Writers of Pro Football Prospectus 2008

Most Recent FO Features

WentzCar16-1.jpg

» Week 11 DVOA Ratings

DVOA has finally climbed on board the Wentz Wagon! The Eagles move into the No. 1 spot, but they aren't the only strong, well-balanced team in the NFL this year. New Orleans, Pittsburgh, and the Los Angeles Rams make this one of the best seasons ever for multiple teams over 30% in DVOA, and Minnesota isn't far behind.

25 Jan 2017

Film Room: Belichick's Defense

by Cian Fahey

It was always unlikely that the Pittsburgh Steelers would upset the New England Patriots in the AFC Championship Game. Entering the game, Mike Tomlin's Steelers had faced Bill Belichick and Tom Brady on six separate occasions. In those six games, the Steelers gave up an average of 33.3 points per game. Tomlin's Steelers have largely used the same game plan against Brady whenever they have faced off: a passive pass rush with zone coverage in behind. That was the same strategy they employed against the Patriots on Sunday. The Patriots, with their second-ranked offense by DVOA, did exactly what they have always done.

With a deficiency of talent on the defensive side of the ball, the Steelers were never going to win a low-scoring game. Their only chance of upsetting the Patriots was to outscore them in a shootout. With Le'Veon Bell, Ben Roethlisberger, and Antonio Brown in the lineup, the Steelers might have had a chance of scoring 30-plus points. Once Bell was injured early in the first quarter, though, the game ended as a contest.

As recently as 12 months ago, the Steelers had the talent to compete without Bell on the field. This year, though, Heath Miller's retirement and Martavis Bryant's suspension have elevated the importance of Bell, Brown, and Roethlisberger to a point where they had to combine to carry the majority of the offensive responsibility. When Miller and particularly Bryant were in the game, teams couldn't tip their defense in the direction of anyone on the field. When those players were taken out of the offense and replaced with inadequate successors, defenses only had to fear Bell or Brown. They could choose who to focus on, or they could hedge against both, forcing Sammie Coates, Jesse James, Eli Rogers, or Cobi Hamilton to win their one-on-one matchups.

On the Steelers' very first third down of the game, a third-and-1, Belichick laid out his philosophy for the day.

The first thing to note on this play is the positioning of Brown. The Steelers put their star receiver on the narrow side of the field, alone. They are attempting to isolate him against a defensive back or create single coverage on the other side of the field. When Rogers goes in motion before the snap, his cornerback follows him across the field while the deep safety doesn't move. Roethlisberger now knows that he has single coverage to his left and that Brown is going to be double-teamed at the snap.

The next thing to notice is the Patriots' front. They are in a conventional alignment, and they aren't moving defenders around to threaten a blitz. Six defenders are tight to the formation, with a nose tackle covering the center and a defensive end in each gap between the guard and tackle. The Patriots are in position to fill every gap, creating a wall if the ball is handed off to Bell.

Belichick has told Roethlisberger to beat him with someone other than Bell or Brown before the ball has even been snapped. After the snap, he becomes even more aggressive.

(Click here if you are having trouble loading the image.)

Before the snap, the left outside linebacker was in position to play the run or rush the passer. He initially moves forward at the snap, but then drops out to cover Bell. The inside linebacker also moves towards Bell at the snap. This allows the outside linebacker to aggressively get to Bell's body without worrying about being quickly beaten.

Meanwhile, Brown is facing a similar situation, as the Patriots use their No. 1 cornerback, Malcolm Butler, to press him at the line while the deep safety waits in position in behind. The wide-open space on the other side of the field is there for Roethlisberger to lay the ball into. The quarterback throws a perfect pass, but Coates can't create separation against the defensive back through his route. He still has a chance to catch the ball, but short-arms it.

Teams don't normally take shots downfield on third-and-short. This play was dictated by the defense though. Belichick bet that the Steelers' lesser weapons couldn't beat his defense's lesser cornerbacks. He was right.

