Writers of Pro Football Prospectus 2008

18 Mar 2011

Worst Cornerback Charting Stats 2010

by Aaron Schatz

On Wednesday, we took a look at the best cornerbacks of the season according to the FO game charting project. Today, we'll flip things around and look at the corners who rank at the bottom of the game charting stats. I've ranked all cornerbacks with at least 40 charted passes, which means 83 total cornerbacks. I've removed passes marked as Hail Mary, Hit in Motion, Tipped at Line, or Thrown Away. I've also removed wide receiver screens, which aren't really a good way to measure cornerback coverage because a cornerback in man coverage is going to (or at least, is supposed to) immediately get blocked out of the play by another wide receiver. Right now I'm only looking at primary defenders, so this is not adjusted to account for double coverage, or plays where the charter marks a hole in zone but lists an appropriate zone defender in the second DEFENDER column. There are no opponent adjustments yet, and pass interference is not yet included.

Typical caveats apply, of course: This is imperfect data based on the game charting project, which means it comes off limited television camera angles. In past years, this data has been very inconsistent from year to year, and we're going to study that in the offseason to try to figure out if we can get more accurate numbers by, say, looking at players over two-year spans.

Success Rate, to remind everyone, is the percentage of passes that don't manage to get at least 45 percent of needed yards on first down, 60 percent of needed yards on second down, or 100 percent of needed yards on third down.

Bottom 10 Cornerbacks in Success Rate, 2010
Player Team Charted
Targets
Yd/Pass Rk Avg. Pass
Distance
YAC Rk Success
Rate
Rk
Jonathan Wade 2TM 40 9.9 77 15.7 2.7 16 38% 83
DeAngelo Hall WAS 88 10.5 81 10.4 5.8 81 39% 82
Jacob Lacey IND 44 9.5 75 10.3 4.5 69 39% 81
Chris Gamble CAR 41 7.7 50 11.5 4.6 70 39% 80
Fabian Washington BAL 43 10.4 79 11.1 5.9 82 40% 79
Mike Jenkins DAL 80 10.5 80 15.3 4.3 66 40% 78
Richard Marshall CAR 82 7.2 40 9.0 3.2 34 41% 77
Shawntae Spencer SF 67 8.5 62 10.7 2.7 17 42% 76
Charles Tillman CHI 84 8.1 58 10.9 3.6 46 43% 75
Terence Newman DAL 83 9.1 71 13.4 3.4 41 43% 74

In the discussion thread for Wednesday's article, someone asked me why I created a minimum of 40 passes when the best cornerbacks are often the guys that opponents won't throw at. This is the reason why. Even with a minimum of 40 passes, you can see the effect of small sample size, as four of the bottom five guys were near 40 passes. Of course, you'll have more of these guys at the bottom than you will at the top, because some of these players had fewer passes because teams pulled them out of the starting lineup. Fabian Washington is a good example, starting the first eight games for Baltimore and then mostly disappearing from the field because he was so horrible in the first half of the season.

This list also shows you just how ridiculous it was to send DeAngelo Hall to the Pro Bowl. Is he really the second- or third-worst starting cornerback in the league? No, his numbers were better than this in past seasons. But he's certainly not one of the best cornerbacks in the league, even after you consider all his interceptions.

The other thing we see on this list was the complete implosion of the Dallas secondary this season. There were some injury issues, but Mike Jenkins and Terence Newman were horrible all season. There have been reports this offseason that the Cowboys may cut Newman, but they can't really cut Jenkins, who is still young and was very good in 2009. They just need to figure out what the hell happened to him last year.

