Chargers Stat of the Day: Defensive Ratings
by John Gennaro on Nov 17, 2009 4:05 AM PST
in San Diego Chargers Statistics
Email
PrintAttention all nerds. Do you head straight to Football Outsiders when you're looking for better-than-average stats? Well, I found another website that you're going to want to bookmark. (Editor's Note: This is not a paid advertisement. I've never even talked with the guys from this website. So if this sounds like a sales-pitch, it's not.)
ProFootballFocus.com is a website that football stat-geeks have been dying for without knowing it. Just about any question you have about the team can be answered within their stats. Want to know who has played the most defensive plays for the Bolts? Eric Weddle, with 544. How about which player has the most QB Pressures? It's a tie between Luis Castillo and Shaun Phillips (both have 10). Wondering who the best MLB on the Chargers is? You'll be surprised.
Let's delve into some of PFF's defensive stats and player ratings...
I was taught a long time ago to never argue with someone who was smarter than me. I almost never abide by this rule, but today I'll make an exception. These guys have done such a great job at collecting stats, that I'm going to assume they do just as good of a job comparing them.
In terms of defense, it appears they look at all of the stats (sacks, QB hits, QB pressures, tackles, penalties, missed tackles, stops) and compare them relative to the number of plays the player had played. So a played like Brandon Siler, who has 10 stops on 104 plays, would be looked at as a more efficient player when he's on the field than Stephen Cooper (22 stops in 454 plays).
Each player is rated based upon their work rushing the passer, stopping the run and in pass coverage. Also factored in are things like Penalties being called against that player (surprisingly, leading the Chargers D in penalties is a tie between Shawne Merriman, Shaun Phillips and Larry English) and Missed Tackles.
The missed tackles stat also needs to be looked at proportionally against actual tackles. For instance, Stephen Cooper leads the team with 10 missed tackles to go with his 48 tackles on the season. However, that's probably not as bad as Antonio Cromartie's 6 missed tackles with only 15 tackles to his name.
So, follow me down the rabbit hole as we look at PFF's ratings of Chargers defensive players by position group. The number next to each player is their PFF Rating, which is roughly explained here. In a nutshell, zero is average across the league and any number above that is above average. Any number below is obviously below average. Also, for the sake of this not being too confusing, I'm leaving out any player who has been on the field for less than 75 snaps.
Defensive Line
- Travis Johnson -1.5
- Ian Scott -3.0
- Alfonso Boone -3.1
- Ogemdi Nwagbuo -3.1
- Jacques Cesaire -9.5
- Luis Castillo -10.3
How about that trade for Johnson? The Castillo number is a bit jarring, but appears to be accurate. Although he's near the top on the team in terms of defensive stops and QB pressures, apparently relative to the number of plays he's played it's just not as impressive as it has seemed. Also, according to their individual rankings he's fairly average at rushing the passer and well below average at stopping the run. For those looking for a comparison, Castillo's 2008 was ranked at 0.8 (or just a little above average) and Jamal Williams' 2008 was ranked at 27.4. 27.4! In 2007, Castillo was well above average at 7.7 when he was on the field.
Linebackers
- Brandon Siler 5.8
- Shaun Phillips 2.7
- Kevin Burnett 2.6
- Stephen Cooper -2.4
- Tim Dobbins -2.7
- Shawne Merriman -3.3
- Larry English -12.2
I told you this stuff was interesting. As mentioned above, Cooper is killed by his missed tackles and by being below-average (-1.5) in pass coverage. Siler is actually the only LB that ranks as above average against the run and in pass coverage this year. Phillips would probably be #1 if it weren't for the 4 penalties he's picked up this season. This group is greatly improved from 2008, when no LB ranked as average or above average (Wilhelm was -16.9 if you were wondering).
Cornerbacks
- Antoine Cason 1.4
- Steve Gregory -1.0
- Quentin Jammer -3.8
- Antonio Cromartie -5.6
Jammer is actually playing better than he did in 2008, when his season ranking was -5.9. The stats seem a little off for cornerbacks in terms of what's "average", but at least they appear to be consistent. Cromartie's 2008 rating? -27.4. He was the anti-Jamal Williams.
Safties
- Eric Weddle 3.7
- Paul Oliver 0.5
- Clinton Hart -1.8
- Kevin Ellison -2.7
Before you start screaming about Hart being rated ahead of Ellison, listen up. Here's what I gather as the explanation by looking at the stats they're presenting. Both players are equal against the run (0.5), and the rookie actually better in pass coverage (-2.3 to -3.7), but the main two factors that catapult Hart above Ellison is penalties (zero for Hart, 2 for Ellison) and pass rushing (1.4 for Hart, 0.1 for Ellison). I can't entirely disagree with that, and considering these guys are charting the plays and I'm not....I'm going to believe them.
