Andy Benoit's podcast on Sunday night, released Monday I think, discussed many topics. One of the topics was his initial thoughts on the Pats D- ATL O matchup. His concern was that you can't really scheme effectively against a team like the Falcons because their skill position players are so good. PIT had the killer Bs but their depth, aside from Williams, wasn't that great. PIT's receiving options, aside from Brown and, to a lesser degree, Rogers (Rodgers?) are unreliable JAGs or replacement level players.
When the Falcons Have the Ball...
- Thread starter Morgan's Magic Snowplow
- Start date
Sanu is good (he's ALWAYS on my fantasy team, I actually really like him), but he's not a world beater: 59 catches, 650 yards, then no WR/TE with more than 35. Rogers is 50 for 600.
Steelers top 3 non RBs behind Brown have 108 catches, Falcons top 3 non-RBs after Jones have 114.
Steelers 2 RBs have more catches than the Falcons 2 RBs and about the same amount of rushing yards.
None of this is to say the Falcons aren't REALLY good mind you.
something something Jamie Collins something.
I like Vehei at FO, but yeah, I agree with Soxy that he's talking about getting caught between the two, when Belichick's thing is to pick the one he thinks is more dangerous and challenge the opponent to beat him with the rest.
This screams to me why numbers like this don't tell us much without the game plan and how plays, even if looking to succeed, are used to set up other plays too.
I understand that the FO DVAO numbers are regular season only but I'm having trouble reconciling the prevailing consensus that the Falcons have an historically great offense with their score of 25.3%.
Yes, that's very strong and good enough for not only best in the league this year but also best since 2013. But a) it's only the 21st best oDVAO in the last 20 years, b) there's been at least a dozen teams with oDVAO above 30 (and 2 Pats teams above 40) in that span and c) the last team with a higher oDVAO to play in the SB were the '13 Broncos who got mauled.
So my question is: is the "we've never seen anything like Matty Ice, JuJo, Sanu, Freeman, Coleman et al and they're just too quick to be stopped" meme just SB hype or is DVOA just flawed in some way?
Yes, that's very strong and good enough for not only best in the league this year but also best since 2013. But a) it's only the 21st best oDVAO in the last 20 years, b) there's been at least a dozen teams with oDVAO above 30 (and 2 Pats teams above 40) in that span and c) the last team with a higher oDVAO to play in the SB were the '13 Broncos who got mauled.
So my question is: is the "we've never seen anything like Matty Ice, JuJo, Sanu, Freeman, Coleman et al and they're just too quick to be stopped" meme just SB hype or is DVOA just flawed in some way?
The Atlanta offense is nothing historic. It's SB hype. And yes, dvoa is flawed but that's a different topic.
Here's the list of the top oDVOA's for the past 20 seasons if anyone's interested.
1. 2007 NEP 43.5 (Brady, Maroney, Moss, Welker, Stallworth, Watson)
2. 2010 NEP 42.2 (Brady, BJGE, Welker, Branch, Gronk/AH)
3. 2002 KCC 35.4 (T Green, P Holmes, T Gonzalez, E Kennison, M Boerigter)
4. 1998 DEN 34.5 (Elway, B Brister (4gs), T Davis, Rd Smith, E McCaffrey, Sh Sharpe)