Bottom 10 Cornerbacks in Yards per Pass, 2010
Player Team Charted
Targets
Yd/Pass Rk Avg. Pass
Distance
YAC Rk Success
Rate
Rk
Kareem Jackson HOU 73 11.1 83 15.3 6.0 83 47% 67
Alphonso Smith DET 47 10.6 82 14.4 4.8 76 47% 66
DeAngelo Hall WAS 88 10.5 81 10.4 5.8 81 39% 82
Mike Jenkins DAL 80 10.5 80 15.3 4.3 66 40% 78
Fabian Washington BAL 43 10.4 79 11.1 5.9 82 40% 79
Rashean Mathis JAC 64 10.1 78 14.1 4.9 79 44% 72
Jonathan Wade 2TM 40 9.9 77 15.7 2.7 16 38% 83
Eric Wright CLE 53 9.8 76 12.5 4.6 71 43% 73
Jacob Lacey IND 44 9.5 75 10.3 4.5 69 39% 81
Jason Allen 2TM 58 9.3 74 16.7 3.4 39 52% 46

Jason Allen actually had his best season in 2010, especially after Miami released him and Houston picked him up. His high yards per pass figure comes from just a handful of plays, mostly by Braylon Edwards. For some reason, Allen was dominated by Edwards, who had receptions of 67, 42, and 27 yards against Allen in two games (one with Miami, one with Houston).

One last table:

Bottom 10 Cornerbacks in YAC Allowed, 2010
Player Team Charted
Targets
Yd/Pass Rk Avg. Pass
Distance
YAC Rk Success
Rate
Rk
Kareem Jackson HOU 73 11.1 83 15.3 6.0 83 47% 67
Fabian Washington BAL 43 10.4 79 11.1 5.9 82 40% 79
DeAngelo Hall WAS 88 10.5 81 10.4 5.8 81 39% 82
Leon Hall CIN 65 8.7 69 14.8 5.2 80 55% 27
Rashean Mathis JAC 64 10.1 78 14.1 4.9 79 44% 72
Vontae Davis MIA 66 7.9 53 10.8 4.9 78 47% 65
Stanford Routt OAK 92 5.7 13 15.1 4.9 77 66% 3
Alphonso Smith DET 47 10.6 82 14.4 4.8 76 47% 66
Tracy Porter NO 50 7.5 46 10.5 4.7 75 54% 33
Darrelle Revis NYJ 57 5.6 11 16.5 4.7 74 70% 1

Complete numbers, updated to account for plays with double coverage, will be in Football Outsiders Almanac 2011 this July (or whenever the owners and players decide to make nice).

Posted by: Aaron Schatz on 18 Mar 2011

48 comments, Last at 22 Mar 2011, 2:32pm by Andy

Comments

1
by SFC B (not verified) :: Fri, 03/18/2011 - 12:10pm

I figured there'd be more Texans. Did they just allow teams to score too quickly to rack up a lot of attempts against?

10
by battlered90 (not verified) :: Fri, 03/18/2011 - 3:04pm

Maybe the lack of Texans success was more due to a big hole at the safety position? Maybe the other Texans CBs are at 11-20 on the list? Two Texans CBs (Kareem Jackson and Jason Allen) are on the lists.

17
by theshadowj :: Fri, 03/18/2011 - 6:43pm

Glover Quin was decent enough to not be on these lists, and I doubt any other CBs had enough attempts to qualify.

26
by Mr Shush :: Sat, 03/19/2011 - 3:09pm

Right. Brice McCain was pretty awful, but probably missed the attempt cut-off. As for the safeties, my perception (watching every game) was that Pollard was absolutely disastrous in coverage, Wilson was adequate-ish but made no plays, and Nolan made some plays but made too many big mistakes. Most of all, though, Jackson was just every bit as bad as these numbers make him look.

2
by Dean :: Fri, 03/18/2011 - 12:24pm

Could you make the argument here that Hall was the worst corner in the league not to lose his starting job?

Given the attention he (deservedly) received around here for his poor play, I'm surprised not to see Dmitri Patterson's name on the list anywhere. Did he not qualify? Or was he just not as bad as these other names?

25
by Mr Shush :: Sat, 03/19/2011 - 3:06pm

"Could you make the argument here that Hall was the worst corner in the league not to lose his starting job?"

