Week 3 Quick Reads

September 28, 2010, 1:26 pm ET

237

Week 3 Quick Reads

Photo: USA Today Sports Images

by Bill Barnwell

Context is everything. In a league where more information and coverage is available on a daily basis than ever before, it's easy to come up with numbers to support most any opinion. Without the context of where those numbers take place and what predicative value they have going forward, though, statistics are toys of manipulation as opposed to illumination.

The most difficult piece of context to account for while watching games is the quality of the opposition. While it's easy to tell how difficult it is to complete a pass against Darrelle Revis as opposed to, say, Sabby Piscitelli, it's extremely difficult to notice the small differences on a play-by-play basis from one defense to another, especially if you've only seen a small portion of the league's defenses actually play.

Those differences can end up having a dramatic impact on performance, though. The table below lists the statistics recorded by quarterbacks in 2009 against eight of the league's pass defenses. It starts with the team with the best pass defense DVOA, the Jets, and ends with the Lions' league-worst pass defense.

Team (DVOA Rank)	Cmp	Att	Cmp %	Yards	YPA	TD	INT
NYJ (1)	258	498	51.8%	2683	5.4	8	17
CAR (5)	305	495	61.6%	3249	6.6	14	22
WAS (10)	314	511	61.4%	3557	7.0	19	11
IND (15)	372	583	63.8%	3631	6.2	19	16
ATL (20)	335	535	62.6%	4041	7.6	25	15
TB (25)	300	481	62.4%	3457	7.2	28	19
CLE (30)	312	524	59.5%	4128	7.9	22	10
DET (32)	369	543	68.0%	4353	8.0	34	9

You might guess where this is leading to. Michael Vick's now played ten quarters of football, and eight of them have come against the pass defenses that ranked 31st (Jacksonville) and 32nd (Detroit) in DVOA a year ago. Jacksonville remained 31st through the first two weeks of the season, and while Detroit has looked improved, there's been more than a few favorable turnover bounces that have gone their way through three weeks.

The truth is that what we've seen out of Vick through those ten quarters has come in contexts extremely favorable to the likelihood of Vick succeeding. His other half came against the Packers in what amounted to a mop-up role against a team that had not prepared for Vick to be a full-time quarterback, and didn't have the spy packages put in that have neutralized Vick in the past. He will not face an above-average pass defense until Week 7, when he gets the Titans; after that, he may not face another one until Chicago in Week 12 or Dallas in Week 14. The Eagles face what appears to be a remarkably easy slate of pass defenses this year outside of, coincidentally, Week 1.

DYAR begins to adjust for opponent strength after Week 4, so you'll begin to see opposition weighted at 40 percent of its eventual strength after next week. (It gets to 100 percent after Week 10.) They will likely take Vick's numbers -- already considered to be slightly above-average at best -- and push them down towards average-or-worse. If Vick's improved as a player, the last two games haven't indicated that through his level of production on the field. He remains inefficient and erratic. He's improved his timing, but it's not as a passer. Outside of Matt Cassel in 2008, no quarterback has probably entered into a better scenario at the right time than Michael Vick with these Eagles after Week 1.

