Writers of Pro Football Prospectus 2008

Most Recent FO Features

JohnsonKer18.jpg

» Seventh Day Adventure: Week 13

The biggest game this week is the Iron Bowl, where the playoff hopes of Alabama, Auburn, and Georgia hang in the balance.

08 Feb 2016

Super Bowl 50 Quick Reads

by Vincent Verhei and Aaron Schatz

As usual, we're running team DVOA ratings from the Super Bowl along with individual numbers for Quick Reads.

One of the big stories for this game was Denver's extreme fortune recovering fumbles. Super Bowl 50 had seven different fumbles, and only two of them were recovered by Carolina. With that kind of fumble luck, some readers might have expected DVOA to actually put Carolina slightly ahead for the game. But no, that's not the case. In both DVOA and VOA without opponent adjustments, Denver did end up as the better team for the night. However, what's surprising is which unit gave Denver that advantage:

DVOA (with opponent adjustments)
TEAM TOT OFF DEF ST
DEN 28% -50% -58% 21%
CAR 12% -30% -58% -17%
VOA (no opponent adjustments)
TEAM TOT OFF DEF ST
DEN 3% -67% -49% 21%
CAR -9% -58% -66% -17%

That's right: according to DVOA, as great as Denver's defense was in Super Bowl 50, Carolina's defense was just as great. And that's after opponent adjustments. Before opponent adjustments, our system actually rated Carolina with an even better game than Denver. After all, Carolina gained more yards per play (4.2 vs. 3.5) and converted more often on third down (3-of-15 vs. 1-of-14).

No, the real difference here came on special teams. Denver was above-average in all five phases of special teams that we measure, while Carolina had a missed field goal from Graham Gano and was poor on punt coverage and punt returns. Britton Colquitt and Brad Nortman both had excellent nights with similar gross averages: 45.9 yards for Colquitt, 45.0 yards for Nortman. However, that weird 61-yard punt return from Jordan Norwood was a huge change in field position, while Ted Ginn's three returns went for a combined 2 yards. Also, Nortman's gross average wasn't quite as valuable as Colquitt's because Nortman's longest punt, 61 yards, ended up netting only 41 because of a touchback.

As for individual numbers, the main issue there is the same as it has been for so much of the 2015 season: how did the Broncos keep winning with a quarterback playing as badly as Peyton Manning played in his final season? And as we all saw, the answer is partly that their defense scores nearly as many points as it allows. Scott Kacsmar pointed out last week that Manning has often played well in postseason defeats, so it is somewhat ironic that he has spent most of this season playing poorly in victory, and the Super Bowl was no exception.

Now, before we go any further, let's set a few ground rules. There have been 54 Super Bowl quarterbacks since 1989, and 27 winners; one of them has to be the worst. And any player would rather be the worst at his position to win a Super Bowl, and not the best player at his position to not win one. But with those ground rules established, we can now say that by almost any measure, Manning is the worst quarterback to win a Super Bowl in at least a quarter-century. According to DYAR, Manning finishes with one of the ten worst Super Bowls for a quarterback we have on record, and he passes Ben Roethlisberger against the Seahawks in Super Bowl XL for worst DYAR for a winning quarterback, and the worst DYAR for any quarterback in 15 years. Manning's counterpart in Super Bowl 50, Cam Newton, also makes the bottom 10, which helps explain Manning's victory. After all, one of them had to win:


Worst Super Bowl Quarterbacks by DYAR, 1989-2015 (Min. 10 passes)
Player Year Team Opp. Result Att. Cmp Pct. Yds TD INT Sk-Yds NY/PP Runs-
YD-TD
Fmb-L PASS
DYAR
RUN
DYAR
TOT
DYAR
Neil O'Donnell 1995 PIT DAL L 27-17 49 28 57% 239 1 3 4-32 3.9 1-0-0 0-0 -30 0 -30
Rex Grossman 2006 CHI IND L 29-17 28 20 71% 165 1 2 1-11 5.3 2-0-0 2-1 -42 0 -42
Cam Newton 2015 CAR DEN L 24-10 41 18 44% 265 0 1 6-64 4.3 6-45-0 2-2 -73 22 -51
Ben Roethlisberger 2005 PIT SEA W 21-10 21 9 43% 123 0 2 1-8 5.2 7-25-1 0-0 -78 20 -58
Rich Gannon 2002 OAK TB L 48-21 44 24 55% 272 2 5 5-22 5.1 2-3-0 1-0 -55 -6 -61
Chris Chandler 1998 ATL DEN L 34-19 35 19 54% 219 1 3 2-13 5.6 4-30-0 0-0 -77 12 -65
Frank Reich* 1992 BUF DAL L 52-17 31 18 58% 194 1 2 2-12 5.5 2-0-0 3-2 -75 0 -75
Peyton Manning 2015 DEN CAR W 24-10 23 13 57% 141 0 1 5-37 3.7 0-0-0 2-1 -115 0 -115
John Elway 1989 DEN SF L 55-10 26 10 38% 108 0 2 4-30 2.6 4-8-1 2-0 -128 10 -118
Kerry Collins 2000 NYG BAL L 34-7 39 15 38% 112 0 4 4-26 2.0 3-12-0 1-0 -295 1 -294
* Did not start game

Remember, these DYAR numbers are adjusted for the fact that Manning and Newton were playing against this year's top two defenses. Without adjustments, things would be even worse.

If you read Quick Reads every week (and you should), you'll know that Manning wasn't especially effective against Pittsburgh or New England in the playoffs either. We've already established that no team since 1989 has won a championship with a quarterback worse than Manning in the regular season or in the Super Bowl itself, but Manning's three-game playoff total of minus-99 DYAR is also the worst of any quarterback who won a Lombardi Trophy, the worst of any quarterback in 19 years, and the second-worst of any quarterback who won a conference championship.


