT's average CB duo ranked in the top 5 based on how effective their front seven was in creating pressure.

When you finally look at the numbers, you see that the whole defense has to improve at the LOS in order to make the turnaround everyone desires. I just want to see Moss, Crowder, Thomas, and Robertson make a positive effect up front so the back end does not have to cover for them again. Look what that did to Al Wilson, Ian Gold, and Even Mobley at LB. It drastically shortens their careers.

[TheReverend]

Great post... but one significant issue with your statement at the end... Mobley broke his neck in a car accident and that's really skewing facts to lump him that group.

[400HZ]

Joyner's metrics always seem too uneven to me. Who was first in CB ypa? That Texans 4th round rookie? I'm sorry, but if your metrics tell you that a rookie and Karl Paymah are the two most successful coverage CBs in the league, then there is a problem with your metrics.

[socalorado]

Originally Posted by 400HZ

Joyner's metrics always seem too uneven to me. Who was first in CB ypa? That Texans 4th round rookie? I'm sorry, but if your metrics tell you that a rookie and Karl Paymah are the two best coverage CBs in the league, then there is a problem with your metrics.

OH NO! You dare challenge "metrics boy"!

[400HZ]

Originally Posted by socalorado

OH NO! You dare challenge "metrics boy"!

There are plenty of very useful ones out there. I'm just skeptical with Joyner's.

[theAPAOps5]

Originally Posted by 400HZ

There are plenty of very useful ones out there. I'm just skeptical with Joyner's.

What are some of the ones you like 400. Just curious

[Mediator12]

Originally Posted by 400HZ

Joyner's metrics always seem too uneven to me. Who was first in CB ypa? That Texans 4th round rookie? I'm sorry, but if your metrics tell you that a rookie and Karl Paymah are the two most successful coverage CBs in the league, then there is a problem with your metrics.

Not really. There is a hype problem with how effective "Big Name" players really are. Guys who gamble so much Like Asante Samuel get overrated because they get INT's and are on Sportscenter highlights. Then, the same people complain about the Probowl being a popularity contest

Also, you are excluding the context I spent so much time setting up. Coverage is HIGHLY dependent on the front seven. Something DEN has not had since 2003. So, in short, please take the whole post in to effect before trying to redress problems already explained.

[TheReverend]

Originally Posted by 400HZ

Joyner's metrics always seem too uneven to me. Who was first in CB ypa? That Texans 4th round rookie? I'm sorry, but if your metrics tell you that a rookie and Karl Paymah are the two most successful coverage CBs in the league, then there is a problem with your metrics.

All stat work needs to be taken with a grain of salt no matter how in depth it is, but there's always some semblance of valuable information to take away from the analysis.

[Mediator12]

Originally Posted by socalorado

OH NO! You dare challenge "metrics boy"!

Thanks for the support!

BTW, I am still waiting for anything relevant from you, but I won't stay up waiting for it

[400HZ]

Originally Posted by Mediator12

Not really. There is a hype problem with how effective "Big Name" players really are. Guys who gamble so much Like Asante Samuel get overrated because they get INT's and are on Sportscenter highlights. Then, the same people complain about the Probowl being a popularity contest

Also, you are excluding the context I spent so much time setting up. Coverage is HIGHLY dependent on the front seven. Something DEN has not had since 2003. So, in short, please take the whole post in to effect before trying to redress problems already explained.

You're making a good point. I wouldn't dispute that. I was talking more about K.C. Joyner specifically, not neccessarily in relation to the point of your post. Most of what he publishes is very in-depth and readable, but he puts out some head scratchers as well. I like football metrics and I appreciate the information they provide - like you said, there are plenty of big name players with over-inflated stat sheets who really aren't that good play-to-play. K.C. Joyner is puzzling sometimes, though. It seems like he overanalyzes. That article he published about the relative importance of each offensive line position, for instance. It was original and made you think, but can you really sit there and say that all 32 NFL franchises have their allocation of resources on their O lines ass backwards? Stuff like that I have a tough time believing.

[400HZ]

Originally Posted by ApaOps5

What are some of the ones you like 400. Just curious

Pro-football-reference comes out with some neat ones. I think from a statistical analysis standpoint, Football Outsiders is unbeatable. Coldhardfootballfacts aren't as in-depth as most, but they are worth reading. And like i said, I read and value Joyner's, but I just think he overanalyzes to a certain extent. I don't know which sets are the most accurate, but those are the ones that typically make the most sense to me.