5. 2011 GBP 33.8 (Rodgers, R Grant, J Starks, J Nelson, G Jennings, J Finley, Jm Jones, D Driver)
6. 2013 DEN 33.5 (P Manning, K Moreno, M Ball, De Thomas, E Decker, Ju Thomas, Welker)
7. 2003 KCC 33.4 (same as 2002 + J Morton)
8. 2011 NEP 31.9 (same as 2010 + Ridley)
9. 2004 IND 31.8 (P Manning, E James, R Wayne, M Harrison, Br Stokley, D Clark, M Pollard)
10. 2004 KCC 31.6 (same as 2003 + L Johnson, D Blaylock - Boerigter)
11. 2012 NEP 30.8 (same as 2011 + Lloyd + Edelman - BJGE)
12. 2011 NOS 30.3 (Brees, M Ingram, P Thomas, D Sproles, J Graham, M Colston, L Moore, R Meachem)
13. 1998 SFF 28.8 (Young, G Hearst, T Kirby, J Rice, T Owens, JJ Stokes, I Smith)
14. 2006 IND 28.5 (same as 2004 + J Addai + D Rhodes - E James - M Pollard)
14. 2005 SEA 28.5 (M Hasselbeck, S Alexander, M Morris, B Engram, J Jurevicius, J Stevens, D Jackson)
16. 2005 DEN 26.9 (J Plummer, M Anderson, T Bell, Rd Smith, A Lelie, J Putzier)
17. 2000 SLR 26.7 (K Warner, M Faulk, J Watson, T Holt, I Bruce, A-Z Hakim, R Proehl)
18. 2000 IND 26.4 (P Manning, E James, M Harrison, J Pathon, T Wilkins, K Dilger, M Pollard)
19. 2009 NEP 26.4 (same as 2007 + K Faulk + S Morris + Edelman - Stallworth)
20. 2006 SDC 25.7 (P Rivers, L Tomlinson, M Turner, A Gates, E Parker, V Jackson, K McCardell)
21. 2016 ATL 25.3 (M Ryan, D Freeman, T Coleman, JuJones, M Sanu, T Gabriel, A Robinson)
1. 2007 NEP 43.5 (Brady, Maroney, Moss, Welker, Stallworth, Watson)
2. 2010 NEP 42.2 (Brady, BJGE, Welker, Branch, Gronk/AH)
3. 2002 KCC 35.4 (T Green, P Holmes, T Gonzalez, E Kennison, M Boerigter)
4. 1998 DEN 34.5 (Elway, B Brister (4gs), T Davis, Rd Smith, E McCaffrey, Sh Sharpe)
5. 2011 GBP 33.8 (Rodgers, R Grant, J Starks, J Nelson, G Jennings, J Finley, Jm Jones, D Driver)
6. 2013 DEN 33.5 (P Manning, K Moreno, M Ball, De Thomas, E Decker, Ju Thomas, Welker)
7. 2003 KCC 33.4 (same as 2002 + J Morton)
8. 2011 NEP 31.9 (same as 2010 + Ridley)
9. 2004 IND 31.8 (P Manning, E James, R Wayne, M Harrison, Br Stokley, D Clark, M Pollard)
10. 2004 KCC 31.6 (same as 2003 + L Johnson, D Blaylock - Boerigter)
11. 2012 NEP 30.8 (same as 2011 + Lloyd + Edelman - BJGE)
12. 2011 NOS 30.3 (Brees, M Ingram, P Thomas, D Sproles, J Graham, M Colston, L Moore, R Meachem)
13. 1998 SFF 28.8 (Young, G Hearst, T Kirby, J Rice, T Owens, JJ Stokes, I Smith)
14. 2006 IND 28.5 (same as 2004 + J Addai + D Rhodes - E James - M Pollard)
14. 2005 SEA 28.5 (M Hasselbeck, S Alexander, M Morris, B Engram, J Jurevicius, J Stevens, D Jackson)
16. 2005 DEN 26.9 (J Plummer, M Anderson, T Bell, Rd Smith, A Lelie, J Putzier)
17. 2000 SLR 26.7 (K Warner, M Faulk, J Watson, T Holt, I Bruce, A-Z Hakim, R Proehl)
18. 2000 IND 26.4 (P Manning, E James, M Harrison, J Pathon, T Wilkins, K Dilger, M Pollard)
19. 2009 NEP 26.4 (same as 2007 + K Faulk + S Morris + Edelman - Stallworth)
20. 2006 SDC 25.7 (P Rivers, L Tomlinson, M Turner, A Gates, E Parker, V Jackson, K McCardell)
21. 2016 ATL 25.3 (M Ryan, D Freeman, T Coleman, JuJones, M Sanu, T Gabriel, A Robinson)
Looking at Passing DVOA (pDVOA) they fair a bit better (11th best in the past 20 years with 53.0) but this year's Pats are right behind (50.5 for 15th). Aside from the usual suspects we have Steve McNair's '03 Titans in 8th with 54.7!