Worst Playoff Runs by DYAR, Super Bowl Quarterbacks, 1989-2015 (Minimum 30 Passes)
Player Year Team SB Result G Att. Cmp Pct. Yds TD INT Sk-Yds NY/PP Runs-Yd-TD Fmb-L PASS
DYAR
RUN
DYAR
TOT
DYAR
Chris Chandler 1998 ATL Lost 3 97 59 61% 728 4 4 7-52 6.5 9-52-0 0-0 100 12 112
Ben Roethlisberger 2010 PIT Lost 3 91 54 59% 622 4 4 9-48 5.7 21-63-1 3-1 44 25 69
Russell Wilson 2014 SEA Lost 3 72 41 57% 724 6 5 10-58 8.1 17-86-1 2-0 22 24 47
Jim Kelly 1991 BUF Lost 3 118 64 54% 665 5 9 7-59 4.8 6-27-0 3-1 10 7 17
Trent Dilfer 2000 BAL Won 4 73 35 48% 590 3 1 10-82 6.1 13-7-0 3-0 14 -9 5
Kerry Collins 2000 NYG Lost 3 97 55 57% 618 5 6 5-26 5.8 11-29-0 1-0 -7 7 0
Neil O'Donnell 1995 PIT Lost 3 125 72 58% 706 3 6 5-32 5.2 6--1-0 1-0 -1 -23 -24
Rex Grossman 2006 CHI Lost 3 92 52 57% 591 3 3 4-42 5.7 4--3-0 3-2 -67 0 -67
Peyton Manning 2015 DEN Won 3 69 38 55% 398 2 0 4-38 4.9 5-10-0 2-1 -106 6 -99
Drew Bledsoe 1996 NE Lost 3 105 59 56% 595 3 7 9-71 4.6 3-4-0 2-1 -125 -27 -152

Quarterbacks
Rk
Player
Team
CP/AT
Yds
TD
INT
Sacks
Total
DYAR
Pass
DYAR
Rush
DYAR
Opp
1.
Cam Newton CAR
18/41
265
0
1
6
-51
-73
22
DEN
Newton's two sack-fumbles -- one recovered for a Denver touchdown, the other recovered to set Denver up at the Carolina 4 -- were probably the two biggest plays in the game, and certainly the biggest negative on Newton's day. Inside the Denver 30, Newton went 1-of-6 for 13 yards with an interception. On third downs, he went 4-of-10 for 61 yards with two conversions and four sacks, including the two fumble plays. Also, since we won't mention Greg Olsen in the receiving tables, we should say here how Denver completely neutralized the Carolina tight end. In nine targets, Olsen had four receptions for 41 yards and only one first down. His six rushes included first downs on second-and-11, second-and-10, and third-and-2.
2.
Peyton Manning DEN
13/23
141
0
1
5
-115
-115
0
CAR
Manning had two first downs on Denver's first drive, two more on their first drive of the second half, and only one other all day. Inside the Carolina 37, he went 2-of-6 for 8 yards with no first downs, one interception, and one sack-fumble. Surprisingly, he was at his best when throwing deep, though deep here is a relative term. His deepest pass went 25 yards, and his deepest completion was caught 20 yards downfield. But on throws that traveled at least 9 yards past the line of scrimmage, he went 5-of-8 for 103 yards and all five first downs. His first third-down play was a 22-yard gain on third-and-4. That was his last third-down conversion of the day; after that, he went 1-of-7 for 1 yard (not a typo) with two sacks, a fumble, and an interception. He was horrible throwing to running backs all year, and that carried into the Super Bowl, where he went 4-of-5 for 10 yards and no first downs throwing to Ronnie Hillman and C.J. Anderson.


Five Best Running Backs by DYAR (Total)
Rk
Player
Team
Runs
Rush
Yds
Rush
TD
Rec
Rec
Yds
Rec
TD
Total
DYAR
Rush
DYAR
Rec
DYAR
Opp
1.
Fozzy Whittaker CAR
4
26
0
1/1
6
0
16
13
3
DEN
Whittaker's big runs were a 15-yard gain on first-and-10 and a 10-yard gain on second-and-10.
2.
Jonathan Stewart CAR
12
29
1
1/2
-1
0
10
16
-6
DEN
Stewart had Carolina's only touchdown, but he had only two other first downs on the day. His longest run gained 12 yards, but he was hit for no gain four times and gained 1 yard twice. That's a median gain of 1.5 yards per carry. And that's bad.
3.
Ronnie Hillman DEN
5
0
0
0/1
0
0
-19
-14
-4
CAR
None of Hillman's carries picked up a first down, three lost yardage, and one was a 2-yard gain on third-and-17.
4.
C.J. Anderson DEN
23
90
1
4/4
10
0
-24
-15
-9
CAR
Anderson had runs of 34, 13, and 12 yards, and Denver's sole offensive touchdown, but those were his only first downs on the day. Meanwhile, 17 of his 23 carries gained 2 yards or less, and eight went for no gain or a loss, including a failure to convert on third-and-1.
5.
Mike Tolbert CAR
5
18
0
0/1
0
0
-33
-30
-3
DEN
Tolbert's five carries, in order: 2-yard gain on first-and-10, fumble recovered by Carolina; 11-yard gain on first-and-10, fumble recovered by Denver; 1-yard gain on first-and-10; 3-yard gain on second-and-7; 1-yard gain on first-and-10.


