Writers of Pro Football Prospectus 2008

Most Recent FO Features

SimBla15.jpg

» OFI: Don't Make Saban Angry

Notre Dame and Baylor entered the one-loss group in what is shaping up to be an extremely tight race for playoff consideration.

25 Sep 2012

Week 3 DVOA Ratings

by Aaron Schatz

A huge 27-3 victory over the previously undefeated Chargers helps put Atlanta in the NFL driver's seat according to this week's Football Outsiders DVOA ratings. The Falcons are number one overall, and with a big lead over the rest of their division, they are also our definitive Super Bowl favorites according to the Football Outsiders playoff odds. They won the Super Bowl in 24.5 percent of this week's simulations; no other NFC team won the Super Bowl in more than 7.0 percent of simulations.

For the most part, the early DVOA ratings agree with conventional wisdom. The differences for the most part come from the current lack of opponent adjustments (which is why these are technically VOA ratings, not DVOA ratings). That changes next week, when we begin to install the opponent adjustments. Once that happens, a team like Buffalo (currently eighth) will drop a bit, and a team like Green Bay (currently seventh) will move up a bit. Still, the top ten makes good sense. The undefeated teams are first (Atlanta) and third (Houston) with Baltimore in between. There are two 1-2 teams in the top ten, but they happen our two Super Bowl favorites from before the season, Green Bay and New England, and their losses have been very close (and, particularly in the case of Green Bay, not actually losses).

(Small aside: Reversing the results of REFPOCALYPSE would mean a small change in the ratings for Green Bay and Seattle. Going from a 24-yard touchdown to an incomplete pass would change the Packers defense and Seahawks offense by about 3.5% worth of DVOA. That seems like a lot for one play, but a touchdown on fourth-and-10 is a pretty big play. If DVOA was strictly based on a win-probability model, the difference would be even larger. The difference would also be larger if not for the fact that a Hail Mary interception does not get the usual interception penalty in the DVOA system. As far as the rankings go, this change would move Green Bay to fifth in overall DVOA, but Seattle would remain 12th..)

Another interesting number belongs to Houston: 32. That's is where Houston ranks in special teams, and it's the reason they don't rank number one right now. Shayne Graham only has seven touchbacks in 18 kickoffs, which would have been great in 2010 but isn't very good with the kickoff at the 35. Trindon Holliday has been caught behind the 20 on three of his five kick returns, and Miami's Marcus Thigpen returned a punt for a touchdown on them.

The one big difference between our ratings and conventional wisdom would be Arizona, ranked 17th, but the Cardinals did move up big this week with their victory over Philadelphia. Their offense is still ranked only 28th, although it's interesting to look at the individual stats. Your eyes do not deceive you: Kevin Kolb is currently second in the league in VOA. That can't last, can it? The important thing to understand is that, despite their improved defense, the Cardinals are riding a somewhat ridiculous run of close victories. Our old pal Bill Barnwell does a good job of looking at the bottom of this piece.

One last team to talk about, and that's Pittsburgh. We all know that New Orleans is having heaps of problems, but I'm not sure people realize just how much Pittsburgh has been struggling. The Steelers are 28th in total VOA through three games, which puts them nine spots below the Saints. They can't really use strength of schedule as an excuse; they just lost to an Oakland team that looked horrendous in the first two weeks. It makes sense that the Steelers are having defensive problems with Troy Polamalu injured; historically, there's always been a big difference between the Steelers defense with Polamalu and without him, as Vince Verhei details in this ESPN Insider piece. Still, even without Polamalu, I don't think anyone would expect Pittsburgh to rank 29th on defense. The offense also has problems, and ranks just 17th. The good news there is that Pittsburgh's offense has been a lot better the last two weeks. The Steelers had -31.5% offensive DVOA against Denver in Week 1, but they've got 15.3% offensive DVOA in their last two games.

* * * * *

All stats pages should now be updated. The "Defense vs. Types of Receivers" numbers are now updated for the first time in 2012. The FO Premium splits database should be updated by later tonight with current 2012 data, and we also finally have "Weekly as of Week X of any Season" numbers updated for the newer version of DVOA, so that should be updated soon too.

Make sure to also check out our brand new SNAP COUNTS page! The NFL is finally making snap counts publicly available, and we're counting them up for you and posting them free. This week, we guys at Juice Analytics have worked on improving the search functionality to make it easier to find specific players. Enter a (last) name in the search box, and it will offer you a list of players to choose from. If you still have trouble finding specific players, try filtering by position or team. We apologize that the page still has "stacked" tables with separate graphics for Week 1, Week 2, and Week 3, but we're working on one big table that will incorporate every week. In addition, we should have a table that gives season totals up by the end of the week. We know some people have had some problems with the new snap-count graphics; please feel free to share your criticisms and suggestions with us at mailbag-at-footballoutsiders.com.

* * * * *

These are the Football Outsiders team efficiency ratings through three weeks of 2012, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)

OFFENSE and DEFENSE VOA are adjusted to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS VOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.

There are no opponent adjustments in VOA until the fourth week of the season, which is why it is listed as VOA right now rather than DVOA. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.

DAVE is a formula which combines our preseason projection with current VOA to get a more accurate forecast of how a team will play the rest of the season. Right now, the preseason projection makes up 55 percent of DAVE. Note that the projection portion of DAVE for the New York Jets has been adjusted to reflect the loss of Darrelle Revis.

To save people some time, please use the following format for all complaints:

<team> is clearly ranked <too high/too low> because <reason unrelated to DVOA>. <subjective ranking system> is way better than this. <unrelated team-supporting or -denigrating comment, preferably with poor spelling and/or chat-acceptable spelling>