Quarterbacks
Rk	Player	Team	CP/AT	Yds	TD	INT	Total DYAR	Pass DYAR	Rush DYAR
1.	Aaron Rodgers	GB	35/45	320	1	1	198	178	20
Rarely will you see a quarterback consistently get seven yards as easy as Rodgers did during the second half of the MNF game. While the Bears were able to get pressure on Rodgers in the first half and force him into some pretty impressive scrambles, the Packers let Rodgers throw out of a quick one-step drop a fair amount of the time in the second half. The result was a final two quarters of 18-of-21 for 138 yards and seven first downs. Rodgers' only real mistake was his penultimate throw, a dubious intentional grounding call. While nobody was in the area of the throw and it landed out of bounds, it was a pretty clear case of miscommunication on a route. Had the receiver ran the route it appeared Rodgers was expecting him to, while the throw would have been off, it would not have been so far off that it would have necessitated intentional grounding. In a league where grounding gets called far too infrequently, this was one case where the flag should have stayed in the refs' pocket.
2.	Joe Flacco	BAL	22/31	262	3	0	190	190	0
The inverse of Vick. Flacco had been derided for a rough start based on two road games against the Jets and Bengals, who are very likely to rate among the league's best pass defenses this year. With a friendlier matchup, Flacco was able to split the Browns apart with one of the greatest 10-dropback sequences you'll ever see. How good was it? Try 10-for-10 for 159 yards, with every pass going for either a first down (eight) or a touchdown (two). He picked up six of the 11 third downs he faced on the day, and went 5-on-6 for 79 yards on the eventual game-winning drive in the fourth quarter.
3.	Tom Brady	NE	21/27	252	3	0	176	172	5
It appears that the Bills pass defense will not be quite as good this season. Brady ate their lunch on Sunday, going 7-of-7 with 76 yards (with one sack) on first downs against them. Eight of his 28 dropbacks came with 15 or more yards to go, and he didn't pick up a successful play on a single one of them; on his other 20 dropbacks, though, he had a success rate of 75 percent.
4.	Tony Romo	DAL	24/30	284	2	0	176	176	0
After a 5-for-19 start to the season, the king of converting third downs returned to his ways against the Texans on Sunday, going 4-of-8 and throwing in a 1-for-2 performance on fourth down in for good measure. It was a case of Romo adapting to what the defense gave him, as the Texans sold out to stop Miles Austin and held him to two catches for 20 yards (and a seven-yard pass interference penalty) on four targets. For this week, Roy Williams and Dez Bryant had enough to beat one-on-one matchups pretty easily.
5.	Drew Brees	NO	30/37	365	3	2	158	158	0
Brees started the game 16-of-18 for 212 yards, but because the two incompletions were interceptions, the Falcons were able to stick around. He was also close to unstoppable on the final two drives of the game, going 10-of-11 for 97 yards with five first downs, including a fourth-and-1 conversion to help push the game into overtime and a third-and-9 conversion on his final pass to set up what appeared to be a game-winning field goal. And the next four weeks for Brees: Carolina, Arizona, Tampa Bay, Cleveland.
6.	Peyton Manning	IND	27/43	325	3	0	157	157	0
In much the same way that we've talked in the past about giving Kyle Orton a pass for his rookie season, it's not really fair to blame Perrish Cox for being the Broncos' third-best cornerback coming out of training camp. And it's unfortunate that Andre Goodman's injury forced him into a starting role. But there's no arguing with the result: In yet another Colts win over the Broncos, Peyton Manning found yet another weak link at cornerback for the Broncos to pick on. While we won't know about the charting for a week or two, both of Manning's final two passes to Austin Collie -- which went for 71 yards and a crucial fourth-quarter touchdown -- came with Cox involved with the coverage.
7.	Mark Sanchez	NYJ	15/27	256	3	0	155	155	0
Jason McIntyre of The Big Lead did not take kindly to our article comparing JaMarcus Russell and Mark Sanchez's rookie seasons last year, and periodically checks in on our "comical" bashing of Sanchez on Twitter when the Jets' quarterback has a big game. Last night's check in suggested that Sanchez played a clutch game because he went 5-of-8 on third down against the Dolphins, getting five first downs in the process. And he's right about that -- Sanchez came up big on third down. On the other hand, he converted 34 percent of his third downs last year versus a league average of 37 percent, and before Sunday, he was at 6-of-17 (35 percent) on third down. I wonder if Sanchez will pick up 63 percent of his third downs going forward.
8.	Kyle Orton	DEN	37/57	481	1	1	154	147	7
That's an impressive yardage total -- even for 57 attempts -- but it's a lot of volume and not all that much efficiency. Take third down, for example: Orton was 7-of-12 for 61 yards, but he only picked up three first downs on those throws. He was 1-of-3 on fourth down, with all three attempts inside Colts territory in the fourth quarter. His 16-of-23 performance on first down had a success rate of 52 percent. The real killer: In the red zone, Orton was 4-of-14 for 20 yards without a touchdown or a first down.
9.	Matt Ryan	ATL	19/30	228	2	0	151	144	7
A quietly effective game that would have looked better had Ryan finished it in overtime; a long bomb to a wide-open Harry Douglas for a game-winning touchdown was thrown about one yard too deep. As it were, Ryan found out very early that the Saints couldn't cover Tony Gonzalez repeatedly and went back to him again and again; on passes that weren't to Gonzalez (we'll get to him later), Ryan went 11-of-21 for 118 yards. His two best throws of the day were both to Roddy White; with Jabari Greer attempting to jam White at the line, White eluded Greer's grasp and gave Ryan a small window of space inbetween Greer and safety Malcolm Jenkins. Ryan hit White the first time on a perfect lob down the sideline for 19 yards, and then finished the drive with an easier touch pass to White in the end zone for a 22-yard score.
10.	Chad Henne	MIA	26/44	363	2	1	140	144	-4
Henne also had a 27-yard defensive pass interference penalty drawn by Brian Hartline that's not included in the numbers above. It's hard to figure out how much of Henne's performance had to do with the absence of Darrelle Revis, but he certainly looked like he had more time in the pocket than he normally would. After two first quarter sacks, the Jets weren't able to get to Henne the rest of the way. He peaked in the second quarter, with an eight-pass stint that saw him pick up 108 yards and a first down or a touchdown on each pass.
11.	Michael Vick	PHI	17/30	291	3	0	126	112	14
While highlight shows featured what Vick did on his three completions of 42 or more yards, he was wildly ineffective at times. On Philly's first four drives, Vick had the 61-yard touchdown pass to DeSean Jackson ... and otherwise went 2-of-8 for six yards with a sack. After an effective two-minute drill to get an 11-point lead, Vick led two possessions that combined for seven plays by going 2-of-4 for 22 yards and a sack, with both completions coming on third-and-long and ending up short of the sticks, leading to punts. He was only particularly effective once he got the ball back again, with 5:30 left in the third quarter. That the Jaguars had only scored three points had nothing to do with Vick; that the Eagles had only mustered 14 was mostly on him. The great arm and athleticism are still there, and Vick has certainly improved his patience in the pocket. But it's easy to do that against a dreadful pass defense.
Rk	Player	Team	CP/AT	Yds	TD	INT	Total DYAR	Pass DYAR	Rush DYAR
12.	Philip Rivers	SD	29/52	455	2	2	124	127	-3