Not if you'd seen Kareem Jackson play.

29
by theshadowj :: Sun, 03/20/2011 - 12:21pm

Didn't he lose his starting job to Jason Allen?

33
by Mr Shush :: Sun, 03/20/2011 - 6:26pm

I think Allen just took over the nickel spot from McCain, but I could be wrong.

3
by Karl Cuba :: Fri, 03/18/2011 - 12:30pm

Look at the descent of the Carolina corners without Peppers, though Charles Tillman does pop up on the list for the Bears. In Tillman's defense, these numbers don't take account of his propensity to punch the ball out of recievers' hands.

I think that the "Dallas' cornerbacks regressed last year" meme should really be aimed at their awful, awful safeties. It's very rare for a corner to be left entirely on his own in man coverage, they will have help from at least a buzzing linebacker or single high safety the vast majority of the time, it's a team game. At the same time, I thought Jenkins was overrated last year and was due a decline of sorts.

I'd love to know if FO has any numbers for Shawtae Spencer over the course of hte season. A year ago he was the niners' best cornerback and he was pretty solid for the first half of this year too. Then he turned in one of the worst eight games I've ever seen, he was pathetic at times (Tampa and Green Bay spring to mind far too readily). I wonder if his dramatic decline was due to some kind of injury but I can't remember anything significant on that front. I'm not sure I'd like either Spencer or Clements back next year and Clements would have to agree to a massive pay cut.

4
by cisforcookie (not verified) :: Fri, 03/18/2011 - 12:37pm

am i the only one fascinated by how deangelo hall is the only guy listed with an average yards per pass greater than his average pass distance? how the hell were the redskins using him last year?

6
by Dean :: Fri, 03/18/2011 - 12:52pm

It makes me wonder if he's even worse than advertized. Maybe I'm misinterpreting the data? Seems to me like he gave up such a high rate of completions and so much YAC that even when you factor in the incompletes as giving up 0 yards, he still gives up more yards per pass the actual distance the ball travelled in the air. It's actually pretty impressive in its awfulness if I'm reading it right.

27
by Chip :: Sat, 03/19/2011 - 4:52pm

And everyone thought Jay Cutler was a prick when (in his 4 INT game against WAS) he said that he would target Hall all day long in another game.

32
by Dave0 :: Sun, 03/20/2011 - 5:29pm

he was both correct and a prick

5
by Joseph :: Fri, 03/18/2011 - 12:43pm

Aaron: considering the presence of Mr. Revis (& Mr. Routt) in the last table, maybe you should look into Adjusted YAC (AYAC). Similar to what you do with DYAR (& DVOA? or is it the other way around?) as far as capping value of a long TD run/pass, maybe you should cap YAC. In other words, is letting receiver X get behind you for a 60 yd TD pass any worse than him getting behind you for a 80 yd TD pass? Or taking a slant to the house because the FS takes a bad angle (NYJ vs. CLE)?
Suggestion #2 for AYAC--maybe, since these are charting #'s, you cap YAC against the CB at the (approximate) point where the safety blows the tackle and give the rest to the safety--esp. for coverage in a zone where the S should be playing over the top of a CB, and then doesn't get over in time because of the play-action fake (NE vs. CHI, last play of 1st half).
Suggestion #3--Considering the play just mentioned, maybe give 1/2 of the YAC to the CB, and 1/2 to the safety.
My gut tells me that for Revis & Routt, their MEDIAN would be 1 yd or more lower than their MEAN--similar to a RB whose YPC can be somewhat distorted by one or two long plays. For Mr. Hall--the numbers just show that when he didn't intercept the pass, he was just plain horrible.

7
by BJR :: Fri, 03/18/2011 - 1:07pm

For whatever reason I always enjoy the 'worst of' articles more than the 'best of'.

8
by MilkmanDanimal :: Fri, 03/18/2011 - 2:04pm

What would the almighty Hole in Zone's numbers for ypp/yac this year, for comparison's sake?