TEAM TOTAL
VOA
LAST
WEEK
TOTAL
DAVE
RANK W-L OFFENSE
VOA
OFF.
RANK
DEFENSE
VOA
DEF.
RANK
S.T.
VOA
S.T.
RANK
1 ATL 45.3% 6 32.3% 1 3-0 22.2% 3 -16.4% 6 6.7% 6
2 BAL 39.5% 5 23.5% 3 2-1 31.4% 1 -0.5% 18 7.6% 5
3 HOU 36.6% 1 19.0% 5 3-0 17.2% 6 -34.3% 1 -14.8% 32
4 SF 23.7% 3 6.8% 10 2-1 18.0% 5 -3.0% 16 2.6% 13
5 CHI 20.2% 14 17.2% 6 2-1 -19.0% 27 -29.5% 3 9.8% 2
6 MIN 18.1% 4 -1.5% 17 2-1 8.8% 11 -7.3% 13 2.0% 15
7 GB 16.7% 7 20.7% 4 1-2 -1.3% 18 -11.6% 8 6.5% 7
8 BUF 16.6% 19 8.9% 9 2-1 7.7% 12 -6.1% 14 2.8% 11
9 NE 15.9% 2 28.0% 2 1-2 22.4% 2 4.7% 19 -1.7% 20
10 NYG 15.6% 22 11.5% 7 2-1 17.2% 7 6.5% 21 5.0% 9
11 DEN 14.4% 9 9.7% 8 1-2 4.3% 14 -8.3% 11 1.7% 16
12 SEA 11.5% 13 -3.1% 19 2-1 -15.5% 25 -19.1% 4 7.9% 4
13 MIA 5.6% 16 -3.8% 20 1-2 -7.6% 20 -11.0% 10 2.2% 14
14 NYJ 4.5% 8 3.8% 12 2-1 -7.3% 19 -2.6% 17 9.1% 3
15 DET 4.2% 17 -0.6% 15 1-2 21.4% 4 10.6% 23 -6.6% 27
16 PHI 0.1% 11 2.7% 13 2-1 -23.2% 31 -34.1% 2 -10.9% 29
TEAM TOTAL
VOA
LAST
WEEK
TOTAL
DAVE
RANK W-L OFFENSE
VOA
OFF.
RANK
DEFENSE
VOA
DEF.
RANK
S.T.
VOA
S.T.
RANK
17 ARI -0.4% 24 -11.1% 24 3-0 -19.5% 28 -17.7% 5 1.5% 18
18 DAL -6.3% 20 -0.8% 16 2-1 -15.6% 26 -15.5% 7 -6.3% 26
19 NO -8.3% 23 4.6% 11 0-3 5.9% 13 16.9% 27 2.7% 12
20 WAS -11.2% 25 -7.4% 21 1-2 13.6% 8 13.5% 25 -11.2% 30
21 IND -12.0% 26 -12.8% 27 1-2 1.1% 16 9.6% 22 -3.5% 22
22 SD -13.3% 10 -8.1% 22 2-1 -10.9% 21 5.7% 20 3.4% 10
23 CAR -13.4% 12 -1.7% 18 1-2 3.5% 15 13.1% 24 -3.8% 23
24 TB -14.8% 15 -9.5% 23 1-2 -21.3% 29 -11.1% 9 -4.5% 24
25 OAK -16.3% 28 -11.2% 25 1-2 9.9% 10 14.8% 26 -11.4% 31
26 STL -18.1% 21 -34.0% 32 1-2 -27.3% 32 -7.6% 12 1.5% 17
27 CIN -22.1% 29 -11.5% 26 2-1 11.1% 9 38.3% 32 5.1% 8
28 PIT -22.5% 27 1.2% 14 1-2 -0.9% 17 20.4% 29 -1.2% 19
29 CLE -24.3% 18 -23.4% 31 0-3 -23.0% 30 -3.5% 15 -4.8% 25
30 TEN -29.5% 30 -14.0% 28 1-2 -12.9% 23 27.4% 31 10.8% 1
31 JAC -34.0% 31 -22.1% 30 1-2 -12.7% 22 18.0% 28 -3.2% 21
32 KC -47.6% 32 -19.5% 29 1-2 -14.7% 24 22.8% 30 -10.1% 28

Posted by: Aaron Schatz on 25 Sep 2012

111 comments, Last at 30 Sep 2012, 10:13pm by Web Agnew

Comments

1
by JIPanick :: Tue, 09/25/2012 - 5:52pm

"...this change would move Green Bay to seventh in overall DVOA..."

But Green Bay IS seventh in overall DVOA as is showing right now...

3
by Aaron Schatz :: Tue, 09/25/2012 - 6:04pm

Whoops. Should be fifth. I'm fixing that now.

2
by RickD :: Tue, 09/25/2012 - 5:56pm

I wonder how much of the collapse of the NE defense is referee-related.

Sorry! You can return to analyzing statistics while pretending that the refs aren't distorting the numbers.

4
by Paddy Pat :: Tue, 09/25/2012 - 6:05pm

WTF? NE went from third defense to... 19th, on the strength of one game? Yeah, they had a bunch of bad plays against Baltimore, but they also made a bunch of good plays, from passes defenced, to an interception, to a key fourth-down stop in the fourth quarter. I thought their performance against a good offense would be around average, maybe 5% in the positive to go with two strong performances in previous games. What gives?

5
by Jovins :: Tue, 09/25/2012 - 6:08pm

I'm too lazy to check, but that one game could make up far more than 1/3 of their defensive plays.

25
by Mr Shush :: Tue, 09/25/2012 - 7:31pm

VOA still, remember, not DVOA - for all the numbers know they could have been playing Jacksonville.

29
by jonnyblazin :: Tue, 09/25/2012 - 8:15pm

Baltimore averaged 7.7 yards per play not including penalties, which is utterly horrendous from a defensive perspective.

38
by MJK :: Tue, 09/25/2012 - 10:49pm

Remember, because DVOA (or, rather VOA) is a per PLAY averaging stat (not per game), it can be more heavily swayed by a good offensive or a bad defensive performance in a single game.

If you have a good day on defense, you'll have only a few defensive plays, and they will look good. If you then have a bad day, you'll have a lot, and a good number of them (at least a third of them) will look bad. Hence, the average of the two days will not be "average" but "bad".

(It should be noted, that if you have a somewhat bad day on defense, but your good defensive plays (especially the drive stopping ones) keep getting wiped out because the refs penalized you for holding a player with your mind, then that game will hurt you EXTRA badly).

For an extreme example, let's say that for game 1 and 2, a team plays lights out defense, and gives up two yards per play on every play. They will face, say, ten posessions per game, so they have a total of 60 awesome plays from those two games. Our eyes see a defense that through two games nevers surrenders a first down and gets two shutouts.

Then they play a third game when they give up 5 yards per play every play. That's pretty bad. Assuming drives start at the 20, they have to play 16 plays per possession. Even if the other team only gets 8 possessions (because they're taking longer) in the game, that's still 128 bad plays.

So even though they played well twice and bad once (which we subjectively think would be a indicative of a pretty good defense) VOA sees 60 good plays and 128 bad ones. In other words, a solidly below average team. Going forward, VOA doesn't think a team has a 2/3 chance of playing well on defense...it thinks the team has less than a 1/3 chance of playing well.

While DVOA and VOA certainly are useful stats, they're not the end-all and be-all. Per-drive stats have their place, too. Note that on a per drive basis, the above example sees 20 excellent drives (defensively) and 8 bad ones, which fits in more with our subjective perception of a team's skill.

11
by Karl Cuba :: Tue, 09/25/2012 - 6:16pm

What do you want them to do, make numbers up? Add their own subjective biases? Ask a magic 8-ball? (just like the replacement refs)

13
by Paddy Pat :: Tue, 09/25/2012 - 6:21pm

No, I'm just surprised by the result and wonder how the numbers add up. I'm not protesting the model. My eyes did not perceive one of the most horrific defensive performances of the season on Sunday night. Bad? Yes. Entirely bad, probably not. I guess I was wrong.

26
by DragonPie (not verified) :: Tue, 09/25/2012 - 7:59pm

I think Karl was talking to Rick

35
by RickD :: Tue, 09/25/2012 - 10:32pm

Was he? If so, it had nothing to do with what I said.

I said I wonder how much of the drop was referee-related. I suspect a good deal of it was.

Perhaps Karl took this to mean a demand that VOA be adjusted to deal with my subjective assessments of bad refereeing. That would be an inference that has nothing to do with what I typed. VOA is what it is. I don't see VOA or DVOA as the end-all be-all that apparently some people do. But I am interested in what moves the number in a certain direction. And the NE defensive number was absolutely hammered this week. If one took the refereeing seriously, this major decline would be warranted.

