Writers of Pro Football Prospectus 2008

27 Jul 2016

Worst CB Charting Stats, 2015

by Carl Yedor

As promised on Monday, we will conclude our look at cornerback charting stats today with a discussion of some of the worst performing cornerbacks from the 2015 season. Again, here is a reminder about the changes to our charting in 2015. In the past, our charting came from a combination of ESPN Stats & Info data and volunteer charting. In 2015, instead of volunteers, we partnered with Sports Info Solutions to do the portions of our charting that don't come from ESPN, including charting of pass coverage and all the metrics below.

Again, the main statistics we will be looking at today, adjusted for the quality of the opposing receiver:

  • Yards per pass, the average yards allowed by the corner when targeted.
  • Success rate, the percentage of targets where the corner prevented a successful gain (45 percent of needed yards on first down, 60 percent on second down, and 100 percent on third or fourth down).
  • Estimated Target Rate, the percentage of possible targets with the corner on the field where the corner was targeted in pass coverage by the opposing offense.

We will again be looking at plays where the cornerback was the primary defender in coverage, leaving out a few types of plays that do not properly reflect a cornerback's ability. These plays include screens, Hail Marys, balls tipped at the line or thrown away, and plays where the quarterback was hit while throwing the ball. The only defensive penalty included is defensive pass interference, and the "possible targets" from estimated target rate leave out the aforementioned omitted plays in addition to passes that were marked "uncovered" or "blown coverage." However, if a play was marked as "hole in zone," the pass play was included.

Traditionally, cornerback statistics are very volatile from year to year, and the best cornerbacks will often have worse results than expected because quarterbacks will be unlikely to target them unless they have already made a mistake. Seventy-five players met our benchmark of either starting eight games or facing 50 passes to be included on the list. One more thing to note: with the number of players ranked, the difference between ranking 50th and 60th is not all that large and is therefore not necessarily an indicator of a huge gap in talent.

In an attempt to partially solve the past problem where nickelbacks frequently ranked much higher than expected in these metrics, we made some changes to the adjustments in the statistics. In the past, adjustments for offensive Team X were done by comparing Team X's No. 1 receiver to all No. 1 receivers, Team X's No. 2 receiver to all No. 2 receivers and Team X's "other" receivers to all "other" receivers. In 2015, we adjusted instead by comparing Team X's No. 1 receiver to ALL wide receivers across the league. Hopefully, this will do a little bit more to penalize corners who only cover slot receivers and boost the adjustments given to corners who primarily cover the opposing team's top receiver.

And now, we will look at the worst players by adjusted yards per target.


20 Worst CBs, Adjusted Yds/Target, 2015
Defender Team G GS Passes Rec Adj Yds Rk
39-B.Browner NO 16 16 69 39 11.0 75
24-B.Flowers SD 11 10 47 28 10.6 74
20-D.Butler IND 14 2 53 35 10.4 73
21-B.Grimes MIA 15 15 75 48 9.8 72
31-B.Maxwell PHI 14 14 72 42 9.5 71
28-G.Toler IND 10 10 76 42 9.5 70
26-C.Williams SEA 10 10 36 23 9.4 69
20-K.Acker SF 15 13 48 27 9.4 68
22-T.Williams CLE 15 15 83 42 9.3 67
41-W.Blackmon WAS 15 10 48 20 9.2 66
39-B.Carr DAL 16 16 62 39 9.0 65
31-A.Cromartie NYJ 15 15 73 33 8.9 64
41-A.Blake PIT 16 16 90 58 8.9 63
31-R.Cockrell PIT 15 7 64 37 8.9 62
23-D.Randall GB 15 9 74 38 8.8 61
23-T.Newman MIN 16 16 72 39 8.8 60
26-T.Brock SF 15 15 56 24 8.8 59
25-J.Powers ARI 13 13 60 33 8.8 58
24-M.Claiborne DAL 11 11 52 28 8.7 57
21-M.Butler NE 16 16 93 46 8.7 56
Minimum 8 games started or 50 targets.

Bringing up the rear for this category was Brandon Browner in New Orleans. This should not come as a huge surprise, given that Browner set the record for most penalties in a season. Only three of his penalties were defensive pass interference, but his general struggles in coverage resulted in an adjusted yards per target allowed of 11.0. If Browner ends up making Seattle's roster for Week 1, the Seahawks will have to hope that a change in role will allow Browner to recapture at least some of what made him a feared member of the Legion of Boom.

Following Browner in the race to the bottom were San Diego's Brandon Flowers (10.6), Indianapolis' Darius Butler (10.4), Miami's Brent Grimes (9.8), and Philadelphia's Byron Maxwell (9.5). This group speaks to the volatility of cornerback stats, given that Browner, Flowers, and Maxwell were all mentioned on the lists of either best primary corner or best supporting corner in 2014.