Five Best Running Backs by DYAR (Rushing)
Rk
Player
Team
Runs
Rush
Yds
Rush
TD
Rec
Rec
Yds
Rec
TD
Total
DYAR
Rush
DYAR
Rec
DYAR
Opp
1.
Jonathan Stewart CAR
12
29
1
1/2
-1
0
10
16
-6
DEN
2.
Fozzy Whittaker CAR
4
26
0
1/1
6
0
16
13
3
DEN
3.
Ronnie Hillman DEN
5
0
0
0/1
0
0
-19
-14
-4
CAR
4.
C.J. Anderson DEN
23
90
1
4/4
10
0
-24
-15
-9
CAR
5.
Mike Tolbert CAR
5
18
0
0/1
0
0
-33
-30
-3
DEN


Five Best Wide Receivers and Tight Ends by DYAR
Rk
Player
Team
Rec
Att
Yds
Avg
TD
Total
DYAR
Opp
1.
Emmanuel Sanders DEN
6
8
83
13.8
0
34
CAR
Sanders caught all three of Peyton Manning's first downs in the second half, on gains of 25, 22, and 16 yards. Those were his only first downs of the game. He failed to convert either of his third-down opportunities.
2.
Corey Brown CAR
4
7
80
20.0
0
26
DEN
Brown had three first downs, including a 42-yarder in the third quarter. He failed to convert his only third-down target in the game.
3.
Devin Funchess CAR
2
5
40
20.0
0
12
DEN
Funchess' two receptions: a 24-yard gain on second-and-12, and a 16-yard gain on first-and-10.
4.
Owen Daniels DEN
1
2
18
18.0
0
2
CAR
Daniels' sole catch was an 18-yard gain on first-and-10.
5.
Ted Ginn CAR
4
10
74
18.5
0
-4
DEN
Ginn's totals include 7 DYAR receiving, -11 DYAR passing for his sack on a failed throwback to Cam Newton. His 45-yard catch in the third quarter was the longest play from scrimmage for either team.


Worst Wide Receiver or Tight End by DYAR
Rk
Player
Team
Rec
Att
Yds
Avg
TD
Total
DYAR
Opp
1.
Demaryius Thomas DEN
1
6
8
8.0
0
-28
CAR
Thomas' only catch was an 8-yard gain on first-and-10. He was the target on three third-down throws, all of which fell incomplete.

Posted by: Vincent Verhei on 08 Feb 2016

86 comments, Last at 11 Feb 2016, 8:49am by ChicagoRaider

Comments

1
by dmstorm22 :: Mon, 02/08/2016 - 2:27pm

This whole Manning season has been an unbelievable performance art. It's like he heard all the people say 'stats don't matter when you don't win' and decided to prove just how right / wrong (depending on your viewpoint) that is.

It's like he said 'fine, I'll just win and play like dogshit'

4
by JFP :: Mon, 02/08/2016 - 2:39pm

Peyton Manning: Husband, son, QB, pitchman, artist, Renaissance man.

20
by Bobman :: Mon, 02/08/2016 - 4:10pm

And if he drank all the Bud he claimed he would, he's also mighty hungover right about now, so add "giant cotton-mouthed headache" to the list. And with that fivehead of his, that's a lot of pain.

21
by Kevin from Philly :: Mon, 02/08/2016 - 4:12pm

Unpaid (apparently) Budweiser shill.

35
by CincySaint :: Mon, 02/08/2016 - 4:34pm

Unpaid (directly) = yes. But he has a financial interest in two Anheuser Busch distributors so there's a reason he said "Bud".

86
by ChicagoRaider :: Thu, 02/11/2016 - 8:49am

And that assumes that he continues to go unrewarded for that pitch. Anyone want to take a bet that he does a Budweiser ad soon? Maybe it was a tryout.

65
by Scott C :: Tue, 02/09/2016 - 12:44am

I don't like Denver. I'm not much of a Peyton Manning fan. I was rooting for Carolina, slightly.

But watching Manning win that way was strangely satisfying as a form of poetic proof that QB quality is not the sole reason for success, despite all the BS about wins, rings, etc being the primary measure of QB greatness.

83
by Dennis :: Tue, 02/09/2016 - 9:47pm

I agree. It's great to see the same people who criticized Manning for not winning more SBs now refusing to give him credit for winning the Super Bowl.

2
by ammek :: Mon, 02/08/2016 - 2:37pm

I'm looking forward to, one day, seeing where Tony Eason ranks in these tables.

Who would be the offensive MVP of superbowl 50? I'd have to plump for CJ Anderson and his <20% success rate. The FO preview which saw similarities with the Giants-Ravens superbowl was accurate, except that Peyton Manning was less productive than Trent Dilfer. (A sentence that has seldom been typed.)

23
by Kevin from Philly :: Mon, 02/08/2016 - 4:13pm

Manning's forehead pales in comparison to TD.

24
by RickD :: Mon, 02/08/2016 - 4:17pm

Eason was yanked early so there will be a question of how seriously his participation counts. Usually Eason is viewed as the worst of all starting Super Bowl QBs, ever.

I love that the game plan was "the Dolphins beat the Bears with short passes, so that's what we'll try to do, even though we're a run-first team." Because, you know, QBs are fungible.

48
by SandyRiver :: Mon, 02/08/2016 - 5:24pm

Zero for 6 plus 3 sacks, one a strip. Small sample but - yecchhh!

56
by Emptyeye :: Mon, 02/08/2016 - 5:55pm

Especially when "The Dolphins" were quarterbacked by an all-time great whose skillness was uniquely suited to attacking the relative non-strengths (Did the '85 Bears even have "weaknesses" as such?) of the Bears, so "Hey Tony Eason and Steve Grogan! You're gonna do the same thing that Marino guy did! Easy, right?"

85
by RickD :: Wed, 02/10/2016 - 5:09pm

Yes, exactly. On the one hand you have John Hannah, on the short lists of "best offensive linemen ever". Admittedly at the tail end of his career, but still bringing it. On the other hand you have Tony Eason.

Who do you design your offense around?

Now it's likely that committing everything to the rush would not have worked. The Bears' just had an all-time great defense there. But at least the Pats would have lost doing what they do best, instead of trying to pretend that Tony Eason could do anything that Dan Marino can do.

71
by mehllageman56 :: Tue, 02/09/2016 - 4:25am

Eason had the worst day of any Super Bowl starting quarterback, but I believe David Woodley was much worse than him. See, I still remember the Mud Bowl.