9
by Aaron Brooks' Good Twin (not verified) :: Fri, 03/18/2011 - 2:12pm

Looks like NYJ played a lot of Cover 0, and Detroit had no able bodies at safety.

It's kind of funny that as badly as DeAngelo Smith played, he had possibly the best defensive game this season, and the freakish stat line at the Pro Bowl.

11
by big_jgke :: Fri, 03/18/2011 - 3:21pm

Odd to see Stanford Routt pop up here as well. What a perplexing season for that guy.

12
by Anonymoose33 (not verified) :: Fri, 03/18/2011 - 3:28pm

It looks like DeAngelo was constantly looking for the pick. So either he got it for a big play or he missed it for a big play.

I thought Haslett at the beginning of the season said that turnovers were more important than yards surrendered, which would explain why Hall remained as the starter even with the horrible stats listed above.

13
by Aaron Schatz :: Fri, 03/18/2011 - 3:52pm

One thing I want to make clear. Both in this article, and in Wednesday's, people seem to be confusing "presentation of numbers" with "analysis." This isn't analysis. This is presentation of numbers. The ranking in a single statistical metric does not tell you a player's worth. You need to look at the whole picture in context. I'm honestly not even trying to do that in this article.

20
by Agamemnon :: Fri, 03/18/2011 - 8:25pm

Bill Polian, is that you?

45
by Theo :: Tue, 03/22/2011 - 12:08am

are you saying these are just a piece of the puzzle?!
I'm amazed.
But I can't look at 2 top row blues on a puzzle without thinking of a sky.

14
by Bright Blue Shorts :: Fri, 03/18/2011 - 4:28pm

Anyone who saw watched DeAngelo Hall play half a season for the Raiders in 2008 after they'd traded with the Falcons for him, knows how bad he is. Absolutely backed off nightmare. 7-yard completion after 7-yard completion. As I recall his season started off badly getting shown up by a Broncos rookie receiver and never got much better.

19
by Shattenjager :: Fri, 03/18/2011 - 7:40pm

The Broncos rookie had 9 catches for 146 yards in that game. And I really wonder if, without that game, people would still be insisting that Eddie Royal was some sort of superstar talent.

31
by dbostedo :: Sun, 03/20/2011 - 5:22pm

I don't know about folks in Denver, but I think most people think of Royal as a decent starting receiver, likely a slot guy. I don't know of anyone who would say he has "superstar talent".

34
by Shattenjager :: Sun, 03/20/2011 - 7:56pm

I don't think most non-Denver fans overrate Royal, or at least not much, but other Bronco fans often act as though he's some sort of superstar talent. Frankly, I don't think he's even a "decent starting receiver."

36
by Danish Denver-Fan :: Mon, 03/21/2011 - 6:24am

I somehwat agree. However, I think a lot of the positive attitude towards him is because he's a nice guy. Saying the right things and doing the right things - refreshing after the Brandon Marshall experience.

Now that huge rookie game (and pretty good rookie season), definitely helped him get a foot in the door, but personality goes a long way with a fanbase.

National fans might not pay attention to what he says and does, so they are better equipped for evaluating him based solely on production.

15
by MJK :: Fri, 03/18/2011 - 5:20pm

Interesting...no Patriots on either list, which is kind of surprising given that they had one of the worst past defenses in the league this season (countered only by the fact that they had decent number if INT's down the stretch).

I wouldn't have expected McCourty to be on the worst list, and am not shocked to not see him on the best list (although I wonder where he was ranked, given how much talk there was of him rivaling Suh for DROY), but I'm kind of surprised that neither Kyle Arrington nor Darius Butler cracked either of the "bad" lists...

Maybe the Pats pass defense problems really were the absolute lack of pass rush, and not the secondary...

38
by RichC (not verified) :: Mon, 03/21/2011 - 10:53am

"Maybe the Pats pass defense problems really were the absolute lack of pass rush, and not the secondary..."

