Writers of Pro Football Prospectus 2008

Most Recent FO Features


» Futures: Nick Chubb & Sony Michel

The Georgia Bullddogs' dynamic duo should be on NFL rosters at some point in the next 72 hours. Which will be the better pro? That depends on what kind of running back you're looking for.

27 Dec 2005

Any Given Sunday: Bills over Bengals

by Ned Macey

If defense wins championships, the Cincinnati Bengals can stop planning for a trip to the Super Bowl. Of course, only 23 of the 37 points scored by Buffalo were actually scored by the offense, but the defense's poor showing is still reason for concern. When the Bengals host their first playoff game in over a decade, they will feature an offense that is among the league's best. Their defense will only contribute as much as their ball-hawking abilities allow.

It might surprise some that in Football Outsiders' DVOA rankings (explained fully here), Cincinnati has the 13th best defense in the league. The official NFL rankings have the Bengals 22nd in yards allowed and 19th in scoring defense. DVOA ranks them higher primarily because of their ability to force turnovers, particularly interceptions. Those interceptions not only prevent points scored and yards gained but set up the offense with short fields, a contribution not measured in the traditional stats.

For most of the season, Cincinnati has struggled against the run: the Bengals rank fourth against the pass but 26th against the run. This has changed over the past four weeks as the Bengals have aggressively stopped the run but fallen victim to big pass plays.

The emphasis on stopping the run started, perhaps not coincidentally, before their division-deciding game against Pittsburgh. Since that time, they have been above average against the run in every game according to DVOA. Before the Steelers game, they had been above average only twice all season and zero times between Week 4 and Week 12.

Their new commitment to stopping the run has taken a toll on the Bengals pass defense, which has been below average three of the last four weeks. Their ability to get interceptions allowed them to escape with a win over Pittsburgh despite 386 yards from Ben Roethlisberger. On Saturday, the Bengals managed only one interception.

Without interceptions, the Bengals defense is severely flawed. They rank in the bottom ten in the league in yards allowed, completion percentage, and sacks. Giving up more big plays in order to make interceptions is a strategy that has paid off for most of the season but backfired on Saturday.

If we remove interceptions from the equation for all teams, the Cincinnati defense's DVOA drops from 13th to 30th. No other team is even close to being as reliant on interceptions; the second largest drop-off is only seven spots.

Football Outsiders has written extensively about the flukiness of fumble recoveries, but we have never asserted that causing fumbles or interceptions is not a skill. Over the course of a season, Cincinnati's aggressive defense has picked off a league-high 31 passes. But when the interceptions do not come, the Bengals let players like Kelly Holcomb suddenly look like they belong in the Pro Bowl.

On Saturday, Holcomb completed 24 of 31 passes for 308 yards. He went often to Eric Moulds, who hauled in 10 receptions. For the first time this season, Holcomb effectively used deep threat Lee Evans, who had 107 yards receiving including a 65-yard catch and a touchdown. The Bills are now 3-3 in games where Holcomb throws at least 10 passes and 2-6 in all other games.

That sounds like a problem with an easy answer: make Kelly Holcomb the starter. But in Buffalo, with former first-round pick J.P. Losman competing with Holcomb, a switch is more difficult. It is a little premature to give up on such a major investment after a handful of starts. Holcomb has limited upside, and the Bills have to decide whether to try and win now with him next season or rebuild the aging defense and allow Losman time to grow.

The problem with keeping Holcomb is that, despite the win on Saturday, this defense has collapsed. Their run defense is 29th in the league according to DVOA, a fact well illustrated on Saturday when Rudi Johnson ground out 88 yards on only 18 carries. The return of Takeo Spikes next season will help, but by itself it won't reverse the defense's fortunes.

One place where Buffalo is not lacking is in its kick return game. Terrence McGee scored his first touchdown, marking his eighth return of at least 40 yards. He is averaging over 30 yards a return.

At Football Outsiders, we calculate the number of points above average that special teams units contribute. The Buffalo kick return team has contributed an impressive 27 points worth of field position. Only Houston -- which with Pro Bowl rookie Jerome Mathis has contributed 27.4 -- has a more effective kick return unit. After those two, the next highest unit adds only 7.2 points. In front of McGee, the Bills feature the best special teams in football for the second consecutive year.

The Bills' season of disappointment leaves the team facing as many questions this off-season as anyone in football. A win this week over a bad New York Jets team would allow them to finish at 6-10, hardly the record a team with playoff aspirations was hoping to achieve. Questions abound about management, coaching, and players. Whether the same group or a new one comes in, the decision about next year's quarterback is a difficult one.