Maxwell in particular did not transition well from his supporting role in Seattle in 2014 to covering receivers like Julio Jones with the Eagles in 2015. After signing Maxwell to a splashy free agent contract under the leadership of Chip Kelly, the Eagles traded Maxwell to the Dolphins in a package for the eighth pick in the 2015 draft. Maxwell will not likely produce at a level commensurate with his salary, but the Dolphins will certainly hope to put Maxwell in a situation to succeed this season. Miami didn't receive Earl Thomas, Richard Sherman, and Kam Chancellor in the deal as well, but Maxwell's true talent level probably lies somewhere between the two extremes of 2014 and 2015.

Flowers was one of the best leading cornerbacks of 2014, so his drop-off in play was not a situation like Maxwell's where he was asked to play as a No. 1 corner when he had only ever previously played in a supporting role. In fact, Flowers finished just behind Sherman in adjusted success rate in 2014 by a fraction of a percentage point, making his drastic decline in performance from top ten to bottom ten more surprising than Maxwell's.

After Brent Grimes' wife Miko delivered her assessment of Ryan Tannehill's play on Twitter during the season, no one would have been surprised if he had been released in the offseason. Objectively, this was also a good decision from the perspective of Brent Grimes' quality of play. Grimes was effectively replaced by Maxwell in Miami, linking another one of 2015's worst cornerbacks by yards per target allowed to the former Eagle.

Darius Butler presents a slightly different case, as the seventh-year player from Connecticut was not a big-money free agent signing. His current contract only carries a $500,000 dead money hit if Butler is released, so if the Colts feel they have better options available to them at the cornerback spot across from Vontae Davis, it would be easy for them to move on from Butler without much trouble.

Next, we will take a look at the worst cornerbacks by adjusted success rate.


20 Worst CBs, Adjusted Success Rate, 2015
Defender Team G GS Passes Rec Adj Suc Rk
25-D.Hayden OAK 16 13 75 45 38% 75
25-K.Jackson HOU 12 10 52 34 40% 74
41-A.Blake PIT 16 16 90 58 40% 73
24-B.Flowers SD 11 10 47 28 40% 72
21-B.Grimes MIA 15 15 75 48 42% 71
20-D.Butler IND 14 2 53 35 43% 70
29-P.Cox TEN 13 13 50 29 43% 69
26-C.Williams SEA 10 10 36 23 44% 68
39-B.Browner NO 16 16 69 39 44% 67
26-S.Moore TB 16 9 47 27 46% 66
31-R.Cockrell PIT 15 7 64 37 46% 65
31-B.Maxwell PHI 14 14 72 42 46% 64
24-M.Claiborne DAL 11 11 52 28 47% 63
41-B.Skrine NYJ 16 8 67 41 47% 62
20-K.Acker SF 15 13 48 27 47% 61
29-X.Rhodes MIN 16 16 76 40 47% 60
24-C.Sensabaugh TEN 16 15 70 38 47% 59
23-N.Carroll PHI 11 11 53 33 47% 58
23-D.Randall GB 15 9 74 38 48% 57
39-B.Carr DAL 16 16 62 39 48% 56
Minimum 8 games started or 50 targets.

As promised on Monday, here we have Pittsburgh cornerback Antwon Blake checking in with the third-worst adjusted success rate among qualifying cornerbacks. With an adjusted success rate just over 40 percent, Blake almost canceled out all the good that teammate William Gay was able to do on the other side of the field. Blake will be taking his talents to Tennessee this season on a one-year contract, so hopefully he will be able to provide more value for the Titans in 2016.

Flowers and Grimes also both made appearances in the bottom five of adjusted success rate, at 40 percent and 42 percent, respectively. While Grimes was released and was subsequently signed by Tampa Bay, Flowers remains with the Chargers on a contract that theoretically runs through the 2018 season. San Diego can't realistically cut Flowers until after this season because of how much guaranteed money he is still owed, so this season may be Flowers' last chance to prove that he deserves to stick around in San Diego. It would not be surprising if Flowers is asked to take a pay cut at the end of the year if San Diego wants to keep him on the roster.

Former first-round pick D.J. Hayden holds the dubious distinction of having the worst adjusted success rate among qualifying corners at 38 percent. Hayden enters the last year of his rookie deal in 2016 without future financial security, given that Oakland declined to exercise his fifth-year option. Fortunately for Hayden, if he performs well this year, he will be able to test the waters of the free agent market at a premium position.

Finishing only slightly better than Hayden was Houston's Kareem Jackson, who posted an adjusted success rate just shy of 40 percent before rounding. Normally having a starting corner perform so poorly would result in a defense that would not finish eighth in defensive DVOA. Of course, normal defenses do not employ J.J. Watt.

Finally, we will take a look at the cornerbacks who opposing quarterbacks attacked the most.