It's easy to imagine what Martavis Bryant would have done in that situation. Coates is only similar to Bryant in stature. Bryant is a much better route-runner, a more fluid receiver, and someone who naturally adjusts to the ball in the air. Coates doesn't beat press coverage the way Bryant does, and has a more mechanical approach to catching the ball. That's largely why he drops so many passes.

This situation wouldn't have presented itself to Bryant though. Unlike most coaches, Belichick isn't stubborn. His play calling adapts to its situation like a chameleon. If Bryant had been on the field, Belichick would have been forced to keep his safety on the wider side of the field, isolating Brown, or forced to play a defense that looked closer to a Cover-2, creating either a running opportunity or single coverage for Bell.

Bell was injured on the second play of the third drive. Up to that time Butler had repeatedly been given safety help against Brown. Bell benefited once or twice before Roethlisberger forced a pass into double-coverage that should have been intercepted.

Roethlisberger had a bad game. In the first quarter alone he missed Brown on a quick throw to the flat, forced that ball into double-coverage with Bell open underneath, and then missed Bell open deeper to check down to a covered receiver. When you are forced to throw to inadequate receivers, your inconsistency has a greater impact because your bad plays remain bad, but more of your good plays are turned bad by receiver error.

When Bell went out, Belichick pushed harder onto the wound by erasing Brown. Without Bell on the field, the Patriots primarily played Cover-2 while giving extra attention to Brown. The closer the Steelers got to the Patriots' end zone, the more aggressive the Patriots became with Brown.

Cobi Hamilton failed to take advantage of two touchdown throws in these situations. Jesse James allowed himself to be stopped just short of the end zone on a play where he should have scored. Sammie Coates ran himself out of bounds before coming back in to catch a touchdown, then failed to catch up to a well-thrown ball because of another poor route in the end zone. Roethlisberger may not have played his best game, but he gave his teammates plenty of opportunities to punish Belichick for disrespecting them. As it turned out, that disrespect was deserved.

The conundrum for the Patriots now is what to do with Julio Jones. Belichick famously undertook a similar strategy against the Atlanta Falcons years ago when they had Jones, Roddy White, and Tony Gonzalez. Gonzalez was made to look like he was fighting to get into a club between two bouncers coming off the line of scrimmage. It was a lot easier to execute that strategy against that offense because of the respective talent and coaching situations. That offense had three stars in White, Gonzalez, and Jones, but it didn't have as good of an offensive line, running game or playcaller.

If Belichick attempts to double-team Jones the way he did Brown, it's unlikely to work. Devonta Freeman and Tevin Coleman are too dangerous, both as receivers and in the running game, for that to work. Furthermore, with how often Matt Ryan throws off of play-action from under center, it will be very difficult for defensive backs to comfortably bracket Jones as he advances downfield.

Shanahan uses hard play fakes out of passing sets that look exactly like the offense's run designs. It gets his quarterback more time in the pocket, and it also distorts coverages by drawing linebackers forward and creating hesitation in defensive backs. Shanahan has used this scheme to great effect with lesser quarterbacks than Matt Ryan. Ryan's numbers have flattered him this year because he has repeatedly been put in situations where his receivers have been wide-open after play fakes. Shanahan repeatedly attacks the space between the safeties and linebackers that is created by play-action.

This play against the Carolina Panthers from early in the year highlights how open Jones gets when the defense aggressively attempts to stop the run.

(Click here if you are having trouble loading the image.)

This play came late in the game when the Falcons were looking to run out the clock, but the Falcons have a top-10 rushing unit so it's not exclusively a situational reaction.

Last week, Jones scored a long touchdown after a play fake on another in-breaking route.

(Click here if you are having trouble loading the image.)