66
by Grendel13G :: Tue, 02/09/2016 - 1:02am

Fun question about the offensive MVP. It would have to be CJ Anderson. His success rate is abysmal, but I would chalk up a lot of that to context (e.g., the o-line being a caricature of its bad self all game, the offensive coaches stupefyingly happy to chew clock, etc). When he actually made it past the wall of bodies on the line, he had some good runs, including some nice broken tackles.

Really the only other offensive player on the Broncos who had a good game was Emmanuel Sanders. And no offensive player on Carolina even came close to the level of play required to win MVP in a losing effort (which seems like such an artifact of a bygone era anyway).

I was actually afraid they might give Anderson the MVP instead of Von Miller (because, you know, offense), which would have been a crime. Luckily, sanity prevailed and the right guy won it.

3
by ChrisS :: Mon, 02/08/2016 - 2:37pm

I am probably being an overly pedantic dick but statements like "Manning is the worst quarterback to win a Super Bowl in at least a quarter-century" rub me the wrong way. It should say "Manning (one of the greatest QB's of all time) played the worst game by a quarterback to win a Super Bowl in at least a quarter-century".

10
by t.d. :: Mon, 02/08/2016 - 2:44pm

These are the guys that created an 'irrational thread' to justify why a less credentialed quarterback was 'on par' with the best guy in the league by their own numbers. Don't expect objectivity on this topic

25
by Bobman :: Mon, 02/08/2016 - 4:18pm

True, but I get what ChrisS is saying in a grammatical context. The Oscar (r) isn't for the best actor per se, but the best performance by an actor in a leading role.

Manning isn't the worst SB QB by a long shot, but his performance yesterday was. A nit-picky point, but one an English teacher would certainly emphasize in grading an essay. If you want objective discussions, this is still probably the best place to go, despite some glitches.

There is plenty of nit-picking on this site, so I am okay with his angst.

And the irrational thread was created to divert the Peytom Branning snowball fights away from the core discussions. I think we actually crashed (or seriously bogged down) their server once. (man... that was fun.) The irrational thread was a survival tactic.

26
by RickD :: Mon, 02/08/2016 - 4:18pm

That's really not an accurate history. The word "tendentious" comes to mind.

17
by anotherpatsfan :: Mon, 02/08/2016 - 3:19pm

Being overly pedantic has never stopped anyone here before. In any event, your corrected sentence (without the parenthetical gloss) does seem to more accurately state the thought that was apparently intended, given the context of the paragraph. The parenthetical is, although true, unnecessary (and borderline redundant when talking about PM). I really don't think that "worst ever" statement was part of any vast anti-Manning conspiracy here (particularly as FO employs Manning's biggest fan, particularly from a GOAT analytics standpoint).

Manning was (at least in part by design) Dilfer-esque yesterday. That's part of the narrative, but it doesn't detract from PM's career greatness.

Could be they were trying to get this (free) info out quickly for our viewing pleasure, so I choose to give them a pass for the clumsy phrasing.

28
by Bobman :: Mon, 02/08/2016 - 4:22pm

Adding to the "design" issues you reference above, CJ Anderson will get unfairly painted as a flop when he was basically asked to dive into the DL three times in a row repeatedly when 100 million viewers at home, and the Panthers' D, knew exactly what was coming. No style points there, coach, but as the Broncos discovered two years ago, style points don't equal rings.

Kubiak just wanted to kill the last 20 minutes of clock and let his D handle the heavy lifting. Worked just fine, but may have resulted in stats that do not accurately reflect the potential of the O players. It also, not coincidentally, prevented any Manning flutterballs from being picked, so I am sure he is happy with that as well. As am I.

5
by jklps :: Mon, 02/08/2016 - 2:40pm

But but but "QB WINS!!!!!!!!!!"

6
by Travis :: Mon, 02/08/2016 - 2:40pm

Not that it matters, but Whitaker had 8 extra receiving yards after a lateral on the Panthers' final drive.

9
by rj1 :: Mon, 02/08/2016 - 2:43pm

I always wondered how advanced statistics account for plays like that where one player catches and then successfully laterals.

82
by panthersnbraves :: Tue, 02/09/2016 - 8:56pm

but he should have gone OOB instead of killing the clock for 3 yards...

7
by BroncFan07 :: Mon, 02/08/2016 - 2:41pm

Heh, that's funny. On a day where Ronnie Hillman literally did zero, he couldn't even pick up another Worst RB award.

46
by Aaron Brooks Go... :: Mon, 02/08/2016 - 5:07pm

It's clear that DYAR doesn't account for pass blocking, because Whitaker was a cancer whenever Carolina needed to pass.

Conversely, I thought Anderson had a really good game. He consistent broke tackles to get an extra yard or two -- that 23-90 would have been 23-40 if Hillman had those carries.

8
by pm :: Mon, 02/08/2016 - 2:43pm

Why is Daniels listed at #5, Ginn at #4 if Daniels is actually #4 on the table?

11
by dmstorm22 :: Mon, 02/08/2016 - 2:47pm

Surprised that Stewart's day comes out as far better than Anderson

27
by RickD :: Mon, 02/08/2016 - 4:22pm

Stewart did get a TD after all.

Anderson had more runs that were impressive, but also a huge number of stuffs. Stewart's problems were in the passing game. Panthers kept giving him the ball and he would be immediately tackled. I think they needed a more creative game plan. I know that's an easy criticism to make and doesn't really give enough credit to a defense that was doing everything well.

12
by hscer :: Mon, 02/08/2016 - 2:57pm

Yup, third downs were awful for Manning. The raw numbers tell us that. I'd be interested to see his DYAR broken down by down (pun not intended). On 11 first down plays he managed gains of 25, 18, 8, 7, 7, 7, and 6. Otherwise, yech. I wish he'd played better, and it sounds like he really did practice well leading up to the game, but the hype about that looks pretty manufactured now.

Something looks very wrong when rushers for 23-90 and 5-0 end up with just about the same DYAR, but I'm guessing Anderson had to have had the less bad DVOA, which still added up over the 18 extra carries?

