Another Raven Thread with Gobs of Numbers

Share this page

Oswlek

Oswlek

Well-known member
Joined
Feb 9, 2007
Messages
16,993
Reaction score
1,491
Points
113
Age
49
Location
In southcarolina's closet
I'm at my longwinded finest in this thread, so things will be broken up in three posts.

A few years back, I used to determine the "true" point values for playoff teams on offense and defense, and then compare the opponents. 2007 was pretty obvious and CHFF stole my schtick afterward, so I haven't actually done it since 2006, but I thought I'd dig it out again for this matchup.

The methodology is pretty straightforward, just determine the average points scored/allowed by the opposition and compare it to what the team in question allowed/scored. If you scored more than the average team allowed, then you get a plus, same for allowing less than your average opponent.

Let's do Baltimore first.

Balt

Points allowed: 16.6
Average opponent offense: 19.4
Defensive Score: +2.8

No surprise here, Baltimore has an excellent defense.

Points scored: 23.6
Average opponent defense: 20
Offensive Score: +3.6
Total Score: +6.4

This did surprise me, and I retabulated the numbers to make sure I was correct. Does Baltimore have a better offense than most realize? We'll dig into that more later.

Patriots

Points allowed: 21.4
Average opponent offense: 21.3
Defensive Score: -0.1

Points scored: 32.1
Average opponent defense: 22.9
Defensive Score: +9.2
Total Score: +9.1

This seems about right. We can rationalize why the defense might be better than it has been all year, but just going by the numbers, I can't argue with these results. It is notable that, while Balitmore did face some lousy offenses (5 at 15ppg or less and none better than 25), NE faced an equally lousy slate of defenses. Only Pitt allowed less than 19.6 ppg and 5 teams gave up 25 or more per game.

NE taking the total score gives them the best odds to win, but it should be noted that teams with balanced scores typically outperform those dominated by one unit. We have competing predictors in this game, based purely on the numbers.
 
Alright, since I was surprised at Baltimore's offense actually contributing more to their total score than their defense, I decided to take a deeper look into things. My eyes tell me that Baltimore's defense might actually have been a big factor in those offensive points. On top of that, I've also had the impression that Baltimore's offense is reliant on deep passes. Ray, Ray, chuck it, so to speak. To test these theories, I went through the drive log of every Baltimore game checking for the following:

50+ yard, 7+ play TD drives
-50 yard TD drives and defensive TDs
50+ yard, -7 play TD drives

7 is somewhat arbitrary, but it felt right to me and ends up being very useful after crunching everything. On top of that, I checked for 40+/- FG drives, thinking that a FG drive of less than 40 yards is less a credit to the offense and more the defense and special teams. I included missed FGs into these numbers to get the most accurate picture of how the offense performed. This gave me 5 categories:

Long TD drives, Defense credited TDs, Big play TD drives and Long/Short FG drives. Here are the results for Baltimore:

Baltimore

Long TD: 17
Defensive: 11
Big Play: 11
Long FG: 21
Short FG: 13

Interesting, but not informative without anything to relate it to, so here are the numbers for NE:

Patriots

Long TD: 38
Defensive: 7
Big Play: 15
Long FG: 23
Short FG: 7

Now, it really isn't fair to only compare offenses to NE (or GB/NO) so we need another one. I settled on Pitt since they are similar in efficiency to Baltimore, per DVOA, as well as having an offense geared around big plays.

Pittsburgh

Long TD: 21
Defensive: 6
Big Play: 12
Long FG: 29
Short FG: 4

Pitt's problems in generating turnovers really jumps off the page when you view it from this perspective. Only 10 short scoring drives all season.

Based on this exercise, it appears that my suspicions were correct. Despite trailing NE 26-34 in total takeaways, Baltimore had 4 more defense related TDs. The FG numbers are even more dramatic. 33% of Baltimore's scoring drives were less than half the length of the field, neither Pitt or NE was over 15%.

Big plays also accounted for 40% of Baltimore's long TD scores, compared to 36% for Pitt and 28% for NE. The distribution further illuminates Baltimore's reliance on the big play. Here are the plays-per-short-drive for all three teams:

Baltimore
3, 2, 3, 3, 4, 5, 3, 1, 3, 4, 3

Pitt
1, 6, 2, 5, 1, 6, 3, 1, 5, 2, 4, 3

Patriots
6, 1, 4, 5, 6, 5, 4, 6, 6, 1, 4, 5, 2, 6, 5

As I wrote before, the 7 play mark was arbitrary, but still illuminating. Baltimore didn't have a single 6 play 50+ yard TD drive all season, and only one lasting 5 plays. By contrast, Pitt and NE (especially NE) have more even distributions that would alter the numbers considerably had I chosen a different seperation point.