Though Ryan didn't line up under center, he still used an aggressive play fake and moved with his offensive line through the handoff action. Meanwhile, Jones was beating the aggressive coverage from the slot before he broke into the space in front of the deep safety. Ryan had a simple completion before Jones took the ball the rest of the way. The Patriots will expect better play from their defensive backs than the Packers got from theirs. Even if the Patriots don't have stars in their secondary, the difference in individual quality is still vast.

Kyle Shanahan has had a great season. He has repeatedly set his quarterback up for success by diagnosing the defensive coordinator's tendencies and attacking them. If Belichick sits in one specific coverage, Shanahan will adjust and find a way to attack it.

Of course, that assumes his original game plan won't be effective. Even though the Steelers offense boasts a good quarterback and a lot of talent, that unit was built on beating you in specific ways. The Falcons offense is better built to adapt to specific looks because they incorporate packaged plays (plays where the quarterback has an option to throw the ball even if it is a called run) to quickly expose alignments that favor the offense.

The coaching matchup when Atlanta has the ball will be the most fascinating aspect of the Super Bowl. Kyle Shanahan and Bill Belichick share few similar traits. Shanahan is the young offensive mind who is about to get his first opportunity to prove himself as a head coach. Belichick has long since answered any doubts about his genius. Furthermore, Shanahan's offenses have consistently been built on the same principles, whereas Belichick has repeatedly redesigned his defensive unit over the length of his career.

Since the Falcons are in a similar situation to the Steelers where they are unlikely to win a low-scoring game, this matchup is not only fascinating but of the utmost importance.

Posted by: Cian Fahey on 25 Jan 2017

47 comments, Last at 27 Jan 2017, 6:05am by Bright Blue Shorts

Comments

1
by Will Allen :: Wed, 01/25/2017 - 3:34pm

I suspect that Belichik will invite Shanahan to run the ball 35 times. The response to that invitation will determine the game, and that response will, in turn, will depend on how may points per drive the Falcons' defense gives up.

2
by Bright Blue Shorts :: Wed, 01/25/2017 - 3:41pm

Having talked elsewhere about BB's XXV defensive gameplan against Buffalo, I was just wondering if he'll break out some version of the 2-5-4 / lots of DBs.

3
by Will Allen :: Wed, 01/25/2017 - 3:53pm

He made the same invitation to Mike Martz, and Martz, despite facing an early Brady offense which was not nearly as efficient as a late Brady offense, was too pigheaded to happily hand the ball off to Marshall Faulk, behind Orlando Pace & Co., for a few hours.

Shanahan won't be Martz, but the Falcons will have to run block, and just as importantly, the Falcons defense has to make the Falcons running game a viable option.

4
by Bright Blue Shorts :: Wed, 01/25/2017 - 4:47pm

Good points Will, I missed the XXXVI SB so never saw how it actually played out.

I do remember a few years ago after failing to win a SB with passing offenses, someone asked him if he'd go back to a running game/hard defense that he'd used in the early years of success. BB essentially said when you've got a talent like Tom Brady at QB, you don't employ him just to hand the ball off.

So again, if you invite the Falcons to run it forces Matt Ryan just to be a hand off machine and automatically takes out Julio Jones and the other receivers.

5
by sbond101 :: Wed, 01/25/2017 - 4:49pm

I think the other interesting dimension to this is that the Pats have had a lot of success stopping the run in light sets/playing the pass because of great gap integrity with the DL and DB's that tackle really well as a group. Bottom line is that even if the Falcons are smart enough to accept the invitation to run, they have to either consistently move Alan Branch with a double team (not easy), or have their running backs break tackles as the Pats DBs try to turn things back to the inside. I don't think either is likely to occur enough to make it a good strategy for the Pats to stop inviting the run.

Contrary to popular opinion I really don't see the Falcons offense out-executing the Pats offense in a shootout as more than a longshot path to victory. If the Falcons want to win this game I really think the best chance is to play hyper-aggressive defense against the short pass (both press coverage, and every marginally-illegal hit on a defenseless receiver you can get) and hope you can get some take-aways out of your safeties on the inevitable deep throws. I really find it hard to believe the Falcons can win in a "clean" game, the Pats simply don't make very many unforced errors.