15
by bobrulz :: Mon, 02/08/2016 - 3:10pm

I wish there was more talk about Kubiak's painfully conservative play calling.

I mean, there was 5 minutes left in the game, Denver up 6, and Kubiak...calls 3 run plays up the middle for like 3 yards. Even on 3rd & long. That at least partially explains the splits.

19
by hscer :: Mon, 02/08/2016 - 3:54pm

I hear you, but he did get 10 plays on third down (which is what I'm talking about), with an average of 7.7 yards to go, and could only convert one, with two turnovers (both in opposing territory) to boot. Not good no matter the second or third down play calling.

22
by clipper :: Mon, 02/08/2016 - 4:13pm

I also am interested in this line of thought. The first drive of the game, Manning came out throwing and they moved the ball well. Seems like when they got into the red zone, they specifically played for FGs. My thought was they should have tried to spread out the defence a little bit and force the Panthers other defensive backs (not Norman) to make plays. They did this on the 2 point conversion and it was a pretty easy pitch and catch. It could be that the Panthers had given up by then, but I wonder if Manning would have been able to rise to the occasion if they had to play a hurry up, pass happy game. Obviously much more potential for turn overs, but also some success? Kubiak's prevent offence was...offensive.

30
by RickD :: Mon, 02/08/2016 - 4:23pm

Peyton's only pick was in the 2nd half, right? After that his job was to not lose, and Kubiak didn't trust him at all.

I didn't have a problem with that playcalling. When your defense is playing as well as Denver's was yesterday, it's OK, perhaps even preferable, to trust them.

43
by tunesmith :: Mon, 02/08/2016 - 4:49pm

I could have sworn that at least one of Peyton's turnovers was after him calling an audible. So Kubiak might have had the better judgment there!

13
by DEW :: Mon, 02/08/2016 - 3:04pm

Anderson's position on this table reminds me again of how DYAR is a counting stat. Basically, he was the only RB on the day who did anything positive. Most of the time, he could at least be counted on for two yards forward instead of Hillman's zero or -1, and he actually contributed a few good runs (plus that 4th-and-1 conversion that got called back for the hold--but if he hadn't converted it, Carolina would have taken the play and Denver wouldn't have even gotten the field goal). Yet he ends up behind Stewart (who spent most of the day being hot, if injured, garbage), and Hillman (who was utterly awful), because all the unsuccessful runs he had keep adding up, whereas Stewart and Hillman got less use and so didn't add as many negative plays.

31
by Bobman :: Mon, 02/08/2016 - 4:26pm

Yup. Blame Kubiak for that, but it was smart football in the end. I don't imagine CJ is complaining today.

36
by deus01 :: Mon, 02/08/2016 - 4:35pm

I don't think it was smart football it just ultimately worked out. For most of the game it seemed like Carolina was just one big play away from taking the lead and Denver was too happy to just knock time off the clock and punt.

40
by Bobman :: Mon, 02/08/2016 - 4:41pm

I did not like the strategy at all, but in the end it worked out due to the awesome performance of the Denver D. Remember it was a nine point game for most of that time, so Carolina was one big play away (I agree) from making it really close, IMO, not taking the lead. My family had no idea why I was still antsy with two minutes left; it was because of that one big play threat that always seemed to be just around the corner, followed by an onside kick.

The downside of playing "actual offensive football" could well have been a fluttering INT that Manning seemed close to in the first half a few times. That might have been enough for Kubiak to make up his mind. So if we're left with Smart vs Lucky I think it was probably smart.

55
by clipper :: Mon, 02/08/2016 - 5:48pm

It very well could have been smart and, given the outcome, was obviously the right way to go. I just remember reading that the two areas the Panthers defence were not good, was defensive backs outside of Norman, and short yardage situations. The Broncos have not been good at short yardage success but they have played well when their second and third receivers have a clear mismatch - thinking of the Green Bay game. This may all be due to wanting Manning to have a good final game, but I have to think that his statistics were as bad as they were because Kubiak dialed things back and became predictable. I think the average football fan could have predicted most of the play calls in the second half for the Broncos - not a secret for success against an excellent defence.

57
by deus01 :: Mon, 02/08/2016 - 5:59pm

That seems to be have been the case for the entire Bronco's season. The offense has played much better when trailing then when they have a lead and they seem to be able to manufacture drives when they need them. It's when they have a small lead that they switch to ultra conservative play calling and become very ineffective.

It's really frustrating when even those of us watching on TV can predict what plays are going to be called.

14
by nat :: Mon, 02/08/2016 - 3:07pm

However, what's surprising is which unit gave Denver that advantage...
That should be no surprise to long-time FO followers.

DVOA and VOA discount fumble recoveries as "fumble luck". Given the equal number of interceptions and the nearly equal number of fumbles, and considering the Panther's much better success at moving the chains, we would have expected the straight up VOA Offense/Defense numbers to favor Carolina. But since Denver's offense is known (based on past DVOA) to be weaker, that brings the two defensive DVOAs to the same value. Comparing unit to unit, the defenses played equally well.

Meanwhile, Denver's special teams played well and Carolina's special teams were horrible. The Panther's offense added some unforced errors, but still outplayed the Broncos' offense, straight up (VOA) or considering the opposing defense (DVOA).

The Broncos' defense was the most important unit to the win, for sure. It had to play very well to match Carolina's defense. But in a unit-to-unit comparison, it's the special teams play that stands out.

32
by RickD :: Mon, 02/08/2016 - 4:26pm

Not all fumble-luck is a 50-50 proposition. My guess is that strip-sacks are far more likely to go to the defense, and those are the two fumbles that really turned the game for Denver.

A lot of the horror of the Panthers' special teams came in two plays: the doinked FG attempt and the "I'm not touching you!" punt coverage play.

34
by nat :: Mon, 02/08/2016 - 4:30pm

Not all fumble-luck is a 50-50 proposition.
True. But VOA takes the type of fumble into account. As I am pretty sure you already knew.