Even if management decides to start from scratch, Losman is as likely to be Joey Harrington as Steve McNair. For a team that has not made the playoffs this century, the prospect of rebuilding with an uncertain quarterback situation is very troubling. If Losman fails to develop over the next two seasons, fans will not only remember fondly the disappointing Drew Bledsoe years, they may start calling for a return of Rob Johnson. If Buffalo can develop an above-average defense and maintain its dominant special teams, going with Kelly Holcomb allows them a chance to be competitive, and when Losman proves himself in practice, they can make the transition to the younger player.

The Bengals hope to explain this game away as a game they controlled but lost on two exceptional plays by McGee, the kick return and a game-clinching interception returned for a touchdown. They did gain more total yards than Buffalo and got more first downs. Still, they were far from dominant, staying in the game thanks to surprisingly stout red zone defense. Three times they held Buffalo to a field goal inside the 10-yard line despite being among the worst red zone defenses in the league.

The Bengals finish the season with a very difficult game at Kansas City. With a playoff berth already secured, they will likely rest some regulars and fall to the Chiefs. A loss coupled with a win by New England over Miami would put the Bengals into the fourth seed hosting Jacksonville in the first round of the playoffs. Such a match-up is problematic for the Bengals because Byron Leftwich throws very few interceptions, only five in over 300 attempts this season.

If Leftwich is fully healthy, Jacksonville poses a serious threat to the Bengals. They would be best served by returning to their early-season strategy, stopping the pass and allowing Fred Taylor to get his yards. This defense is still a work in progress and is only strong enough to limit one aspect of an opponent's attack. Between stopping the run and stopping the pass, they are better off stopping the pass to prevent yardage from coming in chunks. Pressuring the box and allowing Leftwich to make plays down the field will lead to a quick exit for the Bengals in their long-awaited return to the playoffs.

Each Tuesday in Any Given Sunday, Ned Macey looks at the biggest upset of the previous weekend. The NFL sells itself on the idea that any team can win any given game, but we use these upsets as a tool to explore what trends and subtle aspects of each team are revealed in a single game.

Posted by: Ned Macey on 27 Dec 2005

26 comments, Last at 31 Dec 2005, 8:24pm by Ryan Mc


by Craig B (not verified) :: Tue, 12/27/2005 - 3:04pm

This is the first Bengals' game I was unable to watch this year. Is there a reason why they didn't run the ball more often? Rudi averaged 4.9 yard per carry, but only had 18 carries. Was it inconsistency or Bratkowski forgetting about the run again?

by pawnking (not verified) :: Tue, 12/27/2005 - 3:15pm

#1 brings up a potentially interesting article for you FO guys. Is there any measurable stat for playcalling? Looking at this game, you can say that with a better run/pass mix on the part of the Bengals, they would have won, both because of the extra yardage and the less time on defense for the Bengals.

Is there a way to analyze the play calling on other teams' part? Let's say that we all know Andy Reid should have run the ball more this year, but is there a way to quantify how much better it would have been had he done so? Using DVOA, can we look at their success per pass vs. run and project how much better their offense would have been if they had run more?

What about an analysis on the optimal mix of run vs. pass? I believe that the average yardage per pass play is greater than per run play, but are there diminishing returns? Do potential turnovers play into playcalling? Are there downs where calling a run is greatly better than a pass? Vice versa? How can DVOA be used to answer these questions?

by Craig B (not verified) :: Tue, 12/27/2005 - 3:36pm

Possibly answering my own question:
Rudi's longest run was 13 yards, so I doubt inconsistent running was the reason. I'd bet on Bratkowski forgetting about the run once again. He's done it often the past three years.

Disclaimer: Yes, I know that despite his shortcomings he's been a good, if not great, offensive coordinator for the Bengals.

by Matthew Furtek (not verified) :: Tue, 12/27/2005 - 4:13pm

I'm just gonna make a general statment here and say... playcalling is overrated compared to execution...