6
by Will Allen :: Wed, 01/25/2017 - 4:57pm

Yeah, extending the invitation to run does not, to any extent, mean that executing the blocking will be easy. This is the game, assuming the Falcons defense is reasonable about how many points per drive it gives up.

9
by contrarycomet :: Wed, 01/25/2017 - 6:01pm

A cynical observer might call this strategy "Make The Zebras Beat Us" (no criticism of the analysis intended).

Also, once the Pats noticed that the refs were calling a loose game, you'd think they would take the same aggressive approach. It's interesting to try and think about which team would benefit more from loose/tight officiating.

21
by MC2 :: Thu, 01/26/2017 - 6:39am

If the refs allow a lot of contact by the DBs, I think it favors Atlanta, especially with no Gronk. Jones and Sanu are bigger and stronger than any of New England's WRs are.

31
by RBroPF :: Thu, 01/26/2017 - 12:08pm

Unless of course, there's a Michael Floyd sighting.

Though your point is still true.

40
by commissionerleaf :: Thu, 01/26/2017 - 4:31pm

I concur with the Atlanta strategy of playing against the Zebras. New England's offense depends a lot on quick passes under ten yards and fairly quick seam routes - in part because Brady is near-perfect on those throws and near-league-average throwing anywhere else - I would just man up on the outside and play single-high with a safety trolling ten yards to twelve yards deep between the hashes and make Brady throw outside.

I feel better about letting Dion Lewis wheel routes be the Patriots potential route to victory than 15 Edelman slants.

41
by Bright Blue Shorts :: Thu, 01/26/2017 - 4:38pm

New England have been very flexible this year.

You likely saw what happened against the Texans where they just threw deep for big plays. They were doing that early in the Brady-season with long plays to Hogan and Bennett.

42
by sbond101 :: Thu, 01/26/2017 - 8:03pm

That's true, but I think it misses a key point. If you play extremely physical short/intermediate, and more importantly punish receivers (illegally-ish) whenever possible in the short passing game, the Pat's might keep passing short and produce a turnover. Atlanta isn't going to stop the Pats consistently, but If they can frustraight the Brady into a pick or make the receivers drop it a bit, it might be enough in a high variance/low possession game plan. Screw the 15 yard penalties, without an aggressive game plan their probably giving up 350 yards anyways, what's 60 more on a few personal fouls if it gets a possession and a couple drops.

43
by Anon Ymous :: Thu, 01/26/2017 - 8:22pm

I don't think BBS misses the point, he is saying that that strategy would be more effective if NE weren't so capable of exploiting the weaknesses of any defense.

Yes, you could point to the Houston game as an example of NE still struggling after making the adjustment, but Houston has a much better defense than Atlanta, particularly in the secondary and interior pass rush... both of which are necessary for that strategy to be successful.

The only mistake BBS seems to make is crediting Bennett for Gronk's success. Martellus hasn't been a big part of the passing game (as a receiver), it's mostly been quick PAPs. In fact, the deep seam hasn't been utilized much at all since Gronk went down.

47
by Bright Blue Shorts :: Fri, 01/27/2017 - 6:05am

Thanks Anon Ymous ... my memory conflated two games where Hogan was catching deep balls in both ...

Against the Bills both Hogan and Gronk caught 53-yd TDs.

Earlier in the season against Cleveland, Hogan caught a couple of deep balls and Bennett had 3TDs, two of them where short walk-ins but the third was a 37-yarder down the sideline.

29
by Noah Arkadia :: Thu, 01/26/2017 - 11:51am

While success running the ball might not be a given in that scenario, I doubt the Patriots can take the pass away completely, either, even if they overplay it.

32
by Anon Ymous :: Thu, 01/26/2017 - 12:31pm

"I suspect that Belichik will invite Shanahan to run the ball 35 times."