43
by dribble d (not verified) :: Wed, 09/26/2012 - 2:07am

rick... they gave up 7 plus yards a play not including penalties. ray rice average like 6 yards a rush. read MJKs thing and the two together show you why the number is where it is.

not to mention, you're acting like every penalty was a wrong call. it was a poorly reffed game but they got more than half of the calls right.

44
by Paddy Pat :: Wed, 09/26/2012 - 3:22am

I think the Pats' defense played poorly in a lot of ways against Baltimore, but I think the Baltimore offensive line consistently got away with egregious holding. That hurt the NE pass rush and the NE run defense. Baltimore also got help through phantom illegal contact calls and got away with a number of push offs, including some that were away from the play. Imagine that two Blatimore scoring drives might have failed if the game had not featured scab refs. That might effect DVOA a lot. The impression that I had as a casual viewer of the game was that the Baltimore team was better prepared to stretch the rules and see what they could get away with than was NE. Both of the Harbaugh brothers appear to have a healthy sense of how to cheese the scabs. What can one say? NE is certainly not above its own cheating innuendo. All hail to the best cheater?

59
by dcaslin :: Wed, 09/26/2012 - 9:49am

What can we say? Baltimore learned a lot getting screwed by the refs in the loss the Philly. You could practically cut out what Ravens fans said about their loss to Phi and paste it a NE fan's mouth after Week 3.

62
by Paddy Pat :: Wed, 09/26/2012 - 10:08am

No complaints really. I hold Baltimore and the brothers in high regard. It's just interesting to see them on a steeper learning curve than the Pats.

68
by RickD :: Wed, 09/26/2012 - 11:38am

Question for you dribble d:
were the penalties assessed on the Patriots on plays when the Ravens got large yardage or were they assessed on plays when the Ravens got little or no yardage?

Do you understand now how extra penalties might skew a statistic like "yards per play"? If a penalty removes an incomplete pass from the game log, that's going to increase the yards/play.

"they got more than half of the calls right."

When I grade a math test, a 51% is an 'F'.

And no, I'm not "acting like every penalty was a wrong call." Hyperbole doesn't help you.

61
by Karl Cuba :: Wed, 09/26/2012 - 10:06am

"Sorry! You can return to analyzing statistics while pretending that the refs aren't distorting the numbers"

Then what did you mean by that? I don't think it was absurd of me to interpret that as you requesting that they alter their system, which would be daft because as you say, VOA is what it is and you know that when you turn up here. So what are you complaining about then?

71
by RickD :: Wed, 09/26/2012 - 11:47am

Karl, do you work in numerical analysis professionally?

If you are dealing with data that seems fishy, what you do is say that the data is fishy, and that people who are interpreting the data should do so with a grain of salt. That kind of caveat is hardly uncommon in real-world statistical analysis. You don't simply bury your head in the sand and say "well, those are the numbers and we cannot do anything about it."

On the other hand, you don't constantly re-invent your statistical tools from data set to data set.

I'm sorry I have to explain all of this, but it really seems pretty obvious to me. I'm certainly not saying that they should change their statistical tools. You'll know when I'm doing such a thing, because I always use precise language when discussing mathematics. But it's absolutely standard to say "well, these are the numbers we have, they might have been distored for reasons X and/or Y, and let's keep an eye on things."

And the "Sorry!" was because we'd just spent a day and a half on other threads beating the refs to death.

I am tired of people "inferring" meanings from what I say that diverge dramatically from what I actually say. But this is not a problem one has to deal with when interacting with math-types. A mathematician will, rather than making an assumption about what he thinks I am saying, will actually ask for a clarification. It's a far more collegial way to go about matters.

74
by Brendan Scolari :: Wed, 09/26/2012 - 11:58am

I see your point, but how do you make the leap from there to the idea that the Pats defensive collapse is due to the refs?

No penalties on defense aside from pass interference are factored into DVOA whatsoever. And if anything I'd say the refs have been more lenient on allowing DB's to contact receivers. If you're going to say "The Ravens were holding the whole time" I'd only say that you, as a Pats fan, are not in a great position to see that:

a) The Pats were probably holding quite often too.
b) Every othery team in the NFL is probably holding quite often too.

If you think the Pats defense is the outlier in terms of facing inordinate amounts of holding solely against them with the teams they played not doing the same thing in their other games too, then I'd say quite simply that you're biased and in all likelyhood incorrect.

77
by MJK :: Wed, 09/26/2012 - 12:24pm

For what it's worth, the Pats themselves seem to think the bad refereeing is hurting their defense. A number of them spoke to that in the post-game press conference (although most of the players took the high road).

I think the point isn't that their defense is hurt worse than anyone elses' by bad refs...but rather that their defense (and maybe everyone elses') is getting worse the worse the reffing gets.

I haven't looked to see if statistics bear this out, but have scores been creeping up over the first three weeks? With refs calling all kinds of phantom penalties in the defensive backfield and hardly ever calling offensive holding, that's going to hurt a lot of defenses. Subjectively, it seems like penalties that hurt the defense have been called disproportionately high (probably because these are easier calls for the replacements to make), and penalties that hurt the offense (other than false starts) have been called disproportionately low (probably because those are harder calls to make).

78
by dmstorm22 :: Wed, 09/26/2012 - 12:30pm

Scoring's actually gone down slightly each week. It was near 49 points per game in Week 1, and is now under 48 for the season.

I think the refs definitely were told to call more DHolding and IC this week after letting a lot go in Weeks 1-2. It didn't correlate with increased scoring, or even passing I believe.

This year has been odd. Scoring is up (though not where it was last year at this time), but the teams doing a lot of the scoring are strange, like Washington or Cincinnati, while the big three offenses from last year are all a bit down (New England's VOA is fine, but overall scoring rate is down).

81
by Paddy Pat :: Wed, 09/26/2012 - 2:38pm

You obviously didn't watch Sunday night's game, Brendan. Sure, there's holding on most offensive lines, but the Ravens frequently tackled the New England rushers and didn't get called. Other teams have suffered plenty from the scab trends too, but it's not universal.

Some teams are doing a much better job hitting other players and holding and not getting called for it. I think some coaches are doing a more active job encouraging players to take advantage of the scabs than others are. I have seen DBs and WRs punching each other in the middle of plays not called for anything, offensive linemen horsecollaring DEs and driving them into the turf not called, and DBs called for pass interference when they didn't even touch the receiver. Some teams are definitely getting hurt more than others.

87
by JimZipCode :: Wed, 09/26/2012 - 4:23pm

Marshall Yanda got called for his first holding penalties in TWO YEARS. Got called twice.

79
by Karl Cuba :: Wed, 09/26/2012 - 12:42pm

Do you really think there is any need to be condescending? You didn't say what you meant, you weren't being clear and it may be shocking to you but I haven't been hanging on every word you have written. And then you behave like a complete arse.

I'd love to say more but the dratted filter ate my post and there's only so much energy to spend banging heads with a troll.