The Seahawks were terrified of Philip Rivers beating them deep. They played a ton of two-deep coverage and let Rivers just have his way with them underneath. Rivers lived on the hashmarks, completing what seemed like endless dig (wide receivers) and seam routes (Gates) in the second half at will. Once he got into the red zone, though, the safeties didn't have to worry about getting beat deep, and Rivers ended up trying to force those same throws. As a result, he was 2-of-9 inside the red zone, although both completions went for touchdowns.
13.	Matt Cassel	KC	16/26	250	3	1	101	106	-5
Matt Cassel's two biggest plays were two touchdown passes, but it's hard to say that he made them happen. The 45-yard completion to Dwayne Bowe was on a trick play that saw Cassel line up at wide receiver, and the 31-yard pass to Dexter McCluster went six yards in the air. His two other completions of 20 yards or more traveled a combined three yards in the air. In all, 52 percent of Cassel's yardage came from receivers after the catch; the league average for all other quarterbacks in Week 3 was 42 percent.
14.	Sam Bradford	STL	23/37	235	1	1	90	93	-3
DeAngelo Hall's defense did not have much of a game, to be quite honest. With Hall running the ship, you would think that he'd put a particular focus on shutting down opposing wide receivers. Instead, Bradford was 16-of-24 for 195 yards and 11 first downs on throws to his wideouts, including a 14-yard DPI.
15.	Jay Cutler	CHI	16/27	221	1	1	76	58	17
The difference between the Jay Cutler who threw a handful of interceptions against the Packers last year and got away with just one on Sunday night isn't very large. Cutler threw two picks on the final drive alone, with one being called a questionable incompletion and the other a defensive pass interference penalty. He forced a throw into Earl Bennett on third-and-goal from the six-yard line that was nearly picked, and on the ensuing fourth down play, he threw behind an open Desmond Clark. (Clark still should have caught the pass.)
16.	Vince Young	TEN	10/16	118	1	0	70	75	-5
17.	Donovan McNabb	WAS	19/31	236	1	1	68	57	11
McNabb had a success rate of just 39 percent, and he was lucky to have an interception in the red zone saved by some great Chris Cooley defense. He only picked up eight first downs all day, and three of them came in the fourth quarter while the Redskins were down by two scores. The killer: McNabb faced eight third downs and converted exactly zero of them. In all fairness, six of the eight came with nine or more yards to go, but 0-of-8 is bad no matter how you slice it.
18.	Matt Hasselbeck	SEA	19/31	220	1	1	66	61	5
Hasselbeck to the left side of the field, where Antoine Cason starts: 11-of-12 for 116 yards. Hasselbeck to the right side of the field, where Quentin Jammer starts: 6-of-15 for 96 yards (with a 23-yard DPI included.) It might not be time to crown Cason just yet.
19.	Ryan Fitzpatrick	BUF	20/28	247	2	2	57	56	1
This is how bad the Patriots pass defense is going to be this year. Ryan Fitzpatrick -- who has completed 58 percent of his passes historically -- went on a 14-of-18 streak for 178 yards to end the game. And it wasn't in mopup time, either, as only four of the passes came while the Bills were down by more than eight points. The Patriots were able to turn two of Fitzpatrick's incompletions into interceptions, but if Ryan Fitzpatrick and the Bills offensive line can move the ball comfortably on you...
20.	Charlie Batch	PIT	13/17	184	3	2	37	25	12
Batch's two long touchdown throws to Mike Wallace both should have been interceptions, with the first pass getting lost in the sun and the second one tipping off of Aqib Talib's hands and into Wallace's. 12 straight completions is nice, but Batch could easily have thrown four picks in 17 attempts, which wouldn't be. And he's favored against the Ravens next week?
21.	Seneca Wallace	CLE	18/24	141	1	0	32	32	0
Then again, I was worried for Wallace's health on the road against Baltimore, and he made it out alive. He was actually remarkably effective in the first half, going 12-of-14 for 114 yards and seven first downs. In the second half, though, Wallace was sacked twice on his first four attempts, and finished 6-of-10 for 27 yards. Those sacks and a whopping 17 yards on an aborted snap taken at his own 20 meant that the Browns had zero net passing yards in the second half.
22.	Eli Manning	NYG	34/48	386	0	2	29	27	2
Could we crowd-source interceptions? It's pretty clear that Eli Manning isn't responsible for the picks he threw on Sunday, but in our binary system of accounting for them, he either gets one full interception or no interception for each pass he throws. I wonder if we could ask readers to basically place a value on Manning's interception and say "This one looked like it was 25 percent Eli's fault." It might be too much guessing about intent, I suppose, to do with any sort of reliability.
Rk	Player	Team	CP/AT	Yds	TD	INT	Total DYAR	Pass DYAR	Rush DYAR
23.	Matt Schaub	HOU	23/32	241	1	2	26	26	0
Schaub on third down: 3-of-3 for two first downs. Not included in those numbers: Four sacks.
24.	Derek Anderson	ARI	12/26	122	2	1	18	18	0
122 yards on 26 attempts looks downright gruesome, but Anderson drew three pass interference penalties in the first quarter that netted the Cardinals 74 yards. That turns his YPA from 4.7 to a much more respectable 6.8. To follow up the Eli Manning conversation, though, if we're going to give Eli a quarter of an interception for one of his passes, we need to find a way to give Derek Anderson three picks for his bullet screen pass to Tim Hightower.
25.	Bruce Gradkowski	OAK	17/34	255	1	1	15	24	-9
If I told you that Bruce Gradkowski had a game where he threw 13 passes to Darrius Heyward-Bey and nothing more, what would you say the Raiders' odds of having won that game would be? Three percent? Five? And yet, there they were, a chip shot field goal away from beating the Cardinals on the road. Heyward-Bey picked up 51 yards in pass interference penalties, which actually would have gotten him all the way up to 100 yards.
26.	Carson Palmer	CIN	19/37	195	1	2	2	-3	5
27.	Shaun Hill	DET	29/43	237	1	2	-15	-14	-1
Those 43 attempts yielded exactly two completions that traveled further than 11 yards in the air. The Lions had a chance to get within a touchdown in the fourth quarter, but Hill threw an interception in the end zone that just about sealed their fate. He followed it later with a second pick in the end zone, but that was on his final pass attempt and with a minute left.
28.	Brett Favre	MIN	23/34	201	1	2	-20	-20	0
Favre completed only two passes in that 8-to-14 yard intermediate range, although he did get two defensive pass interference penalties on throws to former DPI fiend Bernard Berrian. 19 of his 23 completions were for seven yards or less. This is just about as much of a screen-and-sling offense as you can get, and while part of that is due to what's been an awful offensive line at times, the end sum is not particularly effective.
29.	Josh Freeman	TB	20/31	184	0	1	-36	-29	-8
It will look better once the opponent adjustments hit 100 percent, I promise. Earnest Graham picked up 46 yards on a third-quarter pass. Last week, it took Cadillac Williams 27 carries just to get to 51 yards.
30.	Alex Smith	SF	24/42	232	1	1	-87	-87	0
The triumverate of bad quarterback play -- the anti-Gordie Howe hat trick, I suppose -- is an interception, an intentional grounding penalty, and a stripsack. Smith managed to hit the trifecta on Sunday. Before the meaningless drive that ended with Josh Morgan wrenching his knee for some fantasy points at the end of the game, Smith managed to pick up six first downs on 34 attempts.
31.	Jimmy Clausen	CAR	16/33	188	0	1	-92	-86	-6
I don't recall ever seeing three aborted snaps for one quarterback in one game, but Clausen had three of them through 18 dropbacks and three quarters. Well, theoretical dropbacks, I guess. He turned the ball over on two of them, and he also had a sack and an interception thrown in for good measure. In the fourth quarter, he was reasonably effective, but he was also down two touchdowns for most of it.
32.	David Garrard	JAC	13/30	105	0	1	-128	-120	-9