I'm of the same mind. The examples of this are numerous, but a relevant one that gets overlooked is SB46. NE's defense was objectively terrible and they held a good Giant offense to 13 points deep into the fourth quarter. There was a serious field position cost, which had a large impact on the game, but that likely wouldn't be as much of a problem this year. And the plan might have worked even better had it not been for a surprisingly mediocre Wilfork performance and what was arguably Mayo's worst game as a pro.

Atlanta's offense is much better than NY's, of course, but NE's 2016 defense is correspondingly superior to their 2011 counterparts, and much more capable of slowing down a running game with fewer bodies.

EDIT to add some DVOA context, all listings formatted as (total, pass, run):

2011 Giants offense: (10.5, 30.2, -4.9)
2011 Patriots defense: (13.2, 19.7, 3.6)

Gap: (23.7, 49.9, -1.3)

2016 Falcons offense: (25.3, 53.0, 1.7)
2016 Patriots defense: (-1.5, 13.9, -23.7)

Gap: (23.8, 66.9, -22)

Of the four, only NE's 2016 defense saw a significant improvement in their weighted DVOA. As far as I know, that isn't any more predictive of playoff success than season long numbers, but it is something that I find interesting.

I must confess that this makes it easier for me to imagine a low 20s scenario than I could earlier this week.

7
by galactic_dev :: Wed, 01/25/2017 - 5:12pm

I remember Payton Manning's last playoff as a Colt, the Jets defense skewed so heavily to defending the pass that it forced Manning to keep handing the ball off. Speedily pursuing DBs kept the run from gashing them too badly and they won.

8
by Theo :: Wed, 01/25/2017 - 5:44pm

It's unbelievable, like I said in the game comments, 4 deep and no pass rush is just saying "oh hi, just pick us apart and be comfortable all day, oh and please don't use lube when we mess up our coverage on some plays".

The one and a half play they did a zone blitz, Brady ended up with his ass in the grass...
and next time the Steelers do exactly the same thing, because it's clear they (Tomlin) just don't learn.

11
by RobotBoy :: Wed, 01/25/2017 - 7:14pm

Blitzing Brady as a regular thing is a pretty bad idea as he makes quick decisions and his hot read is somebody like Edelman or James White. Brady's numbers against blitzes are ridiculous. I do agree that mixing coverages would certainly have helped. Yet since Pittsburgh DB's couldn't even maintain their assignments in the one defense they did play, it's understandable why Tomlin didn't try different looks.

22
by Hoodie_Sleeves :: Thu, 01/26/2017 - 10:09am

"Brady's numbers against blitzes are ridiculous. "

Brady's numbers against soft zones with no pressure are pretty ridiculous too.

39
by Theo :: Thu, 01/26/2017 - 4:05pm

I'm not saying to go cover 0 and all out blitzes.
But at least show a game plan.

10
by Joe Pancake :: Wed, 01/25/2017 - 7:05pm

I don't understand the comment "Since the Falcons are in a similar situation to the Steelers where they are unlikely to win a low-scoring game..."

Why is this the case? If the Falcons are able to win in a shootout it will be because their offense was better against the Patriots defense than the Patriots offense was against their defense over the course of the football game. Now suppose the game is low-scoring, if the condition in the previous sentence still holds (i.e., the Falcons can outscore the Patriots), then why do their chances of winning diminish?

Another way to look at it: It might be very unlikely that the Falcons D can stop the Patriots O. But isn't it equally unlikely the Patriots D can stop the Falcons O? If the answer is no, then the Patriots are just better, so I don't see how a shootout helps the Falcons anymore than a low-scoring game. (If anything a high score typically means more drives, which gives the better team more of an opportunity to ensure there isn't a fluky upset.)

To me the take away from this game is that neither team is likely to win without scoring a bunch of points, so both coaches should be very aggressive in trying to get a touchdown every drive. But I don't see why a low-scoring game favors the Patriots.