Still, it was the right move to get rid of Hart. Ellison is an improving player and Clinton was a declining one. Also, it has opened up plenty of playing time for Paul Oliver (who is obviously playing well).
So what do you guys think? Interesting stuff, right? Obviously, like all stats, these are not perfect. From what I can tell, these ratings do not factor in the level of competition being faced (which explains why Gregory would be higher than Jammer and Cromartie) nor do they factor in double-teams (which explains why Luis Castillo is down this year and why Shaun Phillips had such a poor rating in 2008). Still, in terms of judging who is making the most impact with their time on the field....it's intriguing.
Stats and rankings taken from ProFootballFocus.com do not include information from the Week 10 match-up against the Philadelphia Eagles.
32 comments
|
1 recs |
Do you like this story?
Comments
This site was pointed out to me just recently
at another forum. Still digesting – you really have to delve into the explanation of their ratings first, then back to the stats. Otherwise, at first blush, a lot of players that you would consider as quite good look to be very bad.
No statistician by any stretch, so I can’t comment very effectively on the validity of the data, but it does seem to me that they’ve made the effort to include a lot more variables than most sites. I’ll definitely be using the data. Will be interesting to see how this compares with both pacstud & others who have been around longer.
If the thunder don't get ya then the lightning will!!
Robert Hunter
Yeah, I’ve been looking at this site for about 2 weeks trying to figure out what to do with it. There’s just so much there.
The data is valid, I know that much. The ratings are a bit more complicated, but I’m going to assume they’re at least close. It’s really not too far off from the ratings I would’ve given them off the top of my head.
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 17, 2009 4:21 AM PST up reply actions
even on Castillo?
Thought he was having a solid year, one of his best. Admittedly, haven’t studied film, and have only seen a handful of games. Is this a representation of his performance, or do the stats do him an injustice based on his position and role, (such as forcing double teams, etc.)?
The stats/ranking does not take into effect a double-team. I am with you, I think this is Castillo’s best season and I think he’s handing the double-teams well. However, just based off the fact that he’s getting doubled now his numbers were bound to go down.
He’s also being punished for being the only good/healthy D-lineman for a few weeks early in the season. If you look at more recent games (like the Giants game), his numbers seem more on par. He got a 0.2 rating for the Giants game, which is about where he was last season (when he wasn’t getting doubled). Castillo has played a lot of average games while being doubled, and 2-3 games where he was below average (mostly because he’s just not that great at run support and some teams ran all day). If you take away the double, he would at least be up around the “average” rating.
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 17, 2009 5:45 AM PST up reply actions
I’d guess that earlier in the year when the D’line rotation was more in flux and we were getting gashed his numbers would be down. Wouldn’t he now be the focal point on the line as far as opponents game plans go, with Jamal out?
I love stats, but even the best stats in football don’t tell the whole story.
Thanks for your response, and thanks for the post!
Interesting to read the stats
that they have come up with. I posted after the Giants game that I thought Siler deserved more time, based on his play. I am not sure about Oliver yet, but I would like to see him get more time as well, I think we need to have a safety on the field that can man up and cover someone when needed.
I know Weddle and Ellison are pro bowlers and I am an idiot. Just thought I would save somebody some time.
The issue with Oliver is that he is certainly not a SS.
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 17, 2009 9:41 AM PST up reply actions
I would agree
I feel like Weddle and Ellison are both lacking in the coverage department and I would like to see Oliver play a little more. Oliver may not be the answer either. I understand your point; Weddle is not going to be taken off the field for Oliver and Oliver is not a SS, so his time will be limited. (If that was not your point, then maybe I did not understand it).
That was my point, however I don’t agree with Weddle lacking in the coverage department.
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 17, 2009 10:12 AM PST up reply actions
Nor do I
Weddle doesn’t have the greatest closing speed, but he’s got great football smarts & is usually around the ball. When he gets burnt it’s usually bad, but not often.
It won’t happen this year, but I’d like to get a much better look at Spillman. he’s been on ST lately so I need to focus on him some there. In preseason he looked pretty darn good – tackled well, hit hard, & was decent in coverage. With more reps I think his coverage might be quite good – has good closing speed & moves well – fluid hips for a safety.