94
by Purds :: Wed, 09/26/2012 - 6:25pm

RckD, I hate to wade in because like you many of my posts are seen as coming from a specific angle (although in my case, I admit that is probably true!). But, reading what you and others have said in general, I think I get your point but I do have a question about the logic. I think the point is that poor referee calls and non-calls may have conspired to artificially deflate NE's defensive ratings, but here is what I wonder about. Shouldn't the rampant holds by offenses, etc., harm every defense's DAVE or VOA? And, thus, shouldn't the NE rank dropping to #19 be significant because while the ref issues may influence the raw DAVE or VOA, because everyone is playing with the skank refs, the should roughly be ranked accurately? I mean, I can see the bad ref actions hurting the defensive VOA and DAVE for each individual team's defense, but as each team suffers from the same refs, shouldn't the relative rankings still be accurate, or as accurate as VOA and DAVE can get?

6
by Paddy Pat :: Tue, 09/25/2012 - 6:08pm

Does DVOA really think that NO significantly outplayed KC? That's strange and interesting too. I thought KC had performed reasonably this week in the second half and in the fourth quarter.

36
by RickD :: Tue, 09/25/2012 - 10:33pm

Often a team that only shows up for a quarter of the game ends up looking bad according to VOA.

80
by NOLASaintsFan (not verified) :: Wed, 09/26/2012 - 1:02pm

That, and, EVERY referee judgement call in the game went against the Saints. From Pierre Thomas' on the field ruled completion and touchdown being overturned on very unclear replay video that MAY have indicated the bal touched the ground during the reception, to two other touchdowns being called back for reviews that were possibly somewhat questionable. Any one of those three touchdown reviews goes in the Saints' favor, and the game doesn't make overtime and the Saints win.

7
by tuluse :: Tue, 09/25/2012 - 6:09pm

Kellen Davis has played all but 5 snaps on offense so far. The only players to have played more than him are Cutler and 4 of the lineman.

Also, Hester's snaps have gone down each week, while Jeffery's have gone up.

97
by Duke :: Wed, 09/26/2012 - 7:58pm

Which is odd, since I don't recall seeing Davis do much of anything on offense so far.

99
by tuluse :: Wed, 09/26/2012 - 8:45pm

I assume he's been staying in to block mostly, I did find it unusual too he's been on the field for practically every snap.

102
by Dan :: Thu, 09/27/2012 - 3:13am

37 snaps pass blocking, 76 running routes (with 10 targets & 3 receptions), and 94 run blocking. Source: PFF.

8
by Paddy Pat :: Tue, 09/25/2012 - 6:11pm

Frankly, lots of these ratings seem odd. MIA and NYJ feel like they're being overrated here. And I'm sure that TB and DAL are both better than Detroit at this point. Detroit is sort of a one trick pony right now, and those other two teams are really scrappy... I guess adjustments will start to settle this in a few weeks.

32
by J. Oliver (not verified) :: Tue, 09/25/2012 - 9:05pm

I'm not so sure about DAL vs. DET, and I'm a Cowboys fan. Due to O-Line issues, their offense is pretty much hoping that Murray (runs) or Romo (passes) is able to make enough plays on their own to score. With their injuries in the secondary, I'm not sure how good the defense is right now either.

46
by JoeyHarringtonsPiano :: Wed, 09/26/2012 - 5:50am

Did you actually watch Tampa Bay's offense play last Sunday? Like others have mentioned before, a poor performance in roughly a third of your offensive plays (1 game out of 3) will make a big difference. And why would adjustments hurt Detroit all that much? Their worst performance came against San Francisco. And why would it help Tampa Bay? Their best performance came against a Carolina team that frankly looks pretty bad right now.

-I'm not Billy Bad-Ass.

63
by Paddy Pat :: Wed, 09/26/2012 - 10:11am

Good replies, both of y'all. I was probably unduly influenced by week 1, where the defenses of both teams significantly impressed me, and Detroit has seemed like a fragile team in all 3 weeks, as has Chicago for the matter of that, which appears somewhat putrid on the offensive line and rather lacking in defensive depth. I suppose I should go watch Dallas, Tampa, and Detroit's most recent games and see how they look.

67
by Eddo :: Wed, 09/26/2012 - 10:56am

I don't disagree the Bears high perch is precarious, but I don't really think defensive depth is a big issue.

At linebacker, I can see it. Losing Urlacher or Briggs would be huge. Although I do like Nick Roach, having both him and Geno Hayes get tons of snaps would be a bad thing.

The secondary has OK depth. Safety is - and always has been and will be - an issue, but Steltz at least has experience as a starter. Kelvin Hayden is a decent backup corner.

And the defensive line is incredibly deep right now. Possibly the deepest in the league.

75
by JoeyHarringtonsPiano :: Wed, 09/26/2012 - 12:04pm

My subjective gestalt is that being ranked 15-16 sounds about right for Detroit. In week 1, you could tell they were clearly the better team than the Rams, but Stafford single-handedly kept St. Louis in the game with his 3 first half interceptions. In week 2, it was clear that San Francisco was the better team, despite it ending up only a 1-possession game. In week 3, the Lions seemed poised to pull away from the Titans in the second half after scoring 18 unanswered points, but then all the 4th quarter craziness broke out.

I'm hoping the extra dimension that Leshoure brings helps the offense stay consistent, and some health in the secondary will keep the special-teamers out of the starting lineup and improve the pass defense.

-I'm not Billy Bad-Ass.

82
by Paddy Pat :: Wed, 09/26/2012 - 2:40pm

I was unaware of Leshoure's presence when commenting. Am watching the game now on Game Pass, and he's been impressive. We'll see.

9
by Anonymous (not verified) (not verified) :: Tue, 09/25/2012 - 6:11pm

Hey RaiderJoe, you got your wish. Raiders are rated higher than the Steelers. Just not in DAVE

10
by Will Allen :: Tue, 09/25/2012 - 6:15pm

Minnesota at Detroit will be an illuminating game this Sunday, especially for the Vikings pass defense. If the Vikings win while Detroit doesn't successfully toss the ball all over the field, and Ponder plays no worse than an average NFL starter, I'll defintitely say that my projection for the Vikings was way off. My guess, however, is that the Lions win about 27-20, while having a lot of passing success.

39
by LionInAZ :: Tue, 09/25/2012 - 11:40pm

Lions-Vikings matchups over the past decade or so have typically been close fights, even when one seems clearly inferior. I'd take the Lions this week mainly because it's in Detroit, but I'd expect it to be very close, barring injury to a major player like Allen or Megatron.

47
by JoeyHarringtonsPiano :: Wed, 09/26/2012 - 5:58am

The Vikings (especially Ponder) looked so good against the 49ers that I'm really worried about this game. Jared Allen always gives Jeff Backus fits, and he almost single-handedly won both games last year (but luckily for the Lions the Vikings offense disappeared in the 2nd half of the first game, and kept giving the ball away in the 1st half of the second game).

-I'm not Billy Bad-Ass.

12
by buccaneerryan (not verified) :: Tue, 09/25/2012 - 6:19pm

How did Denver's playoff odds jump by 53.5% after losing? Did the Chargers really play that badly?