47
by Aaron Brooks Go... :: Mon, 02/08/2016 - 5:11pm

Remind me -- does it take field position in account? My recollection is that it does not.

50
by The Ninjalectual :: Mon, 02/08/2016 - 5:37pm

IIRC no it does not. But following this super bowl, this offseason might be a good time to persuade the staff to look into this to see if it's an area where DVOA could be improved!

52
by nat :: Mon, 02/08/2016 - 5:41pm

That's a good question. Let's see if Vince or Aaron are still looking in....

OH WISE MEN OF FO! WHAT FACTORS ARE CONSIDERED IN ASSIGNING VALUE TO A FUMBLE?

Such as:
Does actual field position (not just distance from line of scrimmage) have an effect?
Are fumbled interception returns assigned any VOA value at all?
Whatever else you think would help people understand VOA's handling of fumbles...

Answer in whatever detail you feel appropriate, if you have the time. Thanks.

37
by Bobman :: Mon, 02/08/2016 - 4:36pm

I am pretty sure fumble luck was broken down here some years ago--not sure if strip sacks were included (they are pretty rare) but I assume you are right on that. I DO recall mention that fumbled snaps are generally recovered by the offense, so strip sacks might well balance that out.

I still think the bad punt coverage play was because the gunners interfered with the PR (and were certainly inside the "halo") and so they backed off at the last second, changing their momentum, opening a hole for him. I was calling for a flag even before he started his great run. Panther teammates also saw 2-3 Panthers around a PR making what was either going to be a fair catch or a sound beating the moment he catches it, so I can understand them downshifting mentally. Also, it's a situation that gets a FC 95% of the time, and he didn't.

I wonder if it will result in more aggressive gunner play next year, at least in close late-game situations--"I can't risk a big return, so I will run to hit him no matter what. 15 yards is not fatal, but 50 would be. If he FCs, then I get penalized. If not, then he gets vaporized."

53
by LyleNM :: Mon, 02/08/2016 - 5:43pm

You can see on the replay that the gunner did in fact contact Norwood before he caught the ball. I was stunned that there was no flag thrown.

38
by Chappy :: Mon, 02/08/2016 - 4:38pm

Yes, but I wish they posted some historical context about the special teams gap. This year it seems like the difference between the best and worst special teams was about 12%. That means this game was about three times the difference!? I'm sure there is a lot of game-to-game variation, but how many other Super Bowls have had about a 40% swing in special teams? Desmond Howard for Green Bay comes to mind as a special teams showcase, but I'm just not sure how any Super Bowl compares.

16
by Eleutheria :: Mon, 02/08/2016 - 3:13pm

For wide receivers, You accidentally have Owen Daniels at #5 and Ted Ginn at #4, even though Daniels comes first in the chart and has more DYAR.

42
by Ryan D. :: Mon, 02/08/2016 - 4:49pm

I believe they are saying that Ginn was the 4th best receiver by DYAR (ignoring his negative passing DYAR).

18
by galerus :: Mon, 02/08/2016 - 3:40pm

First table states that Cam NEwton has -51 PASS DYAR and 22 RUN DYAR.
But two tables later another one shows -73 PASS DYAR and 22 RUN DYAR for him.

One of these PASS DYAR must not be true.

29
by Kevin from Philly :: Mon, 02/08/2016 - 4:23pm

I filled out a pool card based on the expected suckitude of both offenses: Total under 44.5 (check); Manning under 238 yards (check); Stewart under 70 yards rushing (check); Hillman under 40 yards rushing (double bonus check). Only thing I forgot to anticipate was the kicker's suckitude - total FG both teams over 4 (missed it by one doink!). Dang it.

33
by Bobman :: Mon, 02/08/2016 - 4:29pm

That's nothing! I understand Gary Kubiak won his office pool by predicting Denver would get 1 of 14 third down conversions.

67
by Grendel13G :: Tue, 02/09/2016 - 1:10am

Hilarious. What's even funnier is that Kubiak making that bet in an office pool almost seems more likely than a Super Bowl-winning team going 1 for 14 on 3rd downs.

39
by Mugsy :: Mon, 02/08/2016 - 4:39pm

I get a kick out of any performance that makes Dilfer look good, and makes Sanchez's post season numbers sparkle with efficiency! & people think you gotta have a franchise QB to win! Invest in your D-line and D-backs... and anything can happen!

41
by Mugsy :: Mon, 02/08/2016 - 4:43pm

OMG! It just dawned on me ! If I've learned anything from this ~ Perhaps Ryan Fitzpatrick can win a SB after all!

72
by mehllageman56 :: Tue, 02/09/2016 - 4:31am

He has to make the playoffs first. At least he won ten games this year. Baby steps.

44
by Vincent Verhei :: Mon, 02/08/2016 - 4:55pm

Who would be the offensive MVP of superbowl 50? I'd have to plump for CJ Anderson and his <20% success rate.

Emmanuel Sanders, obviously.

I am probably being an overly pedantic dick but statements like "Manning is the worst quarterback to win a Super Bowl in at least a quarter-century" rub me the wrong way. It should say "Manning (one of the greatest QB's of all time) played the worst game by a quarterback to win a Super Bowl in at least a quarter-century".

Take your pick of any of the following. They are all statistically true, if overly cumbersome to type, which caused me to use the paraphrased version:

* Out of all Super Bowl-winning quarterbacks in the last quarter-century, Peyton Manning on the 2015 Broncos had the worst regular season.

* Out of all Super Bowl-winning quarterbacks in the last quarter-century, Peyton Manning on the 2015 Broncos had the worst playoff run.

* Out of all Super Bowl-winning quarterbacks in the last quarter-century, Peyton Manning on the 2015 Broncos had the worst performance in the Super Bowl.

Not that it matters, but Whitaker had 8 extra receiving yards after a lateral on the Panthers' final drive.