As for the stats, as I said – need to look deeper at them & I’m no statistician. IMO, if it could all be put into numbers, we wouldn’t play the games.
If the thunder don't get ya then the lightning will!!
Robert Hunter
by Buck Melanoma on Nov 17, 2009 3:38 PM PST up reply actions
I did not expect people
to agree with my thoughts on Weddle’s lack of coverage ability. Most people love Weddle as he is easy to like, as you say he is always around the ball and hustles.
In my opinion when he gets burnt its usually because he is being asked to cover somebody, something that he is not that good at doing.
I know I am almost all alone on this one, but I can live with it.
I am more inclined to go with some middle ground here
If you read the quote that ramezes posted it kind of highlights that. Weddle’s coverage skills might actually be quite good (for a safety), but the area he is asked to cover is too large for him. It might be that for FS in the Chargers offense to succeed he’d have to have elite coverage skills, while Weddle’s are just good. Maybe that’s what you meant by “lacking” in your first comment, but the second says that he has a “lack of coverage ability”, which, to me, is a tough sell. So, I’m willing to say his coverage ability is lacking for the Chargers scheme, but that he does have coverage ability. I’ll also go so far as to say, it’ll will be really tough for the Chargers to improve from Weddle at that position because there aren’t many safeties that have that sort of coverage ability.
Wisdom can not be cultivated through ignorance of information.
by Wonko on Nov 17, 2009 4:52 PM PST up reply actions 2 recs
After an initial read of their methodolgies
The one question in my mind is, “How does this relate to winning?” One of the things that brings universality to Football Outsiders (and, mind you, I’m not trying to make them gods of football stats because I’m all for competition I’m just using them to illustrate a point) is that they are all focused on giving players/teams credit for doing things that have historically led to winning football. Baseball stats are similar in that regard where OBP is a basis to all offensive metrics because it correlates the best with scoring runs, which are necessary for winning. Advanced basketball metrics are the same way.
Obviously, doing better than average at the skills of your position is very important and the grading system highlights who is doing this. But, I just feel like the next level here is to figure out how you can use this information to determine how well a team can get points (or defend against points) and wins out of it. Without that, it’s just a lot of fun numbers.
Wisdom can not be cultivated through ignorance of information.
by Wonko on Nov 17, 2009 11:49 AM PST reply actions 2 recs
I couldn’t agree more.
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 17, 2009 2:09 PM PST up reply actions
You're right, Wonko
This is why I can’t place but so much stock in stats. They’re helpful in many cases, but when it comes to the win/loss column, it comes down to some things that just can’t be fully quantified.
How much fun would it be to watch a game & already know the outcome? Sure, there are historic, memorable games that we like to re-watch, but knowing the outcome regularly would suck.
I primarily use stats for a very few reasons – to prove a finite point, to evaluate for the draft, & to some degree to do any post-draft evaluations. I have little interest in any NFL stat during the year except W/L & availability of alcohol on game day.
College is a little different.
If the thunder don't get ya then the lightning will!!
Robert Hunter
by Buck Melanoma on Nov 17, 2009 3:45 PM PST up reply actions
I’ve only looked at the site briefly, but it seems useful. I think you’ll find that if you just go by the numbers you’ll be wrong a lot, though. They can’t take into account some important things especially with regard to the defensive line. Double teams and poor depth can really make a big difference.
Bolts from the Blue // "Game over." - Jamal Williams
Bloody Elbow // "Every normal man must be tempted at times to spit upon his hands, hoist the black flag, and begin slitting throats." - H.L. Mencken
I refuse to believe that Castillo has been the worst defensive lineman this year.
Your eyes tell you he’s clearly been the best.
by Zach (maestro876) on Nov 17, 2009 12:49 PM PST up reply actions
He’s certainly the best on the Chargers this year. I don’t watch other teams’ games closely enough to say for sure where he ranks overall, but I sincerely doubt he’s anywhere near the bottom of the league.
Bolts from the Blue // "Game over." - Jamal Williams
Bloody Elbow // "Every normal man must be tempted at times to spit upon his hands, hoist the black flag, and begin slitting throats." - H.L. Mencken
by Richard Wade on Nov 17, 2009 1:34 PM PST up reply actions
That's what I mean.
Their rankings have him as the worst on the Chargers, and that just isn’t the case.
by Zach (maestro876) on Nov 17, 2009 1:37 PM PST up reply actions
I’m saying their other stats are useful (hurries and whatnot). I don’t think their “overall rating” stats really mean anything substantial.
