Writers of Pro Football Prospectus 2008

Most Recent FO Features


» The Deep Ball Project

Guest columnist John Kinsley breaks down the tape of every deep pass in the NFL in 2017 and comes away with a shocking conclusion: even without Andrew Luck, the Colts had the best long-ball quarterback in the league.

29 Sep 2015

Any Given Sunday: Bengals Over Ravens

by Andrew Healy

Sunday's game between the Cincinnati Bengals and Baltimore Ravens was the usual back-and-forth contest between the two AFC North rivals, with the lead changing hands four times in the final seven minutes. (The win expectancy chart for the game turned out to be pretty exciting.) And while the Bengals won with a pair of go-ahead touchdowns, it was actually a Baltimore touchdown that might have marked the most critical point of their season.

With seven minutes left in Sunday's game against the Ravens, the classic Andy Dalton play happened. On a five-man blitz, Dalton held the ball a beat too long, Elvis Dumervil overpowered Tyler Eifert to force a sack and a fumble, and C.J. Mosley scooped up the ball to put the Ravens in the lead, 17-14. That Eifert had to handle Dumervil alone when the Bengals also kept Giovani Bernard in to block was the first breakdown for the Bengals, but the indecisive second that preceded the fumble stood out more. It was so Daltonesque that it might as well have been wild card weekend.

Seeming to scheme towards Dalton's perceived weakness against the blitz, the Ravens continued to bring the heat on the Bengals' next two drives. They sent four more blitzes after Dalton, increasing their total to 18 on the day. On those plays, however, Dalton lit them up.

Dalton's Big Finish vs. Ravens
Time Down Distance Rushers Result
6:49 1 10 5 Pressure in his face, delivers an 80-yard touchdown to A.J. Green in stride 30 yards downfield
3:38 1 10 5 Pressure up the middle, accurate short curl to Mohamed Sanu who was left open by corner blitz, gain of 19
3:06 1 10 5 Perfect throw to Marvin Jones over his outside shoulder on a go, gain of 31
2:16 2 7 5 Perfect throw to A.J. Green into a small window for the winning 7-yard touchdown

The best of these plays was the first one. Nobody would be surprised that Dalton would execute the first part crucial to the play's success. Before the snap, he likely recognized that he had Green in the slot with free safety Kendrick Lewis lined up in coverage and a single high safety. When Green ran his route to the outside, Dalton knew he would have room to throw.

Dalton's pre-snap read, however, was nearly rendered irrelevant when Dumervil bull-rushed Gio Bernard into his lap, leaving him no room to follow through. As Dalton did most of the game, however, he left his happy feet at home and stayed calm in the face of the pressure, delivering a strike to the streaking Green.

If Able Andy has struggled against the blitz in the past, he showed little sign of any weakness on Sunday. Dalton came up biggest in the fourth quarter, but he mostly shredded Baltimore's blitzes all day.

Dalton Against the Blitz vs. the Ravens
Comp Att Yds Sk/Yds Lost Fum TD INT
10 15 249 2/9 1 3 0

Much of his success was due to the offensive line. I charted Dalton as getting no pressure on ten of those 18 blitzes. But against moderate pressure, he also did pretty well. Here's where we need to get specific as we make the Case for Andy Dalton Having More Hope Against Pressure Than You Might Think.

Point 1: Dalton's failures against pressure depend on how you define it.

According to ESPN Stats and Info (whom we thank for sharing their data), Dalton this year has continued to be terrible under pressure. Their numbers have Dalton posting a 0.5 QBR with pressure (4-for-12, 42 yards, 2 sacks, 1 scramble), compared to a 92.7 QBR without pressure. All quarterbacks do better without pressure, but since QBR only goes from 0 to 100, it is hard to have a much bigger difference than that. And if the numbers are based on just three games, it fits Dalton's pattern from previous seasons, too. That sounds pretty bad, bad enough that where playoff football is involved, you might ding the bell and dismiss the Red Rifle with "Thank you for playing, Mr. Dalton." But let's look a little deeper.