NEP 4 of the top 11 of all time. And they won the Super Bowl zero times those four seasons.
Not just NE. Only one team of the top 13 won the SB ('98 Broncos) and only one other in the top 20 ('06 Colts - and only because they got to play the Bears). The 3 Chiefs teams in the top 10 made the playoffs once and lost their only game.
The sorry tale of the top 20 oDVOAs:
Failed to make the playoffs(2): #3 KCC'02, #10 KCC'04
Lost in the WC(3): #17 SLR'00, #18 IND'00, #19 NEP'09
Lost in the DR(7): #2 NEP'10, #5 GBP'11, #7 KCC'03, #9 IND'04, #12 NOS'11, #13 SFF'98, #20 SDC'06
Lost in the CC(2): #11 NEP'12, #16 DEN'05
Lost in the SB(4): #1 NEP'07, #6 DEN'13, #8 NEP'11, #15 SEA'05
Won SB(2): #4 DEN'98, #14 IND'06
So overall they made the playoffs 18/20 times, were 4-3 in the WC Round, 8-6 in DR, 6-2 in CC and 2-4 in the SB.
Last edited:
And yet four of the top ten defenses by DVOA won the Super Bowl: 2002 TB (2nd), 2008 PIT (3rd), 2013 SEA (7th), 2015 DEN (8th).
Not sure if that really means much though.
Not sure if that really means much though.
Their running game balances their receiving game very well.
Freeman is constant and steady but in games where Julio Jones was not a factor or was contained by the defense, Coleman has had some of his biggest games. Combined the two running backs are nasty to the extreme.
Week 3 Win versus New Orleans
Julio Jones – 1 reception for 16 yards, 0 touchdowns
Devonta Freeman – 152 yards rushing, 5 receptions for 55 yards, 1 receiving touchdown
Tevin Coleman – 42 yards rushing, 3 receptions for 47 yards, 3 rushing touchdowns
Combined RBs - 296 total yards, 4 touchdowns
Week 5 Win versus Denver
Julio Jones – 2 receptions for 29 yards, 0 touchdowns
Devonta Freeman – 88 yards rushing, 3 receptions for 35 yards, 1 rushing touchdown
Tevin Coleman – 31 yards rushing, 4 receptions for 132 yards, 1 receiving touchdown
Combined RBs - 286 total yards, 2 touchdowns
Week 8 Win versus Green Bay
Julio Jones – 3 receptions for 29 yards, 0 touchdowns
Devonta Freeman – 35 yards rushing, 4 receptions for 23 yards, 1 receiving touchdown
Tevin Coleman – Did Not Play – Injury
Combined RBs - 58 total yards, 1 touchdown
Week 12 Win versus Arizona
Julio Jones – 4 receptions for 35 yards, 0 touchdowns
Devonta Freeman – 60 yards rushing, 2 receptions for 17 yards, 2 rushing touchdowns
Tevin Coleman – 9 yards rushing, 2 receptions for 4 yards, 1 rushing touchdown
Combined RBs - 90 total yards, 3 touchdowns
Week 16 Win versus Carolina
Julio Jones – 4 receptions for 60 yards, 0 touchdowns
Devonta Freeman – 53 yards rushing, 8 receptions for 35 yards, 0 touchdowns
Tevin Coleman – 90 yards rushing, 3 receptions for 45 yards, 1 rushing touchdown
Combined RBs - 223 total yards rushing, 1 touchdown
Freeman is constant and steady but in games where Julio Jones was not a factor or was contained by the defense, Coleman has had some of his biggest games. Combined the two running backs are nasty to the extreme.