15
by Thunderbolt of ... :: Tue, 09/25/2012 - 6:27pm

Just guessing, but Kansas City and Oakland really aren't playoff threats based on current DVOA, so the division is basically between Denver and San Diego. Denver didn't lose any games to San Diego in the standings, and although both lost, they did appear to play much better than the Chargers did. At this time 1/3 of the season (this week's game) indicates that Denver is much better than San Diego, so the playoff odds reflect that.

27
by Mr Shush :: Tue, 09/25/2012 - 8:06pm

And while opponent adjustments aren't yet a factor, subjectively allowing for that makes this seem even more reasonable. Denver made a real game of it in Atlanta, and then played a probably similarly good team in the Texans at home and again only lost by 6, while the Chargers got absolutely crucified by the Falcons on their own turf. I can easily see San Diego getting a wild card, but that's the Broncos' division to lose in spite of their significantly harder schedule.

37
by RickD :: Tue, 09/25/2012 - 10:38pm

"significantly harder schedule"

Two teams in the same division play only two different opponents. The within-division strength of schedule differences are mostly negligible. That is, until you start looking at things like "The Patriots get to play the Dolphins twice, while the Dolphins have to play the Patiots twice."

48
by Mr Shush :: Wed, 09/26/2012 - 7:46am

Texans and @Patriots vs. Titans and @Jets (without Revis)? That could very easily be a two game swing.

30
by Karl Cuba :: Tue, 09/25/2012 - 8:25pm

The rest of the media gets all excited by the NFC West and FO responds with a crush on the NFC North, which will get interesting as they play each other this year.

BTW, wasn't Bruce Irvin impressive last night? I thought his best moment was when he got under Bulaga's pads and bulled him right back for his first sack, the Pack were doubling him by the end of the first quarter. When you combine him and Clemons coming off the edge with Mebane and Bryant collapsing the pocket life could get pretty tough for quarterbacks.

83
by Jeff M. (not verified) :: Wed, 09/26/2012 - 2:44pm

I particularly enjoyed his hurdle of the snapper on GB's first FG, even though he couldn't get a second jump in time for the block.

14
by Paddy Pat :: Tue, 09/25/2012 - 6:26pm

Now I'm also confused that Arizona's playoff odds dropped after a strong game against Philly. What does a team have to do to get a little statistical respect?

16
by felden :: Tue, 09/25/2012 - 6:31pm

I imagine it's because the playoffs odds are based on DAVE, and San Francisco is shooting up there as the preseason projections get more and more phased out, faster than Arizona's improving.

19
by LyleNM (not verified) :: Tue, 09/25/2012 - 6:40pm

Not be in the same division with SF and Seattle.

17
by Eddo :: Tue, 09/25/2012 - 6:32pm

The Bears are clearly ranked too high because Jay Cutler is a whiny jackass whose teammates clearly hate him. Ranking teams by the thickness of their jersey numbers is way better than this.

18
by DisplacedPackerFan :: Tue, 09/25/2012 - 6:37pm

So I figured going into the year the NFC West was going to have some good defensive teams, what I didn't think would be true 3 weeks in was that SF would be the worst rated (I know things are going to change).

But I figured I'd look more closely at it.


4 . SEA -19.1 (-13.9 P, -29.2 R)
5 . ARI -17.7 (-11.8 P, -26.7 R)
12. STL -7.6 (-16.7 P, 3.4 R)
16. SF -3.0 (21.8 P, -39.0 R)

That's pretty impressive to have every team in the top half of the league.

Then I noticed that DET was the only NFC North team that wasn't top 16, sitting at 23rd.

Then I noticed that the NFC has teams 2-9, 12, 13, 16. That's 11 of the top 16 teams and NO in train wreck mode is the worst at 27th. That's kinda crazy. It's not like the NFC isn't playing offense, they have 8 of the top 16 teams, with the 3-5 for their top end offensive. I'm in my late 30's I started really paying attention to the NFL in the mid 80's so I got used to NFC dominance early on, but I had gotten used to the AFC being the better conference for much of the 00's. I'm not used to the NFC looking that much stronger.

Final little tidbit, I think someone else posted this on these boards in another thread mid last week, but it's still interesting.

NFC North is 3-3 vs NFC West
NFC West is 3-0 vs NFC East
NFC East is 4-0 vs NFC South

It's an interest "Beat Path" and a bit of scheduling quirk, that currently makes the South the crap NFC division. Atlanta has only played the AFC West, they haven't played an NFC team.

41
by Little Bobby Tables (not verified) :: Wed, 09/26/2012 - 12:31am

I don't think the NFC vs. AFC debate is that clear cut. Sure, the NFC is scary deep -- but the AFC has two of the top three teams in the league, as well as NE team that seems like it will end up in the top five.

20
by ptp (not verified) :: Tue, 09/25/2012 - 6:46pm

Small aside: Reversing the results of REFPOCALYPSE would mean a small change in the ratings for Green Bay and Seattle. Going from a 24-yard touchdown to an incomplete pass would change the Packers defense and Seahawks offense by about 3.5% worth of DVOA

I think the fact that we have one discrete play here with a questionable result (though not even the worst call of the night by far) makes for a tempting "what-if" scenario. Unfortunately this is a massive slippery slope. Change the PI on Chancellor and turn GB's 7 points into a field goal and Seattle's kicking a field goal at the 22 to finish the game, and what does THAT do to their DVOAs? Assume the drive never got *anywhere* because the refs called even a few of the tons of blatant offensive holds GB pulled on Clemons and Irvin that drive and what does THAT do their DVOAs? It's easier to make that one change in the model's results because there's no chance for anything to change after it, but it also seems highly specious as a practice or a temptation to give in to.

21
by Jovins :: Tue, 09/25/2012 - 6:51pm

What if the 49ers weren't allowed to challenge 2 of the 3 Gerhart fumbles? The game likely would have ended without the Vikings fumbling an extra 3 times - that's a huge switch.

40
by LionInAZ :: Tue, 09/25/2012 - 11:50pm

Definitely agree with this point. That bogus PI call against Chancellor turned a failed 3rd down conversion putside of FG range into 1st down within FG range. Everyone seems so focused on that final play that they forgot the terrible call that led gave the Packers the lead to begin with.

42
by evenchunkiermonkey :: Wed, 09/26/2012 - 12:43am

because youve stared me down the slippery slope...

after the packers kick that hypothetical field goal, what if the refs didnt flag Waldens legal hit as RTP and the packers get the game sealing pick? If that doesn't happen imagine the refs callOPI on Rice instead of DPI on Shields and the Seahawks have to nail a 60 yarder for the win.
Its a little like the pawn shop scene from Pulp Fiction. The gimp was out of the box and someone was going to get f#$%ed.

45
by KB (not verified) :: Wed, 09/26/2012 - 5:22am

Very good points. So now GB is up by 3 and McMillian Intercepts the ball and the packers run out the clock. Game over.

The game was just called horribly all around but I'm very surprised to see a fan point to some calls that went for GB. I felt much more went against them. I saw Matthews get held multiple times with no call. At one point he was tackled. Another time I saw Perry getting held for 5 seconds straight.