12
by Will Allen :: Wed, 01/25/2017 - 7:29pm

The fact that the Patriots were good on defense, and the Falcons much less so, suggests that the Patriots are not as unlikely to succeed in impeding the Falcons offense, as the Falcons are unlikely to succeed in impeding the Patriot offense.

13
by Joe Pancake :: Wed, 01/25/2017 - 8:04pm

If that's the case, then the Patriots are just better than the Falcons and are likely to win. But I still don't see why being better helps them more in a low-scoring game than a shootout (which was the claim to which I was responding). As I said above, if anything, it seems that a shootout inherently favors the better team, because a shootout typical means more drives and more drives means less chance of fluky outcome.

15
by Will Allen :: Wed, 01/25/2017 - 8:10pm

In a low scoring game, the team with the significantly better defense is favored. Defense typically gets worse in the 4th quarter, and the lesser the base from which the decline starts, the worse the 4th quarter performance is likely to be

17
by Joe Pancake :: Wed, 01/25/2017 - 9:17pm

"In a low scoring game, the team with the significantly better defense is favored."

1. What are you basing this claim on? It seems to me you always have to take offense into account.

2. Even if this is the case, is the team with the significantly better defense favored *more so* than they would be in a shootout? That's the pertinent question.

"Defense typically gets worse in the 4th quarter, and the lesser the base from which the decline starts, the worse the 4th quarter performance is likely to be."

1. Why does this matter more in a close game than a shootout? Again, if anything it seems like a declining D would be *more* damaging in a shootout because said declining D would be on the field for more drives, and thus it is less likely they will fluke their way into stops.

In general, you seem to be making the case that the Patriots would be favored in a low-scoring game, which I don't disagree with. But I have yet to see a compelling reason why they would be favored more in a low-scoring game than a shootout.

19
by Will Allen :: Wed, 01/25/2017 - 9:51pm

Because a high scoring game would indicate that the generally superior Patriots defense had been negated already.

30
by Noah Arkadia :: Thu, 01/26/2017 - 11:56am

I agree with Joe, if the game is low-scoring then by definition both defenses were successful regardless of their quality. The point here is that if either defense is successful, chances are very high they'll win the game. If both are, all bets are off.

33
by Hoodie_Sleeves :: Thu, 01/26/2017 - 1:45pm

"if the game is low-scoring then by definition both defenses were successful regardless of their quality."

This seems to be largely an argument about semantics and definitions to me (not just by you).

I'd consider 24-10 a low scoring game here (45.6 points is average), and in that case I don't think you could really argue that both defenses were successful.

And I agree with Will here - I think we're much more likely to see the Patriots win 24-10 than we are to see the Falcons, largely because they have similar offenses and the Patriots have much better defense and ST.

14
by Will Allen :: Wed, 01/25/2017 - 8:06pm

Nevermind

20
by RickD :: Thu, 01/26/2017 - 12:45am

The article says in the second paragraph "With a deficiency of talent on the defensive side of the ball, the Steelers were never going to win a low-scoring game. Their only chance of upsetting the Patriots was to outscore them in a shootout. "

The Falcons are unlikely to win a low-scoring game because they are unlikely to be in a low-scoring game. Because they are unlikely to keep the Pats' score low.

Sadly, vernacular English fails to convey the nuances of conditional probability.

"The Falcons are unlikely to win a low-scoring game" could mean either
P[Falcons win && score is low] = small
or
P[Falcons win | score is low ] = small,

where && is AND and | implies a conditional probability.

Seems like the intent of the article is the first statement, while you're interpreting it to mean the second.

At least that's my take.

23
by dmstorm22 :: Thu, 01/26/2017 - 10:56am

Spot on.

If the Falcons win the game, it is not very likely to be a low scoring game. Because the Falcons defense is not good and their offense is very good. Of course, the opposite is true too, even if the Falcons lose, it is not too likely it is a low scoring game.