Playcalling is important, but did play calling cause Palmer to throw an interception instead of throwing the ball away? Did play calling 'cause the special teams coverage to not tackle the return man? Play calling doesn't cause incompletes...

by Craig B (not verified) :: Tue, 12/27/2005 - 4:20pm

Re #4:
In no way am I saying that playcalling caused them to lose. I've known for a long time that they have a poor defense that is very susceptible to the big play. I'm just curious as to why they didn't hand the ball to Rudi more against one of the worst rushing defenses in the league according to DVOA, especially considering the fact that Rudi appears to have had success (4.9 ypc) running the football Saturday.

by Bencoder (not verified) :: Tue, 12/27/2005 - 4:38pm

#2 I second that question. It would be a very interesting analysis, although it sounds like a lot of work to pull off. Perhaps an off-season project?

by Sid (not verified) :: Tue, 12/27/2005 - 5:14pm

What I find amazing is that Deltha O'Neal is in the Pro Bowl. He's just a gambling CB who sometimes gets the INTs, but more often gives up big plays and is burned by his gambling.
O'Neal and James are not that good. They just look good to the morons that measure CBs by number of INTs.

by Sid (not verified) :: Tue, 12/27/2005 - 5:18pm

The Bills are in a tough position. Their idiotic GM Tom Donahoe put them in this position by wasting a first rounder and a 2nd rounder on JP Lostman. It's hard to rebuild around a guy who sucks, but it's also hard to give up so quickly on a guy you have so much invested in.

by Sid (not verified) :: Tue, 12/27/2005 - 5:23pm

Ned, why didn't you mention the possible Cincy-Pitt rematch? It's about as likely as a Cincy-Jax matchup.
I would expect NE to beat Miami and KC to beat Cincy, but maybe BB will do the smart thing and rest starters.

RE: 2

It's a fact that passes average significantly more per attempt than rushes, but passing exclusively may lead to a lot of 3 and outs (not to mention turnovers).

by Kibbles (not verified) :: Tue, 12/27/2005 - 6:35pm

Re #7: Absolutely. Remember, this isn't O'Neal's first pro bowl. He made one early on in Denver off of the strength of a high INT total. A couple of years later, Shanahan had switched him to WR and then sent him packing. I've been surprised with how well he's been playing in Cincinnati (meaning that he's gotten the INTs back), but he's not a pro bowl CB. He's simply a better version of Phillip Buchanon. If they throw to his side of the field, you know that one team is going to get a HUGE play, you just don't know which team.

Also, this was a really insightful article. Probably the best AGS yet. I didn't know that Cincy wasn't still getting gouged by the run, and it's really neat to see how well it coincides with their fall from grace against the pass. It really illustrates the whole "stacking the box and making them beat you with the pass" concept. Also, I understand that Buffalo and Houston got so many points from kickoffs because they give up so many TDs, but that's still incredibly impressive, that their kick return units are worth 20 points more than the #3 unit. Great job.

by Sean (not verified) :: Tue, 12/27/2005 - 7:18pm

I'm going to have to disagree on O'Neal, as his coverage abilities have been vastly underrated as a result of his rocky tenure in Denver. Since he's been in Cincy, O'Neal has played very well, and it goes beyond simply making interceptions. Cincinnati is #1 in the league against an opponent's #1 receiver, with a -44.2% DVOA. Not only is O'Neal picking off passes, but he's also not giving up receptions that convert into first downs. KC Joyner, who uses a completely different set of metrics to measure DB play, things like completion percentages at the various depth ranges and percentages of tight coverage, also singled out O'Neal as someone who has been one of the best corners in the league since he came to Cincy.

by James, London (not verified) :: Tue, 12/27/2005 - 9:21pm

Nice AGS Ned. I know rookies aren't normally the solution to a problem, but is there a DL in the draft who the Bengals might have a shot at that could upgrade the run D next year?

by Sean (not verified) :: Tue, 12/27/2005 - 10:23pm

It's not a particularly good draft for DTs, but there are always guys to be had who can at the very least fit into a rotation. Really, fixing the team is a three or four year process, and the Bengals are well ahead of schedule.

by Sid (not verified) :: Wed, 12/28/2005 - 12:02pm

RE: 10

Yes, O'Neal has always been a gambler. He went to a Pro Bowl or two in Denver, but Denver traded him because they knew he can't cover.

RE: 11

It only looks good because of the large adjustment for INTs. Like Ned suggested, if you took out INTs for all CBs and just looked at it that way, I suspect the Cincy DVOA against #1 WRs would be positive.

by Sid (not verified) :: Wed, 12/28/2005 - 12:05pm

RE: 12

Rodrique Wright out of Texas or Haloti Ngata out of Oregon are the most highly rated.

by Lou in Cincy (not verified) :: Wed, 12/28/2005 - 12:19pm

re #7, Deltha doesn't just intercept passes, he's intercepted TEN of them, which leads the NFL and sets a Bengals record. It's Tory James who keeps getting burned, he fell down on that long pass to Evans.
re # 12, In Marvin we trust... considering the Bengals track record in the draft pre-Marvin Lewis, one or two good players a draft would be a joy. This years class has at least 4 good players contributing. Heck, even Chris Perry turned out to be a find, and boy did we give him crap for that one! Marvin has turned out to be a darn good talent evaluator.