The ESPN charting information is using a strict definition of pressure. On those plays where Dalton is getting pressured heavily, he has mostly been even more atrocious than most quarterbacks. But many times on Sunday, Dalton made great throws when he was not knocked down, but he was under some smaller amount of duress that did not make the cut as being under pressure according to the charting stats. The 80-yard touchdown was just one example. In the third quarter, he threw Green open on a 47-yard play where Daryl Smith made contact with his arm on the follow-through. Green got credit for making a great adjustment, but Dalton's decision and throw there deserve just as much credit.

Point 2: Dalton changed how he played under pressure last year.

FO alumni Doug Farrar and Bill Barnwell both wrote articles exploring Dalton's struggles under pressure before last season. In short, Dalton performed very poorly under pressure in his first three seasons. As Farrar described, Dalton had a -114.7% DVOA under pressure in 2013 and a 27.4% DVOA without pressure. The Bengals averaged 7.9 yards per play when Dalton was pressured and just 1.2 when he was not, per our charting data.

At first glance, it looks like Dalton was just bad again under pressure last season. He posted a -69.0% DVOA when pressured and a 20.3% DVOA without pressure. But most of Dalton's poor DVOA under pressure last year was driven by interceptions. The Bengals actually averaged more yards per play last year when he was pressured than when he was not.

Andy Dalton, With and Without Pressure, 2014

Dropbacks DVOA Yards Per Play INT%
Pressure 75 -69.0% 7.4 8.0%
No Pressure 432 20.7% 6.5 2.5%

What this means going forward is not entirely clear. The optimist would say that Dalton could have gotten unlucky on a couple of those interceptions and the bigger sample with more yards per play is more informative. The pessimist would say "same old Dalton under pressure" in terms of overall efficiency.

For a Bengals fan, maybe it's reassuring just to know that not everything has stayed the same from year to year in the Dalton era.

Point 3: Dalton actually performed better last season against the blitz than when opponents sent four or fewer pass rushers.

If it's true that the blitz destroys Andy Dalton, then the biggest problem is that the Bengals should be very easy to defend, particularly for the better defenses that keep popping up in the playoffs. But, just as he often did on Sunday, Dalton has been succeeding against the blitz in a larger sample for some time.

Andy Dalton vs. Blitzes, 2014

Dropbacks DVOA Yards Per Play INT%
Blitz 154 17.0% 8.3 5.2%
No Blitz 353 2.4% 5.8 2.5%

None of this is meant to give Dalton all the credit for the Bengals' recent success against the blitz, success that continued against the Ravens. But whatever the reason, teams cannot simply blitz the Bengals into submission. And so there is reason to think that, with a healthy A.J. Green and with a potentially different Dalton, this time really could be different for the Bengals.

It is almost impossible not to be skeptical about the Bengals. That well-earned doubt can cause slow-witted prognosticators to pick against the Bengals even when the past may be limited in its usefulness. We all know what happened the last time the Bengals started 3-0 and looked dominant. And things did not go well after Andy Dalton's last stretch of three games averaging more than 10 adjusted yards per attempt. But with a new Dalton, we should be skeptical of assuming we know how this story ends.

By the VOA

Though the Bengals needing a late drive to win, this game featured an unusually dominant performance by Cincinnati. On Dalton's biggest day by passing yardage, the Bengals came out strongly ahead on VOA (just one more week until the opponent adjustments add that crucial D part of DVOA). Cincinnati might have blown it open early if not for a Tyler Eifert drop on fourth-and-1 from the Ravens' 2-yard line late in the first half. (Or "not drop," depending on how you feel about the NFL's rules on what is or is not a catch.)

BAL -1.7% 21.2% -4.5% -27.3%
CIN 20.5% -8.2% 0.9% 29.6%

The Ravens' remaining deficiencies in the secondary played a big role in the defense's struggles. The whipping boy from the AFC Divisional game against the Patriots, Rashaan Melvin, committed two penalties and gave up two receptions in two targets for 48 yards. After Kyle Arrington replaced him, things stabilized. Jimmy Smith is generally not a deficiency, although A.J. Green certainly stole his lunch money on Sunday. He covered Green almost the whole game, although not on the long fourth-quarter touchdown.