If you know the score of the game is going to be low, then yes, the Falcons probably have an equal chance of winning the game.

However p(Falcons win & hold NE < 20 pts) is far less than p(Patriots win & hold ATL < 20 points).

27
by Joe Pancake :: Thu, 01/26/2017 - 11:29am

"However p(Falcons win & hold NE < 20 pts) is far less than p(Patriots win & hold ATL < 20 points)"

Yes, I agree, but this doesn't necessarily mean that the game will be low-scoring. This still allows for a reasonably high-scoring game, and it's easier to see the Patriots winning 45-17 than vice-versa.

"If the Falcons win the game, it is not very likely to be a low scoring game."

I also agree with this, but I think this is true, because it is not very likely to be a low scoring game no matter who wins.

I think I can sum up my point in this way. Instead of the author saying: "Since the Falcons are in a similar situation to the Steelers where they are unlikely to win a low-scoring game, this matchup is not only fascinating but of the utmost importance"

I think it's much more accurate to say: "Since this game is unlikely to be low-scoring, it is paramount the Falcons are very aggressive in trying to score touchdowns every time they have the ball, and so this matchup is not only fascinating but of the utmost importance."

24
by Joe Pancake :: Thu, 01/26/2017 - 11:12am

Yes, I see what you are saying.

However, P[Falcons win && score is low] = small could be true simply because P[score is low] = small is true, and therefore P[Patriots win && score is low] = small could *also* true. So this doesn't say anything about the Falcons chances in a low scoring game versus a shootout, which is why I interpreted things the way I did.

I wholeheartedly agree that P[Falcons win && score is high] > P[Falcons win && score is low]. But I also think the same thing is true if we replace Falcons with Patriots=.

The statement that many (including the author) seem to be accepting as self-evident is P[Falcons win | score is low] < P[Falcons win | score is high], and I have yet to see a compelling argument why this is so.

25
by dmstorm22 :: Thu, 01/26/2017 - 11:28am

It's an interesting question. Anecdotally, it seems true. I can't remember similar great offense/bad defense teams winning low-scoring games in teh playoffs.

Maybe the '06 Colts winning their first two playoff games 23-8 and 15-6? But in those their defense played fantastic.

The 2011 Patriots probably come closest, winning the AFC Title Game 23-20 and then nearly winning the Super Bowl scoring 17 points.

28
by Joe Pancake :: Thu, 01/26/2017 - 11:45am

I think the '06 Colts vs. Ravens is a good example. I'm sure going into that game, people would have assumed that a low-scoring field-goal-only game favored the #1 defensive Ravens against the #2 offensive Colts.

For the converse of this, you can look at the 2011 49ers-Saints playoff game. Who would have picked Alex Smith over Drew Brees given the game was a shootout?

Obviously these are just two examples, but it would be interesting to go through the historical data and see if low-scoring/high-scoring games disproportionately favor the better offense/defense. My guess is no, but who knows?

44
by MC2 :: Thu, 01/26/2017 - 9:56pm

The best example that springs to mind is the '99 Rams, who beat the Bucs 11-6 in the NFC CG. The same thing happened, to a lesser extent, in the SB, when they won 23-16.

45
by dmstorm22 :: Thu, 01/26/2017 - 11:01pm

They had a really good defense. They scored 520+ points, but they gave up just 242.

They were great on both sides. I wanted to exclude those teams in this - much like hte '07 Patriots won the AFC Title Game 21-12.

46
by MC2 :: Fri, 01/27/2017 - 12:00am

Ah, I somehow missed the "bad defense" part of your comment.