20 Worst CBs, Estimated Target Rate, 2015
Defender Team G GS Passes Rec Est Tgt% Rk
28-J.Bethel ARI 16 5 54 26 32.8% 75
24-N.Lawson DET 15 9 58 30 28.9% 74
28-G.Toler IND 10 10 76 42 27.2% 73
29-B.Roby DEN 16 4 61 33 26.9% 72
29-D.Amerson WAS/OAK 16 12 91 43 26.8% 71
23-D.Randall GB 15 9 74 38 26.6% 70
22-M.Peters KC 16 16 116 52 25.8% 69
24-S.Gilmore BUF 12 12 85 41 25.1% 68
31-D.House JAC 16 15 91 47 24.6% 67
24-M.Claiborne DAL 11 11 52 28 24.2% 66
22-T.Johnson STL 14 13 69 36 24.1% 65
41-A.Blake PIT 16 16 90 58 24.0% 64
24-J.Joseph HOU 16 16 77 39 23.6% 63
28-R.Darby BUF 15 15 91 45 23.2% 62
22-T.Williams CLE 15 15 83 42 23.0% 61
26-B.Breeland WAS 15 14 83 44 23.0% 60
31-R.Cockrell PIT 15 7 64 37 23.0% 59
21-T.Porter CHI 14 13 66 32 22.8% 58
21-B.Grimes MIA 15 15 75 48 22.6% 57
20-D.Butler IND 14 2 53 35 22.5% 56
Minimum 8 games started or 50 targets.

In a bit of a bookend to our discussion of Patrick Peterson's fantastic year down in Arizona, one of his Cardinals teammates suffered through a rough season. Justin Bethel was targeted on 32.8 percent of possible passes in his time both on the outside and in the slot. After Tyrann Mathieu's ACL injury forced Bethel into even more extended action, we got to see part of why teams avoided throwing to Peterson whenever possible.

Bethel beats the field by a large margin as the most-targeted corner, with Detroit's Nevin Lawson coming in second at 28.9 percent. For reference, Lawson was closer to Buffalo's Stephon Gilmore in eighth than he was to Bethel in first. Lawson was not terrible by any means, as he finished seventh in adjusted yards per target in his role as a nickel corner, but teams still consistently went after him.

Greg Toler in Indianapolis was third-most targeted, and this should not come as a huge surprise for the same reason that Bethel was so "high" on this list. Vontae Davis has been an excellent performer for the Colts for the past few seasons, so it makes sense that opponents would prefer to go after his less talented teammates.

In fourth we have Bradley Roby from Denver, which again makes sense because the other corners in the Denver defensive backfield in 2015 were Chris Harris and Aqib Talib. Even with all those additional targets, Roby finished in the top 30 in the other two metrics, though that probably had at least a little to do with their studs in the front seven.

Rounding out the top five, we have a player who changed teams and still couldn't catch a break with how frequently he was targeted. David Amerson was cut by Washington in September and then signed with Oakland shortly thereafter. It was interesting that teams chose to go after him as frequently as they did, given that he finished 11th in fewest adjusted yards per target and that D.J. Hayden was in the same defensive backfield.

Posted by: Carl Yedor on 27 Jul 2016

17 comments, Last at 11 Aug 2016, 11:18am by Will Allen

Comments

1
by PirateFreedom :: Wed, 07/27/2016 - 1:28pm

Miko needs to make an FO account and straighten this 42% stuff out

2
by Will Allen :: Wed, 07/27/2016 - 2:32pm

I have a nagging suspicion that when a metric has this much volitility, in terms of who played well, and who played poorly, from year to year, the metric is of limited utility with regard to how these players actually rank with each other. I really don't believe the quality of play varies that much from year to year.

6
by Jerry :: Thu, 07/28/2016 - 8:07am

That's fair, but I think we're still at the point where there's value to learning that a particular metric is of only "limited utility."

8
by ChrisS :: Thu, 07/28/2016 - 4:02pm

That may be true. But there is a lot of variability in year to year win/loss records and while wins are not the best measure of quality they do point to a lot of season to season change in team (made of individual) performance.

11
by Scott C :: Tue, 08/02/2016 - 2:32pm

No. Its fine.

Take Brandon Flowers for instance. He DID play well one year and suck the next. It wasn't some fluke of the metric. Many other metrics agree, or subjective things like PFF's rankings. As a fan, Flowers had many bad breakdowns in coverage last year that he did not before, letting WR's get behind him deep. He was supposedly playing injured some of the year (and was out for several games), and also supposedly did not come into camp in great shape at the start of the year.

Although yes, the variability in the metric makes it less predictable, that doesn't imply it is bad at describing how players played. Flowers played great (and was used well) in 2014, and in 2015 he struggled in a few key ways, leading to the stats here (bad success rate and bad yards/pass his way) but not highly targeted -- he could still cover some underneath routes fine and was picked on with certain route combos or when in certain coverages, not in general.