Five most valuable running backs
Rk	Player	Team	Rush Yds	Rush TD	Rec Yds	Rec TD	Total DYAR	Rush DYAR	Rec DYAR
1.	Jamaal Charles	KC	97	0	57	0	72	36	36
Todd Haley can do whatever he wants after leading the Chiefs to a 3-0 start, but his treatment of Charles remains curious at best. With the team apparently set on establishing veteran Thomas Jones as the primary back, Charles has played the ancillary role through three weeks. So how did he pace all backs in DYAR this week? Small miracles. He converted both the third downs he faced, including a 24-yard run on a third-and-20 draw. Seven of his 12 runs went for five yards or more, and four of them went for double-digit yardage. He was even more effective as a receiver, where all three of his targets resulted in catches, first downs, and gains of anywhere between 17 and 22 yards. What else does he have to do to get a larger share of the workload in Kansas City? Bigger miracles.
2.	BenJarvus Green-Ellis	NE	98	1	6	0	62	56	6
The only way I know of to assure yourself a large majority of the carries in New England is to have every other back get hurt. Kevin Faulk's done for the season, and Fred Taylor went down with an injury during the game on Sunday, which opened up Green-Ellis's chance for regular yardage. He's probably not versatile enough to be on the field for a majority of the snaps, but he's an effective enough runner to get the carries when the Patriots are killing the clock.
3.	Peyton Hillis	CLE	144	1	36	0	58	51	7
For one week, at least, those fantasy owners that have Jerome Harrison on the bench and Eric Mangini in their doghouse owe Mangini an apology and perhaps a DVD of a classic fight or two. Hillis actually picked up serious chunks of yardage against the Ravens, running for 144 yards on 22 carries, with runs of 25 and 48 yards along the way. He picked up the first two third downs he faced, finishing the second one with a one-yard plunge for a score.
4.	Rashard Mendenhall	PIT	143	1	0	0	57	57	0
While the fear has rightly been that Mendenhall will lose goal-line carries to Isaac Redman, Mendenhall scored on a three-yard plunge in the second quarter against the Bucs, while Redman was stuffed on a fourth-and-1 attempt inside Tampa territory in the third. Mendenhall didn't take a single carry on third down all game, which is disconcerting, but when you average 12 yards per carry with a success rate north of 50 percent on your 13 shots at first-and-10, it doesn't really matter.
5.	Marshawn Lynch	BUF	79	0	7	0	35	31	4
If the Bills had known Lynch would actually be a reasonably effective running back, would they still have taken C.J. Spiller in the first round? Could they have possibly drafted an offensive lineman? Would that have made Trent Edwards look effective enough to keep his job? In the end, Spiller had a nice kick return for a touchdown, but he's got 23 touches from scrimmage for 79 yards through three weeks. And they've got three running backs in a league that's begun to realize how fungible running backs are.