16
by alan frankel :: Wed, 01/25/2017 - 8:38pm

Cian, its been awhile. ATL and PIT are two totally different animals, I suspect that the Pats will try to do what Seattle did to Manning, as ATL's offense has a surprising resemblance to that offense in the passing game(all Jacob Tamme jokes in effect). However, I think the key matchup is going to be between the Patriots safeties and the ATL running backs in the run game.In the past Belichek has allowed balanced offenses to run as much as they wanted hoping that even if the other team averages 5 yards a carry assuming they won't be able to 1)string enough of those runs consecutively to drive down the field, and 2)get a 40 + yard touchdown since he is keeping two safeties back. This system is going to face a stiff test against Atlanta, they have two backs who are 1)Fresh,2) extremely talented in the open field on runs and in the short passing game. I will be writing an article on the Medium about this, but I can't wait to see you guys break it down in the "When Pit/ATL have the ball" format

18
by Raiderjoe :: Wed, 01/25/2017 - 9:46pm

Falcs RBs best 1-2 punch in league. will nto be easy for Pates to stop them. Freeman especially tough to bring down. Both very good recievers. Coleman could probably eveb switch to receiver at some point in career just like Bobby Mitchell and Charley Taylor did.
very interesting to see how Pates will stop the4se two guys

26
by alan frankel :: Thu, 01/26/2017 - 11:28am

In a high scoring game there are more possessions, therefore, the better offense wins out. If one offense scores on 30% of their drives and the other scores on 35% you might not see any real change until the 8th or 9tnth drive. So a high-scoring game with a lot of possessions favors the offense. A low scoring game is similar since possessions are shorter there are more of them so it flips the other way. Any time there are a lot of possessions in the game it favors the superior unit

that's why in the old MJD Jags vs Peyton Manning Colts games, The Jags had a chance since they could limit possessions to offset the Manning advantage

34
by Joe Pancake :: Thu, 01/26/2017 - 2:13pm

Ah, but there is a fallacy in your logic. You are assuming a team's chances of scoring are only dependent on their offense, which isn't the case. They are dependent on the *difference* between their offense and their opponent's defense.

For example, suppose the Falcons have a 35% chance of scoring against a typical defense, and the Patriots have a 30% chance. But further suppose that because the Patriots defense is above average the Falcons only have a 32% chance of scoring against the Patriots, while because the Falcons defense is below average, the Patriots have a 34% chance of scoring against the Falcons.

In this case, over many possessions we would expect the Patriots to ultimately win out -- but not because they have a better offense (they don't), but because they have a better *team*. (In fact, I think that is the case in this Super Bowl.)

So in your MJD Jags vs. Peyton Manning examples, I would argue that limiting possessions was smart, not necessarily because the Colts had a better offense, but because the Colts were just better overall. (Although I also suspect that the randomness of any one football game is such that it renders a lot of this type of theoretical analysis moot. I really don't know.)

36
by deus01 :: Thu, 01/26/2017 - 2:17pm

The randomness of a game will have a bigger impact on a game with fewer possessions. E.g. getting one bounce go your way could be the difference in a low possession game.

35
by deus01 :: Thu, 01/26/2017 - 2:15pm

A low scoring game could have longer possession as each team takes a long time to drive down the field. I forget what game it was but I seem to remember one this year where the first quarter expired with each team having only possession.

Really it's just maximizing your teams efficiency. So if you have good offense you want to score quickly in order to maximize your possessions while the other team wants to prolong their drives in order to minimize your advantage.

37
by PatsFan :: Thu, 01/26/2017 - 2:23pm

The SEA/ATL divisional game.

38
by nat :: Thu, 01/26/2017 - 3:22pm

In 2016,
TD drives averaged 3:38 (7.5 plays)
FG attempt drives averaged 3:50 (8.7 plays)
other drives (not half- or game-ending ones) averaged 2:09 (4.5 plays)

Drives tend to fail faster than they succeed.

It's very style dependent, naturally. Many teams can switch to mad-bomber mode, increasing the drive count at the cost of fewer points per drive on average. Many teams can switch to clock-killing mode, reducing the drive count at a similar cost to per-drive scoring. A few teams can vary the speed of their drives without sacrificing per drive scoring.

But mostly, scoring takes longer than failing to score, resulting in fewer drives overall.