I do think thats the problem, but I don't think that Butler/Arrington showing up/not showing up on the list would have any impact on whether or not that really is the problem.

This stat has no way to tell if a corner is getting targeted because hes bad, or hes getting targeted because the guy across the field is better than he is, and he breaks down first when the rush doesn't get there.

Its an interesting stat, but I don't think it really tells you anything in a vacuum.

16
by phillyangst :: Fri, 03/18/2011 - 5:30pm

Whoops! No Dimitri Patterson sightings.

"DVOA loves Philadelphia!"

18
by lionsbob :: Fri, 03/18/2011 - 7:17pm

I wonder how much of Alphonso Smith's bad ranking is from the New England game in which he was torched all game long. He was playing well (it seemed like) before that game.

21
by 57_Varieties (not verified) :: Fri, 03/18/2011 - 8:33pm

How much of this is the player, and how much the scheme. I'm not saying the Steelers had the worst CBs, but I'd guess they'd protect below-average ones better than most. Which leads me to Dallas: why couldn't they protect these guys?

22
by bill (not verified) :: Fri, 03/18/2011 - 11:06pm

Gamble and Marshall? Caro fans: I didn't see one of your games this year? What up?

Bill

35
by ammek :: Mon, 03/21/2011 - 5:15am

The yards/pass figures for both are around average. I'd guess they were playing off the receiver, focusing on stopping the big gain. What were the Panthers' safeties like? Did they blitz often?

23
by Misfit74 :: Sat, 03/19/2011 - 2:20am

I think pretty highly of Chris Gamble and Richard Marshall. I have to think a putrid team and lacking pass-rush helped them both appear on these lists (or fare as badly as they did by the charted numbers).

Though I'm not a Panthers fan, I do like to take note of other potential FAs. Seattle has needed CB help for some time and I thought Marshall would be a nice add if he does, in fact, make it to UFA status.

24
by sam :: Sat, 03/19/2011 - 9:56am

I wonder how much of Rashean Mathis's poor play was his fault and how much was the fact the Jags defense might as well have been playing with 9 guys because of how bad the safety play was this year.

--
sam! or the original sam from the old FO

28
by AnonymousD (not verified) :: Sun, 03/20/2011 - 12:03am

Not sure, but I remember Greg Cosell called Mathis a bad CB in a least one of his podcasts for Fantasy Guru this past season.

30
by morganja :: Sun, 03/20/2011 - 3:20pm

The Panthers corners were constantly playing the run. When the pass did come, there was no pass rush. I think in this case it was a schematic problem with the scheme not working. Apparently other teams use guys called defensive tackles, but John Fox isn't one for embracing revolutionary ideas.

Actually to be fair, John Fox bases his defense on strong defensive tackles. Unfortunately owner Jerry Richardson's toad, Hurney, thought it too expensive to hire defensive tackles, or professional football players in general, while preparing for the lockout which he insists in public he wasn't doing at all. That all of his actions were the same as an owner that was preparing for a lockout was purely coincidental.

37
by Raiderjoe :: Mon, 03/21/2011 - 6:33am

To answer Aaron Schatz wuestion yes, Pro Blow voters on drugs or drunk most of time. Horrible pciks they make everyt year. Many good picks too which expected. A moneky could pick P. Manning to make Por Blow but many of the Pro Blow picks are stupid.

46
by Kevin from Philly :: Tue, 03/22/2011 - 9:20am

"Por Blow" just might be the best RJism ever.

39
by CoachDave :: Mon, 03/21/2011 - 11:24am

Thank you for the validation of Jacob Lacey's suckiness.

What the TV guys saw in this kid this year always escaped my eyes.

40
by commissionerleaf :: Mon, 03/21/2011 - 1:00pm

Stanford Routt may be worth that $31.5 million, if the charting stats are to be believed. His numbers are completely insane.

Average pass distance? 15 yards. Yards per pass? 5.7 yards. YAC? 4.9. Unless I'm misreading the numbers, and yards per pass does not include YAC (which would be silly, but maybe that's how it's handled), that means he allows .8 yards in the air, on average, each time he is thrown at, over upwards of 90 attempts.