[SonOfLe-loLang]

KC seems like a nice guy, but I honestly think his work is pointless. Football is a game, more than any other, where individual performance rely on the other 10 guys around you. There are so many variables that effect every play, its really impossible to single a guy out. All of his statistics and metrics should be taken with a grain of salt.

[NaptownChief]

Originally Posted by 400HZ

Joyner's metrics always seem too uneven to me. Who was first in CB ypa? That Texans 4th round rookie? I'm sorry, but if your metrics tell you that a rookie and Karl Paymah are the two most successful coverage CBs in the league, then there is a problem with your metrics.

I know one thing for sure, stats really suck when they don't tell the story you want to hear.

[DomCasual]

I don't think you can look at last year as anything but an anomaly. That doesn't mean our defense is good. But we couldn't even get enough players on the field at times last year. Does anyone know a precedent for a team changing defensive schemes mid-season?

Last year was a train wreck on defense. It's damned amazing that we won seven games.

[Kaylore]

Metrics have to be taken with a grain of salt. They don't say how a scheme hides some of the problems that players have. A good example is how when Foster was here they ranked him in the top twenty offensive linemen. He's probably going to be out of football next year.

[broncosteven]

3. DEN had No LB's in the top 100 in Sacks.


I am curious to know how many times they actually blitzed a LB last year. Also what were the metrics that he uses for LB's? Same, or just based on total # of sacks?

It is apparent that the D did not take to Bates or his scheme, players thought to much and were not dynamic or fluid in their execution. Execution of all 10 other guys really has to effect the outcome of these metrics.

I really hoped Bates would be serviceable enough and able to be here long enough to get players that would fit his system. I was wrong.

Side note of stat misuse (not that that is what I think this is):

Hitler studied facts relentlessly. He had no formal training in Military tatic's and was only a runner in WWI. He would flabbergast his generals and advisors in the OKH and OKW by citing every stat imaginable. For example he could quote exact # of Panzers produced in 1943 when confronted by his generals in the East when they said they needed more tanks. They could not dispute his stats and when they did test him he turned out to be right.

My point is that Hitler did not account for Human element or human needs especially in east during Barbarossa. Had he kept feeding the supply lines, as well as keep them moving, and addressed winter needs of his army's before the weather turned his plan could have worked.

A great leader would account for stats and the failings of men. Just procurring a bunch of Supermen (either soliders or athletes) won't ensure victory as they still must endure the limits of their Humanity, they are still men and subject to mortal imperfection.

Stats are a tool, not a solution.

[That One Guy]

Mediator says from the start that these stats are influenced by the DL. My question though, if we know that... why did we all keep reading? Who cares where Bailey ranks if he will inherently rank toward the bottom of the league based on the DL he was playing behind? Also, for one to be able to include Paymah with the likes of Bly or even Champ, you'd have to assume Paymah was playing against the same level of skill. Maybe the fact that he only played on sucky players meant they were easier to cover? Maybe the fact that Bailey and Bly were so easily exploited meant that they only went to the schmuck that Paymah was covering in emergency situations when the play had collapsed and it was just a last ditch effort?

When you preface the stats with a half dozen difference outside influences that had nothing to do with the players individual performance and none of the players are being judged on the same scale (ie Champ on #1 WR, then Bly on #1 WR, then Paymah on #1 WR) then it seems like it's absolutely bogus stats and figures that we shouldn't even attempt to take any valid data from.

[That One Guy]

I should add though, that's nothing against Mediator or the guy that drew up the facts. There's probably other situations where these statistics are much more telling, it's just not in a position so easily influenced by outside conditions.

[broncosteven]

Originally Posted by cah412

Mediator says from the start that these stats are influenced by the DL. My question though, if we know that... why did we all keep reading? Who cares where Bailey ranks if he will inherently rank toward the bottom of the league based on the DL he was playing behind? Also, for one to be able to include Paymah with the likes of Bly or even Champ, you'd have to assume Paymah was playing against the same level of skill. Maybe the fact that he only played on sucky players meant they were easier to cover? Maybe the fact that Bailey and Bly were so easily exploited meant that they only went to the schmuck that Paymah was covering in emergency situations when the play had collapsed and it was just a last ditch effort?