What is hard with these stats is interpreting them in context -- how a CB is used, in what types of coverage, against what opponent, etc, make a big impact.

15
by Will Allen :: Wed, 08/10/2016 - 12:33pm

Why don't we see the same degree volitility in other sports, where the performance of the athletes is more easily quantified? Why is there a lot more stickiness among NBA players? Among MLB hitters?

16
by LionInAZ :: Wed, 08/10/2016 - 8:02pm

My guess is that 1) MLB and NBA players see fewer opponent matchups; 2) see matchups more often; and 3) don't depend as much on teammate performance except for team wins.

17
by Will Allen :: Thu, 08/11/2016 - 11:18am

MLB hitters see more opponent matchups, and see matchups less often. Point 3 agrees with my suspicion, which is that the cornerback metric really isn't measuring what it claims to be measuring.

3
by dryheat :: Wed, 07/27/2016 - 4:55pm

Wow....at least once upon a time Grimes had skills that made him worth putting up with his wife. Now? He doesn't seem to be worth the distraction and disharmony that bat-shit crazy woman brings.

7
by Noah Arkadia :: Thu, 07/28/2016 - 10:57am

Completely agree. It could be coincidence, but back when Grimes was playing well for Miami Miko was barely news. Some people found her adorable even. It's when Grimes starting playing poorly that she turned into the harpy from hell. Good luck with that hand grenade, Bucs. Fair warning, it's unpinned.

4
by lokiwi :: Wed, 07/27/2016 - 7:44pm

Rhodes managed to be top-20 on adj. yds/tgt while being bottom-20 in adj. success rate. Is that just a result of a lot of short yardage first downs? Or does success rate flatten out enough in the middle rankings that it wasn't that big a handicap to overcome?

5
by NWebster :: Thu, 07/28/2016 - 6:55am

Any idea when the charting results will be available for download?

9
by dmepolitic :: Sun, 07/31/2016 - 10:32pm

After consistently seeing and commenting about the gap between the tape and media perception of Butler last season. It makes sense that his charting stats are underwhelming. I understand that he gets tough assignments, but he was just burned way too frequently last year to be a legitimate pro bowler (I get that pro bowl voting is normally a weak measuring stick). I think the adjusted/yards target captures that well. I think it is a good corrective to all the hyperbole around butler (which is partly based on a single high profile play).

As always great job football outsiders providing revealing metrics, looking forward to reading your analysis again this season.

10
by joelius@gmail com :: Tue, 08/02/2016 - 2:10pm

True, M. Butler was one of the bottom-20 CBs last year in Adjusted Yds/Target.

But he was also in the top 20 as measured by Adjusted Success Rate:

http://www.footballoutsiders.com/stat-analysis/2016/best-cornerback-stat...

Maybe that averages out to middling, I don't know. Either way, he'll never pay for another meal in New England the rest of his life. Malcolm Butler!

12
by jefeweiss :: Thu, 08/04/2016 - 2:01pm

Interesting that Washington had two cornerbacks that showed up on both the Best and Worst Cornerbacks. You mention David Amerson, but Will Blackmon was on the chart for the worst for Adjusted yards per target (66th) and the chart for the best for Adjusted success rate. (18th)

14
by BigNose :: Wed, 08/10/2016 - 12:04am

Will, and the rest of that unit, got destroyed by deep plays several times, most notably against the Bills and Giants. For instance, Odell's annual circus catch was against Blackmon, although there probably isn't a CB anywhere who could have covered any better. It was just mission impossible with that throw to that receiver. 0:58 in this video. https://www.youtube.com/watch?v=JLPh6_81LtY
But they also made a lot of money plays themselves, and the Skins D were I believe 11th in the NFL in takeaways. Blackmon personally had in some ways a career year in 15, ending the season with the mother of all punchouts in the finale against Dallas, at 2:40 or so in this video. http://www.dallascowboys.com/video/2016/01/03/highlights-redskins-vs-cow...
So a little up and down early on, but someone Skins fans had little reason to complain about at the end of last year. Weird to see him on a ranking of the worst CBs in the NFL tbh.
Amerson though was all bad in DC,and if you want to put a clown nose on him for his early performances in 2015 you're welcome to it. He more or less lost the season opener to Miami on his own, which is why he was cut shortly thereafter. https://www.youtube.com/watch?v=iDKzllfKY3M Enjoy the stellar efforts of #39! Apparently he turned a corner in Oakland and landed a big contract, so something definitely happened to wake him up. Maybe being cut is good for some people.

13
by fredtoast :: Thu, 08/04/2016 - 10:28pm

Where are the "adjustments" made that reward a CB for covering a #1 WR instead of a slot receiver. I don't understand what that means.