I always wondered how advanced statistics account for plays like that where one player catches and then successfully laterals.

Stanford Band Plays always count as completions that are "tackled" at the point where the first lateral is made.

Why is Daniels listed at #5, Ginn at #4 if Daniels is actually #4 on the table?

Whoops. Ginn was fourth in receiving value only when the table was originally made, but fell to fifth with his sack. I will fix.

Yup, third downs were awful for Manning. The raw numbers tell us that. I'd be interested to see his DYAR broken down by down (pun not intended).

First down: 23 DYAR
Second down: minus-12 DYAR
Third down: -126 DYAR

Something looks very wrong when rushers for 23-90 and 5-0 end up with just about the same DYAR, but I'm guessing Anderson had to have had the less bad DVOA, which still added up over the 18 extra carries?

Somebody else touched on this, but Anderson had seven successful carries, and 16 failed carries. And the failed carries hurt more than the good runs helped.

To put another way: Anderson's 34-yard run was worth 16 DYAR. His six zero-yard runs, which obviously had no effect on his yardage total, were worth a total of minus-36 DYAR.

To answer the other part of your question: Anderson's DVOA was -22.4%. Hillman's was -101.4%.

I mean, there was 5 minutes left in the game, Denver up 6, and Kubiak...calls 3 run plays up the middle for like 3 yards. Even on 3rd & long. That at least partially explains the splits.

Denver had four third-down running plays. They came with 1, 7, 9, and 17 yards to go. None picked up first downs.

First table states that Cam NEwton has -51 PASS DYAR and 22 RUN DYAR.
But two tables later another one shows -73 PASS DYAR and 22 RUN DYAR for him.
One of these PASS DYAR must not be true.

That table in the essay was just rife with errors. They have been fixed.

60
by ChrisS :: Mon, 02/08/2016 - 6:45pm

I choose "Out of all Super Bowl-winning quarterbacks in the last quarter-century, Peyton Manning on the 2015 Broncos had the worst regular season, playoff run and super bowl performance". I love quick reads and your insight and responsiveness are great.

74
by rj1 :: Tue, 02/09/2016 - 8:57am

"I always wondered how advanced statistics account for plays like that where one player catches and then successfully laterals."

"Stanford Band Plays always count as completions that are "tackled" at the point where the first lateral is made."

Do you give any credit though for how much the guy runs after receiving the lateral? It's something I've wondered because it does not fit neatly into traditional statistical counting, and I play rugby union in the U.S.: that type of play is our bread and butter moving the ball. In rugby terms, in effect what American footballers 99.9% of the time choose to do is crash the ball into contact instead of utilitizing their teammates to run further with open space. Doing that play successfully should be a high DYAR play for the guy that throws the lateral (when it looked he could only gain 5 yards he turned into 15), but you'd still need to give credit to the guy that caught it in a "yards after catch" sense. The practical issue is you significantly increase your risk of a turnover which is more damning in American football than it is in rugby union. But this is in part why I hope the South Africans or someone else in the Commonwealth decides to play American football one day: you could see a new way to play the game.

80
by Vincent Verhei :: Tue, 02/09/2016 - 6:25pm

The problem is those plays only happen in end-of-the-game scenarios. They are very rare in general, and successful plays are even more rare than that. If anything, my educated guess (as someone who watches a lot of football) is that they end up losing yards more often than not. But they are so infrequent there is no way to study them.

45
by Never Surrender :: Mon, 02/08/2016 - 5:01pm

Is it really "extreme fortune" that Denver recovered five of seven fumbles in the game? If you assume that recoveries are 50/50, in games with seven fumbles total you'd expect one team to recover at least five of them about 45 percent of the time, no?

Wouldn't "extreme" be something more along the lines of, say, recovering all seven, or recovering six out of seven despite the fact that three of the fumbles featured only one Denver player anywhere near the ball and always a crowd of Panthers.

49
by The Ninjalectual :: Mon, 02/08/2016 - 5:35pm

This is a good and oft misunderstood point about statistics. Its like how you'll see winning lottery numbers like 1,2,5,13,22. That doesn't loom random, laypersons say! But in truth it's very likely to come in clusters of similar numbers. What people feel like random should look like (2, 14, 21, 35, 46) is a very unusually even distribution.

Still, one team had to make more recoveries than the other. Denver undeniably benefitted from good luck more often than Carolina, but I agree it's not the gigantic disparity some make it to be.

54
by deus01 :: Mon, 02/08/2016 - 5:46pm

Especially since the chance of a recovery isn't really 50-50 a higher probability on a few of them (i.e. the strip sacks) makes it much more likely that they would get 5 of 7.

58
by Eddo :: Mon, 02/08/2016 - 6:32pm

Actually, when Chase Stuart looked at fumble recoveries in 2012, QB sacks came out to be the closest to 50/50 of all.

Play type: defensive recovery rate
----------------------------------
Aborted snap: 24.4%
QB sack: 50.8%
QB run: 45.4%
QB run (negative): 33.3% (***)
Non-QB run: 62.1%
Reception: 60.0%
(Overall: 50.1%)

(***) From the article: "this could be on bad handoffs, simply dropping the ball, etc."

61
by Dan :: Mon, 02/08/2016 - 7:07pm

Note that each team had 2 strip sacks, and Denver recovered 3 of the 4.

51
by hscer :: Mon, 02/08/2016 - 5:40pm

Denver's luckiest recovery was by far the one after the Newton interception. Maybe Denver keeps it to 16-10, but 16-14 would have made them super nervous.

You could argue that that fumble should never have even happened, which would make for a hard time deciding which one of the other 4 Denver recoveries was particularly lucky.

62
by Cythammer :: Mon, 02/08/2016 - 7:40pm

Except that if Carolina had recovered that fumble, they would have been the one's getting absurdly lucky. Nothing says 'luck' like gaining big yardage to set up first and goal on a play that began with your QB throwing a pick. A turnover for both teams, but with Panthers being the ones who gain a big advantage overall if that had been what happened.