Week 3 Win versus New Orleans
Julio Jones – 1 reception for 16 yards, 0 touchdowns
Devonta Freeman – 152 yards rushing, 5 receptions for 55 yards, 1 receiving touchdown
Tevin Coleman – 42 yards rushing, 3 receptions for 47 yards, 3 rushing touchdowns
Combined RBs - 296 total yards, 4 touchdowns
Week 5 Win versus Denver
Julio Jones – 2 receptions for 29 yards, 0 touchdowns
Devonta Freeman – 88 yards rushing, 3 receptions for 35 yards, 1 rushing touchdown
Tevin Coleman – 31 yards rushing, 4 receptions for 132 yards, 1 receiving touchdown
Combined RBs - 286 total yards, 2 touchdowns
Week 8 Win versus Green Bay
Julio Jones – 3 receptions for 29 yards, 0 touchdowns
Devonta Freeman – 35 yards rushing, 4 receptions for 23 yards, 1 receiving touchdown
Tevin Coleman – Did Not Play – Injury
Combined RBs - 58 total yards, 1 touchdown
Week 12 Win versus Arizona
Julio Jones – 4 receptions for 35 yards, 0 touchdowns
Devonta Freeman – 60 yards rushing, 2 receptions for 17 yards, 2 rushing touchdowns
Tevin Coleman – 9 yards rushing, 2 receptions for 4 yards, 1 rushing touchdown
Combined RBs - 90 total yards, 3 touchdowns
Week 16 Win versus Carolina
Julio Jones – 4 receptions for 60 yards, 0 touchdowns
Devonta Freeman – 53 yards rushing, 8 receptions for 35 yards, 0 touchdowns
Tevin Coleman – 90 yards rushing, 3 receptions for 45 yards, 1 rushing touchdown
Combined RBs - 223 total yards rushing, 1 touchdown
Wonder if it says more about roster construction (build for depth, don't overspend on premium guys) than anything else.
But the 11th best dDVOA, the '00 Ravens, also won a championship.
Compared with the top 20 offenses, 2 more of the defences failed to make the playoffs (#3 Buf'04, #5 Chi'12, #15 Bal'99 and #18 Bal'04) but their playoff records were better: 5-1 in WC, 11-4 in DR, 7-4 in CC and 5-2 in SB for a .718 win pct vs .571 for the defences.
1 tbb 02 31.8 ch
2 Pit 08 29.0 ch
3 Buf 04 28.5 f 9-7
4 Bal 08 27.8 cc
5 Chi 12 26.7 f 10-6
6 Sea 13 25.9 ch
7 Den 15 25.8 ch
8 NYJ 09 25.5 cc
9 ten 00 25.0 dr
10 Bal 03 25.0 wc
11 bal 00 23.8 ch
12 Phi 08 23.6 cc
13 Bal 06 23.0 dr
14 Mia 98 22.4 dr
15 Bal 99 20.8 f 8-8
16 Pit 10 20.7 ru
17 Chi 06 20.1 ru
18 Bal 04 19.9 f 9-7
19 tbb 99 19.4 cc
20 chi 05 19.4 dr
2 Pit 08 29.0 ch
3 Buf 04 28.5 f 9-7
4 Bal 08 27.8 cc
5 Chi 12 26.7 f 10-6
6 Sea 13 25.9 ch
7 Den 15 25.8 ch
8 NYJ 09 25.5 cc
9 ten 00 25.0 dr
10 Bal 03 25.0 wc
11 bal 00 23.8 ch
12 Phi 08 23.6 cc
13 Bal 06 23.0 dr
14 Mia 98 22.4 dr
15 Bal 99 20.8 f 8-8
16 Pit 10 20.7 ru
17 Chi 06 20.1 ru
18 Bal 04 19.9 f 9-7
19 tbb 99 19.4 cc
20 chi 05 19.4 dr
Before we accept DVOA as gospel though we should also validate that it is truly capturing the top 20 offenses. What does it say that the Greatest Show on Turf offense was only #17 in DVOA, even with the highest rushing DVOA in history?
That team had a very good but not historically great passing DVOA (they were only 3rd in passing that season). Warner threw a lot of INTs in 2000: 18 in only 11 starts. Also they played a weak schedule that knocked them down a little.
Any sense of how serious this injury is to Mack and what this might do for the Pats' interior rush?
It's an ankle sprain - he played well against Green Bay after it happened, so I'm not expecting it will bother him after two weeks of rest.
I didn't watch much of that game but I did happen to see that play. It didn't look like much on replay, and he only missed one play.
He's not practicing this week so it's not nothing.