Bolts from the Blue // "Game over." - Jamal Williams
Bloody Elbow // "Every normal man must be tempted at times to spit upon his hands, hoist the black flag, and begin slitting throats." - H.L. Mencken
by Richard Wade on Nov 17, 2009 1:52 PM PST up reply actions
I know the guys who run the site and I had a similar argument with them last year when Hart was grading out as better than Weddle. Its over analysis in my opinion but its nice to be able to check the stuff like player participation and some other things. The system is let down by being able to 100% factor in scheme, opposition and responsibilities although the guys who do it all know their stuff and do their best
I’ll ask one of the guys why Castillo rates so low as I agree he’s probably been the best d-lineman this year and certainly far from the worst. If anyone has any other questions regarding the grades I can certainly pass them along
Sounds like
they still need some tweaking.
by Neoplatonist Bolthead on Nov 17, 2009 1:42 PM PST up reply actions
I’d like to know the result of your conversation with them regarding Weddle vs. Hart this year.
In reality, they should explain how each position is graded. Seeing Hart above Ellison is also a bit of a headscratcher.
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 17, 2009 2:12 PM PST up reply actions
It wasn’t this year, it was 2008 and at the time the gradings hadn’t been done on every game over the period of time we were arguing about. This was the response
“Taking Weddle and Hart as this for starters, I don’t doubt for a second that Weddle is a better player in space than Hart, but we’re not judging that, we’re judging how well their doing their jobs. Weddle is being asked to play in much more space than Hart and match up against receivers 1 on 1 far more frequently than Hart is. Hart couldn’t do that job, no doubts, but he isn’t being asked to do that job. What those gradings show is that Hart is playing more effectively as the up safety playing in limited space more frequently on TEs closer to the line than Weddle is doing his job covering deep downfield on receivers in more space.
There’s no doubt Weddle is a more talented player than Hart, but we’re not scouts, we’re analysing games and from breaking plays down Weddle is not as effective in the role that he has been given as Hart is in the role he has been given. That compares to last year where Weddle was playing closer to the line and lining up in the slot in nickel and dime sets, he was very effective in that role according to our analysis. We can’t say that he’s become a worse player over the off-season, all that we can see and say is that Weddle is far less effective in his new role and isn’t performing to the level that he managed last season."
I can sort of see the points but that is the main downfall of the system – it can’t factor in complexity of roles, only specific player performance
There is an explanation of the site here (under Grades)
http://profootballfocus.com/about.php
I know John can agree with that analysis
Wisdom can not be cultivated through ignorance of information.
Yup
"I did not invent the wheel, I was the crooked spoke adjacent." - Aesop Rock
by John Gennaro on Nov 18, 2009 2:19 AM PST up reply actions
Why can't they factor in something like double-teams?
Is it because there’s no official measurement of that (and if so, why not)?
Passion Play - follow the annual quest for the premiership in all its horror and glory, http://spunc.com.au/members/hunter/product/9780980517965/
I'm assuming
You’d need something to give player’s a bonus for what they do in a double team. So, holding your ground in a double team would be similar to beating your man single teamed or something. The hard part would be identifying what in a double team is equivalent to something else single teamed.
Wisdom can not be cultivated through ignorance of information.
I find Cason's rating a bit strange
It’d be nice if they could factor in opponents (DVOA style?), but Cason also seems to have made more critical errors, which is maybe where Wonko’s point re the relation between these numbers and winning comes into it. Another great stat would be to judge the effectiveness of the secondary as a whole (or any other line), when a particular player is on the field. So Cason’s numbers might be ok, but if he is trying to do a bit much and that throws things out of whack elsewhere, then it’ll come up in the numbers. I’m sure you know that the Houston Rockets apparently picked up Shane Battier because he made everyone else play better.
Also, Ellison has missed some important tackles, but we give him kudos for creating pressure and have some decent reasons to hope that he can improve on this, whereas Hart didn’t seem to have much upside.
Passion Play - follow the annual quest for the premiership in all its horror and glory, http://spunc.com.au/members/hunter/product/9780980517965/
Comments For This Post Are Closed
Secondary Sidebar
User Tools
Related FanPosts on Bolts From The Blue
FanPosts
Community blog posts and discussion.
Recent FanPosts
BFTB Apparel Store
Certain photos copyright © 2012 by Associated Press or Getty Images. Any commercial use or distribution without the express written consent of Associated Press and Getty Images is strictly prohibited.