No matter what the worst call I may have ever seen was that PI called on Shields. He played perfect defense all while being held and thrown to the ground. How that gets called on him no1 may ever know. Maybe they meant to call offensive PI but somehow got it confused. For the Refs sake I hope so because that would be less embarrassing than somehow believing Shields actually interfered with Rice.

51
by Will Allen :: Wed, 09/26/2012 - 8:46am

What they likely saw was the receiver's jersey being stretched by Shields' hand, which, by itself, would have made for a tight, but acceptable DPI call. What was missed, of course, was the far greater contact intitiated first by the receiver. It's not surprising that inexperienced people fail to see the big picture.

I thoouht no call was the best call, followed by OPI. DPI should have been a nonfactor, of course.

52
by justanothersteve :: Wed, 09/26/2012 - 9:00am

There was a lot of uncalled holding on both sides. Bulaga looked better in the second half in part because he started doing the same holding the Seattle OTs were doing. I agree the PI call on Shields was horrible. There was also the PI call that went against Seattle. In slow motion, I thought the Seattle defender got there a fraction of a second early, but that is almost never called regardless of the level of official.

I'm still not happy how the game ended. All I can do is think of it as the Packers asterisk game. The Seahawks have the Testeverde "touchdown". The Bears have the Majkowski questionably legal forward pass. The Raiders have the Immaculate Reception (and the Tuck Rule). I don't know if this game will have a name for it. This morning I heard it called Goldengate. In any case, they need the regular refs back. I'm sick of the Replacements.

58
by Anonymous Coward (not verified) :: Wed, 09/26/2012 - 9:44am

From the mid 3rd Quarter to the last two drives the Packer offense was given 4 mulligans on failed third downs due to bad calls. I think it's perfectly reasonable to see this game as a 7-6 Seattle win if the regular officials are working

22
by Roch Bear :: Tue, 09/25/2012 - 7:07pm

I commented last week about how disorienting it was to find the Bears offense rated higher than their defense or special teams. The statistical gods have reeled them in and now the Bears are number 2 in teams, 3 in defense and 27 in offense. Now I think there are a bunch of underdamped statistical gods.

23
by Dennis Soh (not verified) :: Tue, 09/25/2012 - 7:17pm

Do the rankings consider strength of opposing team? Like for example, does Baltimore get a bonus on offense for playing the #2 defense (PHI)?

The #1 ranking for Baltimore's offense seems a little generous.

24
by DisplacedPackerFan :: Tue, 09/25/2012 - 7:28pm

Right now, no they don't. The current numbers are just what percentage better than a baseline created from I believe 2004 - 2007 data, a team performed. Starting next week, since there will be more data, the numbers will include opponent adjustments, so GB should see their offense look better because they played the #3, 4, and 16 defenses. So in your question, the Balt offense has played the #32, 19, and 2 defenses. Assuming they ran the same number of plays in each game, their offensive ranking will drop.

Though it won't be full effect adjustments until week 10 I believe ( I could look it up but I'm lazy right now).

28
by dmstorm22 :: Tue, 09/25/2012 - 8:12pm

So, right now the Texans are #4 in pass and rush defense, and #3 in pass offense and #7 in rush offense.

What other teams have there been to be top-10 in each category? Has there ever been a team to go Top-5 in each?

The Texans just seem so incredibly complete right now. They've won three games without Andre Johnson really a key factor in two of them. They have two good to great running backs. Their defense is playing incredibly well right now. They seem angry that they lost a great shot at a Super Bowl last season.

31
by Will Allen :: Tue, 09/25/2012 - 8:43pm

Hmmmmm........I do think Jerry Jones would be made miserable if the Texans were to win a Lombardi.......I believe I have another team to root for.

33
by Mr Shush :: Tue, 09/25/2012 - 9:10pm

The Texans look like a terrific team and a legitimate Super Bowl contender, and through three weeks JJ Watt would be my choice for DPOY and maybe even MVP, but that #4 run defense ranking is a mirage. Nose tackle and ILB2 are two of the four weakest spots in the starting lineup (or indeed regular rotation - the others are CB2 and RT) and a team that can stay close enough to the Texans to be able to run regularly - especially up the middle out of a pro set - would demonstrate that in a hurry. The Dolphins actually did, except that they put the ball on the ground twice in 17 carries, and were then so far behind that they ran twice in the entire second half. Don't get me wrong - we're not talking about the no-Bob-Sanders-2008 Colts - but the Texans are genuinely very good at passing, running, rushing the passer and covering, but only average at stopping the run. Oh, and pretty lousy on special teams.

34
by dmstorm22 :: Tue, 09/25/2012 - 10:19pm

I was just going by the VOA ratings. Just by watching them on a limited basis, I would say you are right.

85
by fracguru :: Wed, 09/26/2012 - 3:48pm

People keep commenting about the Texans run defense up the middle. I've heard for years how the Texans HAVE to get a big 350lb NT to stuff the middle. It seems so easy- run up the middle- and yet game after game the Texans don't seem to get punished for this weakness. Also, it's not as if NT and LB2 are the only guys that can make a tackle on a run up the middle. Against Miami the Texans started bringing the strong safety up to run support. They also have a pretty good player at ILB1 who makes plays.

I'm not saying you are wrong, but how long do we have to watch before we decide that being average against the run up the middle isn't a big deal?

100
by Mr Shush :: Wed, 09/26/2012 - 9:49pm

Oh, no argument here. If you're going to have a weakness, run defense is the one to have, and it's not even like they're terrible at it.

The Dolphins game was a bit flukey, though - Bush pretty much ran all over the Texans in the first half, but those turnovers before half time and the points off them took the run off the table in the second because the lead was so big. Forcing 2 fumbles on 17 carries is simply not repeatable.

The Texans can win the Super Bowl without a dominating nose tackle. They may be the best team in the league even without one. I'd rather have an upgrade at right tackle or corner, or a real difference-making safety, or a true long term replacement for Andre Johnson, or any number of other things. That still doesn't mean it wouldn't be a nice thing to have.

49
by BigAl (not verified) :: Wed, 09/26/2012 - 7:51am

Again the Cardinals are causing a unexplainable glitch in the DVOA matrix. Undefeated, with all wins over higher ranked teams, yet still sitting ranked comfortably outside the top 15. Schatz no longer carrying the mantle of "they can't be winnning according to my spreadsheet", but passing the mantle to a less convincing Bill Barnwell. What's most reassuring that this deja vue experiencing correlates directly with the Cardinals Super Bowl run a few years back. It would be interesting to compare that years seasonal trend to whats the stats are projecting this season for the Cardinals

50
by Mr Shush :: Wed, 09/26/2012 - 8:35am

Not even remotely similar. That year, DVOA (rightly) thought the Cardinals were a very good team up until the point where they stopped playing (because the division was locked up and they had no shot at a bye) after which it (rightly) noted that they were playing like crap and (wrongly but understandably) concluded that the truth was somewhere in between their early season excellence and their late season crappiness. This year, VOA may be under-rating the Cardinals so far due to the lack of opponent adjustments, but they have also been rather lucky and had some narrow wins. The reality of the Cardinals seems to be great defense, lousy offense, great special teams. They'll probably end up around .500. They might sneak a playoff spot. They're not going to the Superbowl.