Quick Hits

  • The close margin also was not helped by some questionable strategic decisions from Marvin Lewis and Hue Jackson. Most notably, with two timeouts late in the first half, they let the clock run after Gio Bernard was tackled at midfield with 0:28 seconds left, potentially leaving three points on the field.
  • Kelechi Osemele had three penalties, including a crucial hands-to-the-face that wiped out a fourth-down conversion on the Ravens' last drive. Geno Atkins also looked explosive at times. On a third-down stop of Lorenzo Taliaferro in the second quarter, Atkins looked particularly like his old self.
  • John Harbaugh made a silly challenge in the fourth quarter on a spot, but he also had two great moments. Most importantly, he went for it on fourth-and-5 with 6:37 to go in the third quarter and the Ravens down 14-0. He was rewarded with a Steve Smith 50-yard touchdown. Even more impressive, he squirrelled an extra play in a game where his team was trailing. The third quarter appeared to end on a play where Flacco was roughed. Upon returning from break, referee Walt Anderson announced that the Ravens had exercised their right to have an untimed down tacked on to the third quarter after the defensive penalty. It's obviously a small difference, but the kind of thing that some coaches find and some do not.

The Keep Looking at Wins Stat of the Week

After three weeks, Steve Smith ranks third in the NFL with 349 receiving yards. In recent years, Smith has been unable to keep up his production later in the season. There have been only ten games since 1998 when a receiver 36 or older got 15 or more targets.

36-Year-Old Receivers, 15-Plus Targets in a Game (1998-2015)

Age Date Week Tgt Rec Yds Y/R TD
Jerry Rice 36-033 11/15/98 11 19 10 169 16.9 1
Jerry Rice 37-053 12/5/99 13 15 9 157 17.4 2
Jerry Rice 39-337 9/15/02 2 19 11 94 8.6 0
Tim Brown 36-097 10/27/02 8 17 13 144 11.1 0
Jerry Rice 40-042 11/24/02 12 15 7 110 15.7 1
Jerry Rice 41-070 12/22/03 16 16 10 159 15.9 0
Terrell Owens 36-300 10/3/10 4 15 10 222 22.2 1
Tony Gonzalez 36-258 11/11/12 10 15 11 122 11.1 2
Steve Smith 36-131 9/20/15 2 16 10 150 15.0 0
Steve Smith 36-138 9/27/15 3 17 13 186 14.3 2

Smith has pulled that feat off twice in the last two weeks. While stats guys everywhere love them some Steve Smith, the Ravens' offense needs another target to emerge and quickly. First-round draft pick Breshad Perriman cannot get healthy fast enough.

Posted by: Andrew Healy on 29 Sep 2015

31 comments, Last at 30 Sep 2015, 8:22pm by TimK


by saladin65 :: Tue, 09/29/2015 - 1:27pm

Anyone checking Daltons similarity scores could have seen this coming of course.

by Eddo :: Tue, 09/29/2015 - 1:41pm

Can you elaborate, please?

by saladin65 :: Tue, 09/29/2015 - 3:28pm
by Blotzphoto :: Tue, 09/29/2015 - 7:04pm

I'm not surprised by that Joe Montana comp at all. Troy Aikman is another similar player.

by Anon Ymous :: Tue, 09/29/2015 - 1:41pm

I know it isn't a focus of this article, but Cincy was victimized by ridiculous TD-stealing call. That would have been a much bigger deal had they not pulled off the comeback.

by jonnyblazin :: Tue, 09/29/2015 - 2:17pm

The pass to Eifert? It was an obvious call. The TE was contacted by the DB with only one foot down, and didn't maintain possession upon falling to the ground.

by Anon Ymous :: Tue, 09/29/2015 - 2:25pm

He got three feet down and only went horizontal because he was reaching for the end zone. The defender's contact didn't cause a bobble, nor did it knock Eifert off balance. As such, there was no reason to even use the "going to the ground" rule.

Don't take my word for it, check it out for yourself: https://youtu.be/wo3hCCJRD0M?t=1m16s

by TimK :: Tue, 09/29/2015 - 2:47pm

Thanks for the link, not seen it myself until now. Definitely got both feet down (one of them twice) and had enough control of the ball to reach it well past the goal-line. But under current rules it was probably the right call.

by deus01 :: Tue, 09/29/2015 - 3:00pm

I don't understand why turning and reaching for the end zone doesn't make it a catch. If a receiver changes the direction of their body and the ball after making a catch that should indicate that they've secured the pass and should be treated the same way as any runner.