On the other hand, are there any likely free agents at DT who could help us out?

by Sid (not verified) :: Wed, 12/28/2005 - 5:38pm

RE: 16

Deltha O'Neal intercepted 9 passes in 2001 with the Broncos. Tory James intercepted 8 passes last year for the Bengals.

by Lou in Cincy (not verified) :: Wed, 12/28/2005 - 7:57pm

re:16 All I'm saying is if Deltha had 6 or 7 Ints and that was all he did then #7 might have a point. As it is, 10 picks is a lot, and deserving of praise.

Tory played great last year, but he has been burned a few times this year

by thad (not verified) :: Wed, 12/28/2005 - 9:41pm

According to DVOA the Bengals defense is 5th against the pass and 26th against the run
If you go check out two minute warning
the Bengals have the best cornerbacks in the league

by dave (not verified) :: Wed, 12/28/2005 - 9:42pm

Regarding the Bengals' improved run D (and worsened pass D) over the past several games, one wrinkle not considered is the injury to DT Bryan Robinson. The 290-something-pound former Chicago defensive end went out 5 weeks ago with a foot injury (and looks to be back for the wild card game), and was replaced with 320-something-pound run-stuffer Shaun Smith.

In the last 5 games, Smith has recorded as many tackles (18) as Robinson did in his 10 games. The guys playing on either side of him, DE Justin Smith and DT John Thornton, as well as the guy right behind them, MLB Odell Thurman, are also posting slightly better numbers (and lots more assists, relative to total tackles, suggesting they're getting in on more plays) over the last five games.

Overall, the Bengals d-line has been a bit more than 2 tackles/game better (mostly attributable to Shaun Smith) over the last 5 games versus the first 10. That would likely jump them up a few notches in the run defense department right there.

by Sid (not verified) :: Thu, 12/29/2005 - 12:02am

RE: 19

Far from it. There is a lot more to playing CB than getting INTs. INTs are few and far between even for players like Deltha O'Neal (he gets about 1 every 2 games). What about all the other times the CB is thrown to?
The Bengals have a pair of CBs that gamble way more than any other pair of starting CBs. They are far from being the best in the NFL. Personally, I'd take Samari Rolle and Chris McAlister of the rival Ravens.

by Sean (not verified) :: Thu, 12/29/2005 - 7:47pm

Sure there is more to playing CB. But an INT stops an offensive drive cold. Considering that each drive averages out to a team scoring between 1.03 and 2.69 points per drive, each time you can decisively end an offensive drive with zero points, you are making about as significant a play as it's possible for a defender to make. Does that mean that corners should be judged strictly by how many picks they have? No, absolutely not. But I don't see the point of trying to judge how O'Neal plays without the interceptions; it's like only taking statistics for the Eagles defense when they aren't blitzing- both player and team deliberately take a high risk/high reward approach, and they should be judged on what the overall effect is.

by Sid (not verified) :: Thu, 12/29/2005 - 8:32pm

I'm just saying that he isn't a great cover corner. He generates more INTs than other CBs, which is great, but I don't see him as an elite CB.

by thad (not verified) :: Thu, 12/29/2005 - 9:17pm

What the two minute warning site does is look at every pass thrown to the top two wide recievers in each game. Then they calculate the passer rating on just those throws. Also pass interference is counted as a completed pass. So far this year the top two WR's of the Bengals opponents have only 8 td's which tells me that O'Neil is not getting burned that much.
Here are some other years
2002 Bucs 53.2
2003 Pats 39.2
2004 Bengals 57.9.
I am not saying their charts are the end all and be all of corner back ratings but I think they have some merit.

by Sid (not verified) :: Fri, 12/30/2005 - 12:38am

Doesn't DVOA do the same thing?

by Ryan Mc (not verified) :: Sat, 12/31/2005 - 8:24pm

As somebody who has watched 13 of the Bengals games this year I think both CBs have been magnificent. They are "supported" by one of the weakest pass rushes in the league (26 sacks, 4th worst in the league, and that includes an 8 sack game against Houston) and have still done a great job in coverage most of the season. It is the lack of pass rush that is catching up to the Bengals. QBs have been very comfortable in the pocket against Cincy this season, and even the best corners can't cover forever.