81
by Independent George :: Tue, 02/09/2016 - 6:52pm

Marlon McCree weeps manly tears at you.

59
by anotherpatsfan :: Mon, 02/08/2016 - 6:37pm

The chance of Denver doing it was about 22 percent (half the chance that one of the two teams would do it). While it is not very surprising that a team winds up with 5/7, the team that does can be considered lucky. But luck happens.

Agree it would be interesting to see some enhanced fumble luck research (thanks for the link Eddo).

77
by Arkaein :: Tue, 02/09/2016 - 4:31pm

I think "extreme fortune" in this case might mean extremely beneficial, not extremely unusual.

Using a simplified 50/50 model, Denver received +3 to their turnover differential essentially from luck. That's a big difference from even +1, much less -1 or -3, all of which would be as or more likely.

63
by Cythammer :: Mon, 02/08/2016 - 7:43pm

I think both QBs were significantly better than their stats, even if they were both still not good. But neither had any real help at all from their running game, not much from the receivers, and suffered from bad line play. For that matter the coaches were hamstringing their offenses too, with bad playcalling on the Carolina side, and an unimaginative, give-up approach for much of the second half for Denver.

64
by SirimarA :: Mon, 02/08/2016 - 8:33pm

"[A]ccording to DVOA, as great as Denver's defense was in Super Bowl 50, Carolina's defense was just as great."

This statement should raise eyebrows from the FO staff. You all watched the game. Do you REALLY think the Panther's defense was "great"? Confidence in one's statistics to cut through the inherently biased view of football we all tend to have is one thing, but overconfidence is another. I think this game can shine light on a particular flaw with (omission from?) weighted defensive DVOA.

Weighted DVOA is intended to assess how "good" a performance was, accounting for degree of difficulty. Weighted DVOA controls for degree of difficulty in an incomplete fashion, accounting only for the opposing team's offense. But the opposing team's offense is not the only factor which makes a defense's job more or less difficult. Poor offensive performance often "poisons" a good defensive performance. There is only so many times a defense can produce good results if their offense keeps going 3-and-out and the defense gets no rest. Weighted DVOA simply doesn't account for the fact that there are situations where almost any defense (even a great one) will fail. Without accounting for how difficult a particular defensive performance is, one cannot assess how "good" that performance was. Because DVOA does not account for increased difficulty that comes with a poor offense it misses just how "good" the Broncos performance was on Sunday.

70
by tunesmith :: Tue, 02/09/2016 - 4:25am

I'd love to see a defensive DVOA split of "while ahead" versus "while behind" or something. Or whatever would control for strategic clock killing. Denver's defense was against a team that was trying desperately to score. Carolina's defense was for the most part against an offense that was trying half-heartedly to kill the clock, happy to punt it away.

73
by Duff Soviet Union :: Tue, 02/09/2016 - 7:12am

"Weighted DVOA is intended to assess how "good" a performance was, accounting for degree of difficulty. Weighted DVOA controls for degree of difficulty in an incomplete fashion, accounting only for the opposing team's offense. But the opposing team's offense is not the only factor which makes a defense's job more or less difficult. Poor offensive performance often "poisons" a good defensive performance. There is only so many times a defense can produce good results if their offense keeps going 3-and-out and the defense gets no rest."

Is there any evidence for this at all? You could just as easily say that a team having a good offense causes their defense to "slack off" knowing that they can afford it now and then, while a great defense with a bad offense will never do that.

And yes, I thought Carolina's defense was great, even considering that Denver's offense sucks.

75
by SirimarA :: Tue, 02/09/2016 - 12:02pm

There is absolutely evidence:

http://www.fieldgulls.com/seahawks-analysis/2015/1/29/7947705/rhythm-fat...

While it may be that what you are saying has some truth, and there is some negative psychological affect from good offensive performance and positive psychological motivation stemming from poor offensive performance, I highly doubt that psychological affect is enough to overcome the physical challenges of being put out on the field repeatedly without adequate rest. Do you have any evidence for your claim?

84
by Dennis :: Tue, 02/09/2016 - 9:53pm

Without even looking at DVOA or any other advanced metrics, I thought Carolina's defense pretty clearly outplayed Denvers. The only Bronco's offensive TD came on a four yard drive, and they needed a third-down penalty to even get that. Carolina outgained them by quite a bit. The difference was Denver got the big turnovers.

68
by Grendel13G :: Tue, 02/09/2016 - 1:17am

I love that Manning out-Dilfered Dilfer! What a way to end a HOF career.

(PS, Peyton, please do hang up the cleats. I think you're the GOAT, but this season... yeesh...)

69
by Paul R :: Tue, 02/09/2016 - 2:02am

Peyton Manning--Plays poorly, wins Super Bowl, retires with honor.
Tom Brady--Plays like a sissy, gets testicles rubbed in his face, booed at same Super Bowl.

Debate over.

76
by garion333 :: Tue, 02/09/2016 - 12:58pm

Under Cam's breakdown: "In nine targets, Olsen had four receptions for 41 yards and only one first down. His six rushes included first downs on second-and-11, second-and-10, and third-and-2."

For clarity sake it should probably say "Newton's six rushes" since it reads like Olsen had six rushes and that's just not true. ;)

78
by commissionerleaf :: Tue, 02/09/2016 - 4:54pm

I thought Cam Newton's press conference was perfectly fine except that the reporters asked stupid questions.

I say this as a person who was rooting for the Broncos and thinks Newton is overrated by the NFL fanbase based on QB wins.

I also think Cam Newton has the best physical tools of any quarterback in history, but that's completely different from how he has actually played to date, which is uneven.

79
by Will Allen :: Tue, 02/09/2016 - 5:18pm

Short of a player committing felony during a press conference, I really don't care what they do. On the other hand, if a player were to lift Roger Goodell's wallet during the trophy presentation, I'd think that was pretty entertaining.