If that play was consistent with the rules then the rules are stupid.

by tuluse :: Tue, 09/29/2015 - 3:03pm

Eifert never had balance, before his second foot lands the defender makes contact, thus he has to control the ball all the way through the catch.

It's an annoying rule, but that's how it's written.

by Anon Ymous :: Tue, 09/29/2015 - 4:04pm

He did. When Eifert gets hit, it bends his upper body forward, but his legs remain completely balanced. If he had been 20 yards from the end zone, he would have attempted to shrug off the tackler and start running. The only reason he ends up vertical is because he made the conscious decision to dive into the end zone; it was never part of the catch process.

Frankly, I'm not a huge fan of the rule as is, but in this case it wasn't even accurately applied.

by jonnyblazin :: Tue, 09/29/2015 - 5:17pm

"The defender's contact didn't cause a bobble, nor did it knock Eifert off balance."

It doesn't matter. It's not the refs job to judge whether a defender knocks a receiver down, or contacts a receiver who then immediately goes down on his own volition. Contact + down receiver + no possession = no catch. It's not that complicated a rule.

by Anon Ymous :: Tue, 09/29/2015 - 5:30pm

It does matter when the receiver didn't start going down until after making the catch.

by tuluse :: Tue, 09/29/2015 - 5:32pm

The defender makes contact before it's a catch, only one foot is down.

As soon as that happens, he has to control the ball for the "entire process".

by Pat :: Wed, 09/30/2015 - 9:26am

His legs aren't balanced. They show no evidence of balance. They move continuously in the opposite direction of the top of his body. That's not 'balance', that's 'falling down'.

You might be saying "he's balanced" because his knees bend and show that he's pushing off the ground (see 1:19 in that video). But that's just him *trying* to gain his balance. He never does - because the defender crashes immediately into his knee and kicks it off the ground.

The back foot then slips, he ends up horizontal, with both feet off the ground, and loses the ball. He's falling down. I don't get it. He catches the ball while falling down, and loses it in the process. No way that's a catch.

by Anon Ymous :: Wed, 09/30/2015 - 10:04am

You are incorrect. Eifert's upper body bends, but his legs are perfectly balanced. The only reason he ends up horizontal is because he *chose* to end up horizontal; it was an intentional dive. Had he been anywhere else on the field, Eifert would have just shrugged off the tackle attempt and run forward.

"His body starts is rotating counterclockwise the entire time from when his feet touch the ground. He's falling. The entire time. And then as he's falling, he drops the ball. It's incomplete. Easy."

This isn't an accurate representation of what happened. If that were the case, I wouldn't have said anything about it.

by Noah Arkadia :: Wed, 09/30/2015 - 12:03pm

In the case of Anon Ymous vs the Internet, this court rules in favor of the Internet. Seriously, to your eyes he might be throwing himself to the ground looking for the endzone, but that makes no difference. Since it happened after contact and contact happened before he got two feet down, he needs to maintain possession all the way. Like jonny said, it's not that complicated.

Who, me?

by ChrisS :: Wed, 09/30/2015 - 12:57pm

He was hit prior to getting both feet down. At 1:17 in your linked video he is being contacted by the defender and his left foot has not touched down. If you are hit by a defender prior to getting both feet down, by rule you must control the ball all the way through contact with the ground. No catch. Now if you want to argue whether the rule is stupid I might agree with you.

by Pat :: Wed, 09/30/2015 - 9:18am

nor did it knock Eifert off balance.

What are you talking about? Eifert never has his balance. At all. Look at his legs - they go in one direction, continuously : away from the end zone, opposite the top of his body. He's falling the entire time. If you look at his forward leg you can see that he tries to start keeping himself upright, but the defender knocks out his legs and he slides down to the ground.

His body starts is rotating counterclockwise the entire time from when his feet touch the ground. He's falling. The entire time. And then as he's falling, he drops the ball. It's incomplete. Easy.

Moreover: It was 4th and 1. 1:54 to play.

Eifert had the first down as soon as he caught the ball.

What the hell was Eifert thinking reaching for the end zone?!

by Dan :: Tue, 09/29/2015 - 3:20pm

It's an incomplete pass.