Also yeah he returned after one play but you have lots of adrenaline at that point. Could be the type of thing that is more painful after the game.
Who knows. But I'll be watching Branch/Brown vs. Mack a lot next Sunday (and Flowers on passing downs).
Also yeah he returned after one play but you have lots of adrenaline at that point. Could be the type of thing that is more painful after the game.
Who knows. But I'll be watching Branch/Brown vs. Mack a lot next Sunday (and Flowers on passing downs).
Neither is Jones. Makes no sense to practice either of them this week. Not nothing, but certainly not serious enough to worry about I wouldn't think.
Might we see Matt Ryan legging out a few carries for 1st downs? Two or three 3rd down conversions at key points of the game could be damaging.
90 total yards from Freeman/Coleman and then 35 yards from Jones - I'd certainly take those numbers next week. Interesting that they got 3 TD's though. Gabriel, Sanu and Hardy all had better games than Jones, combining for 15 rec and 187 yds and 2 td's. That game has a weird box score - the Falcons must have been hyper efficient to get 38 points. It doesn't look like there were any big punt/kick returns, and only 1 fumble by the Cards (and recovered by Cards) along with 1 int by Ryan. There was also only 47 rushing yards outside of Freeman and Coleman.
http://www.nfl.com/gamecenter/2016112700/2016/REG12/cardinals@falcons#menu=gameinfo|contentId:0ap3000000746899&tab=analyze&analyze=boxscore
IIRC, the initial injury report said Mack DNP because of an "ankle". Somewhere in the middle of the week the wording was changed and it is now listed as "fibula". A ligament injury (ankle sprain) being changed to a bone injury is not likely good news. Maybe I'm reading too much into it. We'll know for sure if he misses practice next week. I couldn't find any photos of Mack from this week. Is he in a boot?
I noticed this too, plus it's the same fibula he broke and had season-ending surgery on in 2014.
He'll get shot up and play so he should be OK for the 1st half. What will be tricky is when the injury stiffens up on him during the long halftime. I imagine he'd have to either put a heating pad on it or ride the stationary bike.
I'm guessing Dimitroff felt encouraged toward greater accuracy by hearing all week about possible Seattle/Pittsburgh sanctions for inaccurate injury reports.
When the Falcons have the ball I'm likely to consume beers at roughly twice the rate I do while they are on defense.
Is there a good place to find stats about the Falcons formation tendencies? For example, how often do they use one TE? Two? Three? How about when they use Coleman and Freeman in the same backfield? Do they ever use a FB? 3 WRs? 4? Etc
I've been wondering if the Falcons could pull a page from the Pats playbook against the Steelers and go with a 20 set with both Coleman and Freeman plus 3 WRs. Spread out the Pats, make them go dime, and then run.
I've been wondering if the Falcons could pull a page from the Pats playbook against the Steelers and go with a 20 set with both Coleman and Freeman plus 3 WRs. Spread out the Pats, make them go dime, and then run.
Interesting article on the Pats D. Seems like Chung and McCourty and the Queens of the chessboard.
Fighting Flexibility With Flexibility
Fighting Flexibility With Flexibility
I doubt we see Jones used this way. He's played 64 defensive snaps all season and 32 of those were back in Week 5 against Cleveland. Plus that would be a small lineup that could be run on.
A team whose QB is clearly inferior to Matt Ryan (bot not awful) that also has had a fairly diverse and productive group of offensive weapons is the 2014 Cincinnati Bengals. Obviously these offenses are different in terms of production, but there's some overlap in terms of receiving depth:
AJ Green
Marvin Jones
Mohammed Sanu
Giovani Bernard
Jeremy Hill
Tyler Eifert
Can people smarter than I - there are lots of you - remember how the Pats approached defending that's group? I realize Matt Ryan is light years ahead of Red Rifle, so maybe this is entirely irrelevant to think about, but I figured I'd put it out there. Thanks.
AJ Green
Marvin Jones
Mohammed Sanu
Giovani Bernard
Jeremy Hill
Tyler Eifert
Can people smarter than I - there are lots of you - remember how the Pats approached defending that's group? I realize Matt Ryan is light years ahead of Red Rifle, so maybe this is entirely irrelevant to think about, but I figured I'd put it out there. Thanks.