Better yet, his averages look remarkably similar to Nnamdi's partial season.

41
by Andy (not verified) :: Mon, 03/21/2011 - 10:23pm

If I had to guess, YAC was per completion (as it's generally kept that way for offensive receivers).

My own question - is average pass distance the avg. distance in the air? Where that + avg YAC equates to the average completion length? Wondering, because then you could take (YPA x Trg)/(avg dist + YAC) to get the number of completions from these.

42
by Andy (not verified) :: Mon, 03/21/2011 - 10:45pm

Ideally I think having CB passed vs. data split as yards down field (behindLOS, 1-10, 11-20, etc) as commonly found with WR, then taking a Vertical Y/A (yards/att for passes past 10 yards down field). I know KC Joyner of ESPN does this. It's easy to replicate VYA for WR since that data's readily available (well, SI & other STATS using sites keep WR attempts-split for the current year, nowhere free has the attempts in their splits for past years), but almost impossible for DBs unless you are the one charting the data.

One way to "better" YPA, would be to look only at completions+total PD, and divide yards allowed by that. I say this, because most incomplete passes that make up the total targets are likely the fault of the QB, not anything special the DB did. For WRs, catchable balls usually include just their REC + drops. It would be more complicated with DBs, but without actual film you can't really argue "he covered him so well the QB didn't want to throw it in close, and the ball ended up uncatchable -- but it's still because of the DB". So yards/(cmp+PD), to find the average amount of yards allowed every time there was a chance of a completion & not giving benefit to guys who just played crappy QBs. Granted, this would not favor a guy like Nnamdi whose value comes from not being thrown at - but without film to determine when he actually has a guy locked down so well a target would be stupid (on the QB's part) or when the QB didn't bother because somewhere else was so broken down it was a good place to go**, it's impossible to correctly determine his value. Really, coverage talent in general is probably the toughest aspect of football to express in metrics.

**Personally, I think it's a combo of the latter and the fact that Nnamdi plays RCB where mainly #2 WRs line up. QBs would rather go to their #1 being covered by OAK's #2 CB, or whoever their LBs are covering (I don't know about '10, but in 09 all 3 were terrible in coverage) -- as Revis is just as good and he still was targeted 153 times over the past season.

43
by Andy (not verified) :: Mon, 03/21/2011 - 11:06pm

Ok, after looking at some data from last year (from another site), and taking Yards Allowed - total YAC /receptions allowed (ie: average distance in air per pass-completion), the HIGHEST among 107 CBs was 12.6, I am beginning to think that the "Avg. Pass Distance" you list DOES INCLUDE average YAC? As 23 of the 43 guys listed on the Best & Worst CB lists have an APD above 12.6.

However, this makes D. Hall's APD being lower than his YPA confusing (and if I'm not mistaken, physically impossible). Because if you take (YPA*Att)/yards-per-completion, you get more completions than targets.

44
by tuluse :: Mon, 03/21/2011 - 11:39pm

I think your logic is flawed. Even if the worst CB in the league allowed 12.6 yards per completion, he's not allowing 100% completion rate. So incomplete passes could easily bump it up 3 yards per attempt.

47
by Andy (not verified) :: Tue, 03/22/2011 - 2:29pm

How is that related to my last post in the slightest? The first paragraph was just in regard to what they meant by Avg Distance (was it including YAC, or just yards in the air? Based on numbers from 09 that I looked at, I'm inclined to believe it includes YAC). The second part points out that one problem if it DOES include YAC, then DeAngelo Hall's Avg Distance (including YAC) would was lower than his YPA allowed - and this is not possible (Yards/completion should be higher than yards/attempt because you are dividing yards by a bigger number).

48
by Andy (not verified) :: Tue, 03/22/2011 - 2:32pm

**correction: dividing it by a SMALLER number (less completions than attempts)