If it was at midfield, would you really rule that a catch & fumble rather than an incomplete pass? He's hit as he's making the catch, and the ball comes out before he hits the ground.

(The fact that it happened at the goal-line rather than at midfield doesn't have any impact on whether or not it's a complete pass - the rules are the same in both cases.)

by deus01 :: Tue, 09/29/2015 - 3:30pm

If it was at midfield I would say it was a catch and fumble. It might take a few fumbles but receivers would be more careful reaching for a first down if it started happening. It would then be the same as if a running back was reaching and the ball was knocked out by the tackler's foot as he fell.

by RickD :: Tue, 09/29/2015 - 6:07pm

I would say he was hit after he'd made the catch. And it would be a fumble elsewhere on the field.

by TimK :: Tue, 09/29/2015 - 2:43pm

It does seem ridiculous to me that someone who has enough control to twist round and stretch out with the ball doesn't get the same 'touch the plane of the goal' rules as a running back twisting and stretching - thinking of the Detroit TD on Sunday night as an example.

I can understand how someone making a diving catch in the end zone (or even across the goal-line) but losing the ball as they land can be said to have not completed a catch, but if you have good enough grasp to twist and stretch out then you've got the ball and made the catch - what more "football move" is there than straining for the extra inches to score? I know they've removed football move and replaced it with "demonstrating he is clearly a runner", but taken literally the new rule could be used to deny any diving catch in which the catcher is tackled down before being able to get up and make a step...

Similar things happened with a sideline catch on Sunday night - Sanders caught the ball, pulled it in, had both feet down and then when he hit the ground OOB the ball moved enough (yes, it moved a fair amount, but he'd already brought it in to his body) without ever touching the ground, for it to be called not a catch. Personally I'd like to see 'pulling the ball in' in some way to be considered one sign of completing a pass, then once you've done that you don't get penalised for an slight bobbles etc after establishing some control of the ball.

I guess a catch is one of those things that is almost better defined by listing what is isn't rather than what is required. Don't let the ball bounce off the ground for a start ;) - though I'm glad a ball can graze the ground on a diving catch now as long is it is held firmly before the ground comes into play.

Some of this reminds me of cricket where players have a habit of celebrating a catch in the field by hurling the ball straight back up in the air. There have been some players who have lost control of the ball whilst doing this and have had the catch adjudicated as a drop because of this. It all comes down to how long and how do you hold something for it be considered controlled as a catch. If someone is going OOB and the ground cannot cause a fumble then I feel that pulling it into the body is a good sign of control even if it might move a bit hitting the ground as the player might be swapping hands to brace their landing. Sadly I can't think of any definition that won't have contentious edge conditions, but the current case seems to have some glaring issues.

by RickD :: Tue, 09/29/2015 - 6:05pm

It's ridiculous that the NFL rule about whether a catch is legal can baffle a couple dozen football fans. It really needs to be simplified.

Suggestion: it's a catch when the receiver has control of the ball.

Establishing position should be a different topic. But if the receiver has control of the ball, and he's hit "before both feet hit the ground" - hey, fumble!

If "going to the ground" is a process that isn't complete until the player is out of bounds or already in the end zone, why should we care about it? This notion that a guy has already scored a touchdown, unless he drops the ball after he scored, simply makes no sense.

The rule book feels like it has too many authors.

And yes, Sanders also made what looked like a catch to me on Sunday night.

by tuluse :: Tue, 09/29/2015 - 6:21pm

I approve of how you want the rules to be, but that's not how they are now.

by TimK :: Tue, 09/29/2015 - 6:33pm

Sadly for our enjoyment, but possibly happily for the general quality of the officials (and conversation around here ;) ), most people aren't arguing the calls, so much as decrying the rules that make the calls seem so hard to understand.

Given the amount of television footage these days maybe someone can simplify the catch rules by making it 'you must have control of the ball' and then using specific examples to show what is and isn't considered control that can help fairly consistent application in edge cases.

by Blotzphoto :: Tue, 09/29/2015 - 7:01pm

It's also not legal to kick the ball out of the receivers hands, which is how Eifert lost control of the ball ;)

by jonnyblazin :: Tue, 09/29/2015 - 7:11pm

"Suggestion: it's a catch when the receiver has control of the ball."