Even further, I randomly chose a couple 7 or 8 play "longer" TD drives to see if a big play still made an appearance, ignoring NE due to their obvious lack of reliance. Of the three chosen for Baltimore, two had plays of 35+ yards. By comparison, the only big plays in the Pitt selections were a fake punt and a 26 yarder that is specifically denoted "pass short" in the PBP.
 
So what does all this mean for Sunday?

If you go by the points for/against analysis, NE would be expected to score between 25-30 and Baltimore somewhere around 24. A close game that could go either way, with the slight edge favoring NE.

Using the second gives a different impression. The odds of Baltimore cracking the 20 point barrier without short fields or big plays is minimal. This is why this game ultimately comes down to pressure. NE is going to use their Marshall Faulk circa 2001 plan to take away Rice, while having a deep safety at all times to stop the long ball. Whether Flacco can make the intermediate plays is all a matter of time, because there will be receivers open.

On the flip side, NE isn't going to rely on the run too much to end up with 3 RB fumbles, so Baltimore going to need to get their turnovers via interceptions. Receiver fumbles are also possible, of course, though less likely to lead to short fields since the pass must have already been completed. This receiver corps is too good for Brady too have time and still feel the need to force the issue.

Pressure. The QB hit the least walks off victorious.

<iframe width="420" height="315" src="http://www.youtube.com/embed/gvll72tHTyQ" frameborder="0" allowfullscreen></iframe>
 
Oswlek said:
I'm at my longwinded finest in this thread, so things will be broken up in three posts.

A few years back, I used to determine the "true" point values for playoff teams on offense and defense, and then compare the opponents. 2007 was pretty obvious and CHFF stole my schtick afterward, so I haven't actually done it since 2006, but I thought I'd dig it out again for this matchup.

The methodology is pretty straightforward, just determine the average points scored/allowed by the opposition and compare it to what the team in question allowed/scored. If you scored more than the average team allowed, then you get a plus, same for allowing less than your average opponent.

Let's do Baltimore first.

Balt

Points allowed: 16.6
Average opponent offense: 19.4
Defensive Score: +2.8

No surprise here, Baltimore has an excellent defense.

Points scored: 23.6
Average opponent defense: 20
Offensive Score: +3.6
Total Score: +6.4

This did surprise me, and I retabulated the numbers to make sure I was correct. Does Baltimore have a better offense than most realize? We'll dig into that more later.

Patriots

Points allowed: 21.4
Average opponent offense: 21.3
Defensive Score: -0.1

Points scored: 32.1
Average opponent defense: 22.9
Defensive Score: +9.2
Total Score: +9.1

This seems about right. We can rationalize why the defense might be better than it has been all year, but just going by the numbers, I can't argue with these results. It is notable that, while Balitmore did face some lousy offenses (5 at 15ppg or less and none better than 25), NE faced an equally lousy slate of defenses. Only Pitt allowed less than 19.6 ppg and 5 teams gave up 25 or more per game.

NE taking the total score gives them the best odds to win, but it should be noted that teams with balanced scores typically outperform those dominated by one unit. We have competing predictors in this game, based purely on the numbers.
Click to expand...



It depends on if you go by yards or points. If by yards the pats played several top 10 defenses, now we always say that points is all that matter, but considering one team that played twice the jets, it seems their offense hurt them more and caused points to be given up than their defense did. What I do know is this, the best offense the ravens played this year blew them out in the chargers.
 
Oswlek said:
So what does all this mean for Sunday?

If you go by the points for/against analysis, NE would be expected to score between 25-30 and Baltimore somewhere around 24. A close game that could go either way, with the slight edge favoring NE.

Using the second gives a different impression. The odds of Baltimore cracking the 20 point barrier without short fields or big plays is minimal. This is why this game ultimately comes down to pressure. NE is going to use their Marshall Faulk circa 2001 plan to take away Rice, while having a deep safety at all times to stop the long ball. Whether Flacco can make the intermediate plays is all a matter of time, because there will be receivers open.

On the flip side, NE isn't going to rely on the run too much to end up with 3 RB fumbles, so Baltimore going to need to get their turnovers via interceptions. Receiver fumbles are also possible, of course, though less likely to lead to short fields since the pass must have already been completed. This receiver corps is too good for Brady too have time and still feel the need to force the issue.

Pressure. The QB hit the least walks off victorious.

<iframe width="420" height="315" src="http://www.youtube.com/embed/gvll72tHTyQ" frameborder="0" allowfullscreen></iframe>
Click to expand...



Well maybe....they talked about this last night on late hits...if the Ravens force all the throws to the outside, meaning they force throws to wideouts on the pats, then the pats do not have a great receiving core. The Ravens will try to take the middle of the field away, not run as much cover 2, and force the pats to make plays to Branch and Ocho, which if we have those odds we might be in trouble.
 