60
by Kurt :: Wed, 09/26/2012 - 9:49am

Probably not, but last year a team rode a great defense, mediocre-to-bad offense and okay special teams to overtime of the NFCCG.

70
by Mr Shush :: Wed, 09/26/2012 - 11:41am

There's a world of difference between mediocre-to-bad and atrocious. Kolb/Skelton behind a terrible offensive line is the latter, even with Fitzgerald. And the 2012 Cardinals play in a much harder division than the 2011 49ers.

72
by MJK :: Wed, 09/26/2012 - 11:55am

I don't think this year's Cards are last year's Niners.

The Cards defense definitely looks good...but the Niner's defense last year was amazing.

On offense the disparity is greater. Even though the Niner's offense was middling last year, they had some legitimate playmakers--Vernon Davis and Frank Gore--and a QB who wasn't lighting things up but was almost perfectly mistake free. The Cards have, by my count, just one legitimate playmaker (Fitzgerald) and don't have a QB that is in Alex Smith's league when it comes to limiting mistakes.

And remember, as good as the their special teams are, they didn't make Gostkowski magically hook the ball. They won that game due to a random event totally out of their control. Even leaving their close win over the Seahawks aside, if they were 2-1 instead of 3-0, their ranking wouldn't look so funny.

In addition, see my comment above about a few bad defensive drives spoiling your VOA despite good overall defense. The same works for offense in reverse. Good offensive drives skew your VOA. So a team that is good because it has a good defense but a bad offense will likely be ranked lower by VOA then by conventional wisdom, W-L, or by per-drive stats.

89
by jpr (not verified) :: Wed, 09/26/2012 - 5:35pm

Actually, your Gostkowski argument doesn't help the rest of your argument. These per-play averaging stats. If you eliminate the fumble prior to the missed FG, the Cardinals have now won the game handily and without a distinctive lucky break, but you're not impacting VOA enough to move the Cardinals up the list significantly. Its also early, but I don't think the Niners' defense last year was significantly better than the Cardinals' has looked since the middle of last year (in the last 9 games last year, 3rd in TDs allowed, 1st in 3rd down %, 2nd in yards per pass, and none of those have dropped off this year).

91
by JIPanick :: Wed, 09/26/2012 - 5:49pm

"They might sneak a playoff spot. They're not going to the Superbowl."

Based on the last decade, in order for your second statement to be true, your first must be false.

96
by Mr Shush :: Wed, 09/26/2012 - 7:49pm

I disagree. Every team since the 2002 Bucs to win a Super Bowl has had at least a very good quarterback. The only team to make a Super Bowl without good quarterbacking was the 2006 Bears, and their offense overall was mediocre, not terrible, while their defense and special teams were absolutely outstanding, not just very good. They were a dominant defense-first team, not a marginal one. Just because teams can sneak into the playoffs and make or even win the Super Bowl doesn't mean that every team that sneaks into the playoffs can realistically make or win the Superbowl. The Cardinals have almost nothing in common with the teams that have had this kind of success.

98
by dmstorm22 :: Wed, 09/26/2012 - 8:32pm

Forgetting Mr. Jake Delhomme, are we?

Plus, Brady in 2003 ranked 9th in DYAR and 13th in DVOA. Eli Manning ranked 40th in DYAR and 34th in DVOA in 2007.

56
by Eddo :: Wed, 09/26/2012 - 9:24am

Mr. Shush went into this in more detail, but are you saying that if Gotskowski doesn't shank a fairly easy field goal, you'd be OK with the Cardinals' rating? Because they're only undefeated due to an incredibly unlikely play that they had virtually no influence on.

53
by langsty :: Wed, 09/26/2012 - 9:08am

How many times have the Falcons been #1 in DVOA?

65
by Peregrine :: Wed, 09/26/2012 - 10:27am

First time that I can remember.

Just hope that the run defense and run offense can hold their own when the glut of takeaways stops (11 in three games). Pass protection has been better than last year, but oddly with the same personnel that got Ryan sacked 13 times in the first three games. Nolan has worked wonders for Moore and DeCoud.

Meanwhile, it seems the special teams coach finally realized the best kickoff strategy was to order Bosher to kick it out of the end zone. Only took 18 games or so, but that's what he's paid to do.

86
by Dan :: Wed, 09/26/2012 - 4:22pm

Atlanta is +10 in turnover differential.

No other team even has 10 takeaways.

88
by JimZipCode :: Wed, 09/26/2012 - 4:28pm

How many times has the Ravens offense been #1 in DVOA/VOA?

90
by Aaron Schatz :: Wed, 09/26/2012 - 5:42pm

This is, in fact, the first time in the history of DVOA that Atlanta has ranked number one overall.

Baltimore's offense actually led the league in DVOA in the first season after moving from Cleveland, 1996. It was a strange alternate universe where the Ravens' defense couldn't stop anyone and Vinny Testaverde was an MVP candidate.

54
by Theo :: Wed, 09/26/2012 - 9:09am

The Steelers are playing without Polamalu, Smith, Farrior and Harrison on defense compared to last year.
Their replacements; Allen/Mundy, Hood, Foote, Carter, aren't as good (although Ziggy Hood is for real). Add the fact that Clarke, Hampton and Keisel are not the players they used to be.
So focussing on the loss of Polamalu means missing the bigger picture here.

57
by Anonymous Coward (not verified) :: Wed, 09/26/2012 - 9:40am

Ziggy Hood is for real what? For real the worst starting 3-4 end in the NFL? He's been awful all season

55
by usernaim250 :: Wed, 09/26/2012 - 9:18am

Redskins DVOA, is not looking good when opponent adjustments come in.

Defense VOA is 13.5, and the three opponents offense VOA are 11.1, -27.3, and 5.9, for a total of -10. Given that DVOA collapses big gains, they are if anything overrated, since they have shown that they will consistently give up big plays to even the worst offenses (the Rams). They are intentionally cultivating boom and bust with cover-0 blitzes.

Offense VOA is 13.6, opponents Defense VOA are -7.6, 16.9, and 38.3, for a total of 47.6, so actually they look below average.

Overall they look like a bottom five team. But they think they are a couple of plays away from playoff contention. I suppose you have to think that way when the truth hurts.

As a long suffering fan, I like what RGIII brings, but when you the team you traded with was worse than you before the trade, promptly embarrasses you after the trade, and is still taking two more of your first rounders, I have to question ever giving that much away when you are rebuilding. The Redskins still have too many huge gaping holes (DHall, both safety spots, nickel corner, RT) and "just a guys" (all those 6th and 7th rounders and street FA on the O Line and in nearly every non-starting spot on the depth chart) to give away four blue chip picks on one. A team that is already solid could do that, but not one that is still three good drafts away from a respectable roster.

If you look at their depth, they only have five guys who were picked above the fifth round. At least they are younger, and some of those young guys might be useful or even good. But basically it's the same old say mold. Aaargh.

69
by Mr Shush :: Wed, 09/26/2012 - 11:39am

The trade's about what you get out of Griffin for 15 years, not 1. If he's as good as he looks like he will be, it's a no-brainer.