So if a receiver leaps in the air, catches the ball for a split second, and then has the ball stripped out by a defender, that's a fumble?

Then there'd be a lot more fumbles among receivers, especially in the middle of the field. That might seem like it would be more entertaining, but then offenses might become more risk averse and avoid those kind of throws.

As it is now, QB's are feasting upon LB's and safeties in coverage because they can't knock out receivers anymore. My guess would be this rule change would bring back some balance by giving an advantage to the defense.

by TimK :: Wed, 09/30/2015 - 10:26am

It all depends what 'control' is considered to be.

I'd say that getting the ball grasped in a hand or between hand and body or two hands and being able to further adjust how you hold it counts as control. So, yes, if someone leapt up, brought the ball down, landed and then had it knocked free I'd be happy to call that a catch and fumble. If the receiver never lands, before being hit and it coming free, then they've not established possession on the field of play, but once someone has taken control over the travel of the ball, and touched down the requisite contact point(s) then it would be theirs to lose. I don't think a 'split second' is enough to show control. But a second certainly can be if the ball has a different trajectory controlled by the catcher.

If that makes for a few more fumbles and some more catches too, then it is a risk/reward balance that seems to add excitement and reward athletic attempts. For example the one handed catch brought into towards the torso and then stretched out for the goal line with both hands would be control enough for me. There can always be arguments about a purely one-handed catch (most people would have trouble gripping a fast thrown ball well enough in one hand to convince they had control without pulling it in in some way I think), but in the case of the 4th down pass in this game, I don't think balance should matter, he caught the ball well enough to hold it try to stretch it out over the goal line, whilst doing this his feet touched the ground enough times, so I would be happy to see the rules reward that s a catch and a touchdown. Had he done it to stretch the ball past the first down marker then it would have been a fumble (a la D Thomas for Broncos vs Lions). Had he been diving and caught the ball in-mid air and crossed the goal line without touching the ground then lost the ball into the out door on landing it would be an incompletion.

Of course none of us is likely to be on the competition committee anytime soon, so we will probably have to put up with yet more refined hair splitting in a steadily more incomprehensible rule book.

by Noah Arkadia :: Wed, 09/30/2015 - 12:15pm

And therein lies the question, what is considered control. Mr. Blazin is correct in that it's not as simple as that. If you say "adjust how you hold it" then you have plays like the Eiffert non-catch where you can't tell if he adjusts how he holds it or not. Does reaching forward (or towards the body) count as adjusting if you lose the ball in the process?

And what about diving catches where you bring in the ball towards the body but it's coughed up when hitting the ground? That shouldn't be a catch, in my book, regardless of whether the receiver has been touched or not.

Let's not lose sight of the fact that due to the fact that rules are by nature completely arbitrary, there's always going to be gray areas. I actually like the way the rule is defined now, it's an overkill method but it does make things very clear (once you get it).

Who, me?

by TimK :: Wed, 09/30/2015 - 8:22pm

I agree with 'coughing it up' invalidating the catch - to me that means losing control and the ball hitting the ground. But I feel the ball moves somewhat when the player hits the ground without the ball itself hitting the ground then that should be OK.

To be honest, perhaps having not seen it live, the Eifert non-catch was less irritating to me than the Sanders non-catch in the Sunday night game. That was a case where everything looked to me like a great catch but I knew that under the rules as written it was unlikely to stand, and I couldn't see anything really the player could have done. Fortunately neither of these plays were crucial in the end - at least we aren't shaking our heads at another 'Fail Mary' type play. Also some things have improved in the last few years, I was so glad to see the back of the 'force out' rule and the allowing some part of the ball to graze the ground provided control had been established and was maintained makes sense when the ball is the size and shape it is too.

I think it the case of the edge conditions where things seem to be different between runners and recievers, or between different lines on the field of play that causes problems. Perhaps that would be the best simplification to make. I've been involved in trying to draft codes of conduct and technical documentation in my time, so I know how impossible it is to cover everything in a way that doesn't have edge conditions. But there is a point when the documentation gets too complicated and is probably better paring down to basics and starting again, and several NFL rules seem to be getting to that level.