Do these numbers take into account when the defense scores, and when the offense gives up points (like a pick 6)? I've noticed that the numbers at NFL.com don't take that into account. It just lists points scored under it's offensive stats, (even though the defense, and special teams) scored some of the points.

A defense that gets turnovers (and especially if it scores), can make an offense look better than it actually is. Also an offnes that turns the ball over a lot can make a defense look worse than it is.
 
dchester said:
Do these numbers take into account when the defense scores, and when the offense gives up points (like a pick 6)? I've noticed that the numbers at NFL.com don't take that into account. It just lists points scored under it's offensive stats, (even though the defense, and special teams) scored some of the points.
Click to expand...

Read on. That is discussed in depth in the second post.
 
Oswlek said:
Read on. That is discussed in depth in the second post.
Click to expand...

My question wasn't addressed in the second post. It's about the points allowed and points scored numbers that you are showing. Does the points allowed number consist of all points that the team gave up, or is it only the points that the defense gave up?
 
dchester said:
My question wasn't addressed in the second post. It's about the points allowed and points scored numbers that you are showing. Does the points allowed number consist of all points that the team gave up, or is it only the points that the defense gave up?
Click to expand...

Points allowed/scored does not take those figures into consideration because I'm not trying to determine the quality of the offense or defense. From a team-based perspective, it really doesn't matter how points were scored, only that they were.

I also don't account for special team scores as well, because, again, they are pertinent from the hollistic point of view.

In the second post, I do dig into Baltimore's offense, and how much the defense helped things.
 
Oswlek said:
Points allowed/scored does not take those figures into consideration because I'm not trying to determine the quality of the offense or defense. From a team-based perspective, it really doesn't matter how points were scored, only that they were.
Click to expand...

Oh, it seemed like you were trying to draw conclusions about the offenses and defenses from the data. My mistake.
 
Call me surprised that Balt had more long plays than Pitt.
Kudos for all that work, Os!
 
If Baltimore does take away the middle, which seems like a good idea although I don't know if it is possible, what would the Patriots counter with Os? Outside runs with Ridley? Welker and Branch and maybe even Ocho splitting wide?
 
grog said:
If Baltimore does take away the middle, which seems like a good idea although I don't know if it is possible, what would the Patriots counter with Os? Outside runs with Ridley? Welker and Branch and maybe even Ocho splitting wide?
Click to expand...

I'm more of a high level analyst. ;) AWTE is probably a better person to ask, so hopefully he stops by.

I have to think that every single defense the Patriots have faced since Dallas has "taken away the middle". Does anyone really believe that opposing DCs watch film of NE and say to themselves, "we need to focus our coverage outside the numbers"?

Brad Jackson on Baltimore had an interesting comment on FAN yesterday. He said after struggling against Dall/Pitt/NYG, NE went back to the drawing board and started running the offense through the TEs. His contention was that, prior to then Welker was the guy who they designed around, with the TEs cleaning up the scraps.

I have no way of confirming this, but it sounded good to me and Brad was articulate and insightful enough with his other points that I'm inclined to believe him.
 
The number that jumps out at me is Balt having such a high percentage of big plays. The Pats seem to give up easy scores after a big play, either through a DB getting burnt or committing a huge PI. Hopefully the Pats can limit Balt's big plays.
 
bmooney54 said:
The number that jumps out at me is Balt having such a high percentage of big plays. The Pats seem to give up easy scores after a big play, either through a DB getting burnt or committing a huge PI. Hopefully the Pats can limit Balt's big plays.
Click to expand...

Actually Gil Brandt said last night that of the "Big Plays" the pats have given up, which is over 30 yards, they only gave up TD's on two. As with other games, this comes down to what the offense does, if the offense is moving, if they are scoring and not turning the ball over, the pats defense is not going to lose the game, they have not all year.
 
bmooney54 said:
The number that jumps out at me is Balt having such a high percentage of big plays. The Pats seem to give up easy scores after a big play, either through a DB getting burnt or committing a huge PI. Hopefully the Pats can limit Balt's big plays.
Click to expand...

1 thing in our favor is this group of refs calls PI less than any other group this year.
 
midgar8784 said:
Actually Gil Brandt said last night that of the "Big Plays" the pats have given up, which is over 30 yards, they only gave up TD's on two. As with other games, this comes down to what the offense does, if the offense is moving, if they are scoring and not turning the ball over, the pats defense is not going to lose the game, they have not all year.
Click to expand...

Exactly. Despite giving up less than half the big plays that NE did, Pitt actually allowed more big play TDs. This is because the Patriots don't give up the "guy running free down the sideline" big plays (which are the ones that Baltimore thrives on) they tend to give up a 15 yard seam route that turns into 42 with a missed tackle.
 
You must log in or register to reply here.
Back
Top