73
by RickD :: Wed, 09/26/2012 - 11:56am

Agree completely. I'm on the RGIII bandwagon.

And the problems with the Redskins right now are more on the defensive side than the offense.

76
by Brendan Scolari :: Wed, 09/26/2012 - 12:16pm

Agreed.

92
by bravehoptoad :: Wed, 09/26/2012 - 6:02pm

It's definitely not a no-brainer.

http://harvardsportsanalysis.wordpress.com/2012/03/10/evaluating-the-red...

For the Redskins to get the equivalent value from RGIII as they spent acquiring him, he must produce at least as much as Tom Brady.

Here's another assessment:

http://codeandfootball.wordpress.com/2012/04/30/the-break-even-points-on...

...the Redskins are betting that RG3 will have a Sonny Jurgenson-esque career, and not just his Washington Redskin career, but his Eagles and Redskins career, in order to pay back the ‘AV debt’ that has been accrued by this trade.

In other words, a hall-of-fame career is the minimum requirement for the RGIII trade to break even.

93
by tuluse :: Wed, 09/26/2012 - 6:14pm

That first article is interesting, but I'm not sure it's the right conclusion.

RG3 has to be as good as Tom Brady to beat the expected value of their draft picks.

However, I'm not sure using expected value is the right course of action. Firstly, it's splitting expected value over 3 people. One super valuable player is worth more than 3 average starters. For example: Hunter Hillenmeyer (34), Israel Idonije (30), Danieal Manning (29) add up pretty close to Lance Briggs (90). All 3 are nice players that you don't mind starting and love as backups. However, Lance Briggs is more valuable than all 3 put together even though they slightly edge him in CAV.

103
by Arkaein :: Thu, 09/27/2012 - 10:33am

Right, it's basically the same as the principles behind VOA and YAR. You don't evaluate top players by the number of yards they produce, you evaluate them by the number of yards they produce relative to an average or replacement player.

The marginal performance gains at the top end of the spectrum are worth a lot more than similar deltas in the middle of the spectrum, which is also why the difference in pay between top and average players is a lot more than the difference between average and bad players.

104
by bravehoptoad :: Thu, 09/27/2012 - 1:19pm

I think the flaw in this argument is that you're not counting the two other players who'll have to start next to Briggs. Do you want Hillenmeyer, Idonije, and Manning, or Briggs AND two street free agents, both of whom are most likely to be below replacement value?

The correct answer to that depends on your team. In this case, the Redskins are not a team loaded with depth, shedding lower draft-picks like water.

At any rate, the decision of 3 firsts + a second for RGIII (and three undrafted rookies) is clearly not a "no-brainer."

107
by tuluse :: Thu, 09/27/2012 - 1:40pm

The latter by by far. I've seen this team without Urlacher, and I've seen it with 2 replacement level safeties manning the back end. Losing Urlacher was more devastating than losing averagish safeties.

Also, you're offering a false dichotomy. There is no reason the players the Redskins will use to replace the 1st round picks they lost will be below replacement level. Below average probably, but if they can't find players at or above replacement level they have more problems than losing 2 first round picks.

'At any rate, the decision of 3 firsts + a second for RGIII (and three undrafted rookies) is clearly not a "no-brainer."'

I never said it was, in fact I think they did pay too much. However, I just found a flaw in that article I wanted to point out.

108
by bravehoptoad :: Thu, 09/27/2012 - 1:45pm

No, not you -- it was Mr. Shush's comment about the "no-brainer" that I was initially responding to.

95
by Mr Shush :: Wed, 09/26/2012 - 7:28pm

In addition to Tuluse's observations about AV effectively over-rating longevity and under-rating peak performance, I believe it systematically undervalues top flight quarterbacks, especially in the modern game.

101
by Eddo :: Wed, 09/26/2012 - 10:18pm

In addition to tuluse's and Mr Shush's comments, I'll also add that roster spot availability is a factor.

If Griffin provides 100 units of value, and the picks combined would have provided 120 units, Griffin's still likely more valuable. This is because, along with Griffin, you get three other players. It's not that hard to find three players to make up that value.

Also, you'll want to factor in that it would be easier to retain Griffin for longer, being that he's one player, than a combination of four players that have equal total value. If Griffin's good, the Redskins will likely get 10-15 years out of him. If they had four draft picks instead, they might get 10-15 out of one, 5-10 out of another, and only 3-4 out of the other two.

105
by bravehoptoad :: Thu, 09/27/2012 - 1:25pm

If Griffin provides 100 units of value, and the picks combined would have provided 120 units, Griffin's still likely more valuable. This is because, along with Griffin, you get three other players. It's not that hard to find three players to make up that value.

If it were that easy to find replacment-level players, then the middle rounds of the draft would be of virtually no value. Everyone would spend them swinging for the fences with high-ceiling, low-floor kind of people that would rarely hit.

In other words, along with Griffin, you're likely getting three other players that are below replacement value.

106
by tuluse :: Thu, 09/27/2012 - 1:35pm

Replacement level players are easy to find. I think you're using the wrong wording here. Do you mean average starters or even average backups? Because replacement level is lower than that.

109
by Mr Shush :: Thu, 09/27/2012 - 1:49pm

Right. I mean, that's the definition of replacement level, no? You can sign one off the street at any time.

It gets even more complex when you have to start considering cap room. If you think (as I do) that top quarterbacks are systematically underpaid, Griffin will still be giving you surplus value in cap terms throughout his career. The other hypothetical players probably won't beyond their first contracts.

110
by bravehoptoad :: Thu, 09/27/2012 - 1:50pm

Yes, sorry, average player, since value over average is what they're measuring when they try to determine the value of the RGIII trade. RGIII plus three below-average players for 3 firsts and a 2nd -- that's really the trade.

64
by sanwekwe :: Wed, 09/26/2012 - 10:24am

Setting aside DVOA implications, how different would the playoff odds be if the officials had ruled it an interception at the end of MNF? If GB was 2-1 right now and Seattle 1-2?

66
by Charles Jake (not verified) :: Wed, 09/26/2012 - 10:55am

I'm glad to see the Bears so high (thanks to D and ST as usual), but opponent adjustments and the decline of DAVE are going to kick them square in the nuts.

84
by Jim W. (not verified) :: Wed, 09/26/2012 - 2:51pm

Dallas' false start penalties must be killing their offense's VOA. By my tally they have accumulated 13 false starts (!) so far. That's both pretty incredible and alarming. Their actual pass and run VOA is basically the same as Green Bay's. Green Bay's offensive VOA is -1.3 (18th) while Dallas' is -15.6 (26th). Gotta be the false start penalties.

111
by Web Agnew (not verified) :: Sun, 09/30/2012 - 10:13pm

"DAVE is a formula which combines our preseason projection with current VOA to get a more accurate forecast of how a team will play the rest of the season." Hmmm. That must be a different use of the phase "more accurate" than I generally see. As I mentioned before, DAVE is based upon a biased pre-season perspective. Is St. Louis the 32nd best team in the league? Send DAVE back to formula. Why use analysis to create DVOA and then burden it with opinion? Now let me go look up "more accurate" again...I must be confused.