No Eifert yet but they did have Gresham who at that time was probably a better player than what Atlanta has at TE. Marvin Jones was hurt and didn't play in the game.
Box score and game recap here.
That was after the week of "Jimmy Garoppolo should be seriously looked at for first team QB" to "Brady's all done" from the likes of Donovan McNabb and Trent Dilfer. Patriots would have run through Trump's border wall that night.
@dbn took a statistical look at the Patriots' "bend but don't break" D in his ITP debut:
http://insidethepylon.com/nfl/teams-nfl/afc-east/new-england-patriots/2017/02/01/new-england-patriots-bend-dont-break-defense-myth-reality/
http://insidethepylon.com/nfl/teams-nfl/afc-east/new-england-patriots/2017/02/01/new-england-patriots-bend-dont-break-defense-myth-reality/
I thoroughly enjoyed that article. I wonder if another contributing factor to the phenomenon of a low points-allowed to yards-allowed ratio could be the amount of time that the Patriots are ahead in the score, which makes the opponent more predictable (pass-heavy) and therefore easier to defend downfield (e.g., play action does not work so well)? In this situation you would expect the following:
- Less of a pronounced BbdB differential in the first quarter, before there is a large lead
- Less of a pronounced BbdB differential in seasons where the Patriots had more close games and/or losses
That is a good suggestion. If I can get ahold of raw data (and thus can break it down by quarter) I'll take a look.
FO does provide "average lead at beginning of a defensive drive" data, but averaged over the entire season. I didn't use it in the analysis, for the practical (read: lazy) reason that it was in a separate table on their website and I hadn't grabbed that data, but I did eyeball the numbers for the seasons where the Patriots were particularly BBdB, and nothing stood out to me.
Aside: every season, the NFL-averaged "average lead at beginning of a defensive drive" isn't zero as I'd expect, but rather always positive (typically ~ +1 point). I had to think for several minutes about how this could be. I came up with:
FO states that take-a-knee drives have been excluded from their data. If the winning team has the ball last and takes a knee, that drive is excluded from their opponent's (i.e., the one behind) defensive drive stats. If the losing team has the ball last and the clock runs out while they are trying to score, that drive is included in the opposing (i.e., the one with the lead) team's defensive drive stats.
If the Pats kick off Sunday, they'll have 0 at the beginning of the defensive drive. If the Falcons go down and score, they'll be +7 at the beginning of their first defensive drive. Then if the Pats scored a TD, back to 0 on NE's 2nd defensive drive. Then the Falcons score a FG, it's 3. The average is positive because the team that just had the ball might have scored.
Hmm, I think you are correct. While it's easy to have scenarios where, for any given game, the average (over all drives for both teams) defensive lead is exactly zero, it's much, much easier to have it be net positive than it is to have it net negative. Sparing everyone a long explanation, every game starts with a drive where the lead is zero, and then the first drive after a score it is positive. From there it can alternate in equal magnitude but opposite sign, or alternate from positive to zero, but only if the team with the ball last in the 1st half and first in the 2nd half is just right, it can't be net-negative. So, over the course of a 256 game NFL season, I suspect you have many games where it is zero, some where it is positive, and very few where it is negative.
Thanks for the article -- it clearly presents the data and analysis. I especially enjoyed your use of Gnuplot to create the graphics 
PGPLOT with a C wrapper. I'm old-school.Thanks for the article -- it clearly presents the data and analysis. I especially enjoyed your use of Gnuplot to create the graphics
Yeah, that's really good stuff. I think the money quote is this:
That is such an important point that seems to get lost, even in some of the discussion and commentary I read from intelligent football writers. Yes, the Pats defensive numbers are artificially inflated due to having the benefits of a consistently great offense and special teams. So what? That offense and special teams ain't going anywhere. It's, essentially, a baked-in part of the defense. That's what Belichick means when he talks about complimentary football. Offense, defense, special teams.... to act like any of these units perform in isolation of the other two units is a fallacy that needs to be destroyed. It's, like, all one great big beautiful whole, man.