When you preface the stats with a half dozen difference outside influences that had nothing to do with the players individual performance and none of the players are being judged on the same scale (ie Champ on #1 WR, then Bly on #1 WR, then Paymah on #1 WR) then it seems like it's absolutely bogus stats and figures that we shouldn't even attempt to take any valid data from.

I do find it interesting in that it is a window into what those Quality Control guys look at week in and week out. I think I have read that they prep all these stats into a format that the upper assistants use to prep for games.

I find it interesting even though I am a casual fan.

[That One Guy]

Originally Posted by broncosteven

I do find it interesting in that it is a window into what those Quality Control guys look at week in and week out. I think I have read that they prep all these stats into a format that the upper assistants use to prep for games.

I find it interesting even though I am a casual fan.

But I liken it to a mileage study where some cars were being tested going uphill, some going downhill, some full of cargo, some without cargo, some heading into the wind, some at their back, etc... and then they throw em all together and some crappy Kia SUV comes out on top. We know the SUV doesn't get the best mileage and once you look at the stats, you suddenly want to slap yourself for even conducting the test. There's no way you can give these things any credibility whatsoever. Just too many variables.

[broncosteven]

Originally Posted by cah412

But I liken it to a mileage study where some cars were being tested going uphill, some going downhill, some full of cargo, some without cargo, some heading into the wind, some at their back, etc... and then they throw em all together and some crappy Kia SUV comes out on top. We know the SUV doesn't get the best mileage and once you look at the stats, you suddenly want to slap yourself for even conducting the test. There's no way you can give these things any credibility whatsoever. Just too many variables.

He did infer that the #'s were skewed by lack of play in Front 7:

"4. Together, it means the CB's performance was extremely influenced by the lack of pressure from the front seven."

Anyway the best thing about these stats is that they are 2007's #'s and not current #'s.

It is just a measuring stick to compare progress.

[That One Guy]

Yeah, I pointed that out. I may be the only one that has an issue with em, I don't know.

And yes, I do see the stupidity in continuing to post in a thread that I say basically has no merit.

[Mediator12]

Originally Posted by cah412

I should add though, that's nothing against Mediator or the guy that drew up the facts. There's probably other situations where these statistics are much more telling, it's just not in a position so easily influenced by outside conditions.

You are completely missing the point. It is a case of not seeing the forrest, because the trees seem funny. The whole point of looking at these stats, even though they are heavily dependent on other variables, is to curb expectations and grade true performance. Do not put so much stock in the metrics measuring ability, they measure performance. Performance IS tied to all 11 players. What these say is that Champ Bailey and Dre Bly were not as effective as their talent in DEN's scheme and personnel in 2007. Nothing more, nothing less.

We already Know that Champ is NOT as effective behind a piss poor DL and neither is Dre Bly, so cut them some slack when they get beat covering for 4 seconds. Half of the teams Sacks were Coverage sacks, so give the Secondary credit for taking away reads to create pressure on the QB. Do not think the pass rush is there or the run defense is there UNTIL is proves otherwise!

The analysis of this is simple, the front seven was completely ineffective last year and needs to improve dramatically for the "D" to get back to just average in both aspects. People forget to measure how poor the pass defense really was last year because the running game was so miserable. The poor CB numbers are obviously a product of their poor play as both had career worst numbers. Now, does some of the blame rest on getting older and scheme, absolutely. However, this is a problem that needs to be addressed as a unit. Again, that is why this is relevant.

[maher_tyler]

Originally Posted by Mediator12

Not really. There is a hype problem with how effective "Big Name" players really are. Guys who gamble so much Like Asante Samuel get overrated because they get INT's and are on Sportscenter highlights. Then, the same people complain about the Probowl being a popularity contest

Also, you are excluding the context I spent so much time setting up. Coverage is HIGHLY dependent on the front seven. Something DEN has not had since 2003. So, in short, please take the whole post in to effect before trying to redress problems already explained.

I agree with the D line and them getting no presure which makes Champ an even better corner, thats why i hate arguing with people who aren't Denver fans why hes hands down the best CB in the league, possibly ever...Denver has never had a very good d line since he's been here yet he still puts up DMVP numbers every year...maybe not every year but for sure respectable numbers

[RunSilentRunDeep]

A rising tide is going to lift all boats this season. Between better players, better scheme (for the talent in it) and improved cohension the defense is going to take a big step forward. Don't forget Champ had a bum leg for a lot of last season.