Least valuable running back
Rk	Player	Team	Rush Yds	Rush TD	Rec Yds	Rec TD	Total DYAR	Rush DYAR	Rec DYAR
1.	Keith Toston	STL	22	0	0	0	-28	-28	0
22 yards on 11 carries is ugly, but without a fumble, it seems pretty innocuous; bad, sure, but not bad enough to serve as the worst performance of the week. So how did Toston get here? He did his Matt Forte impersonation. With two carries from the 1-yard line in the second quarter, Toston rushed for -1 yards and then -3 yards.

Five most valuable wide receivers and tight ends
Rk	Player	Team	Rec	Att	Yds	Avg	TD	Total DYAR
1.	Anquan Boldin	BAL	8	11	142	17.8	3	88
Much like Flacco, Boldin had a streak of downright dominance on Sunday. He caught seven consecutive targets for 135 yards, yielding four first downs and three touchdowns. He was thrown four passes on third down and they were all catches and conversions. Even two of his three incompletions went 25 yards in the air or more. That 88 DYAR total is the best performance of a receiver so far this season, and it would have ranked fifth a year ago.
2.	Lance Moore	NO	6	7	149	24.8	2	77
Moore had become a bit of a forgotten man last year after the emergence of Robert Meachem, but he remains a capable part of the offense and had just a huge performance on Sunday. In addition to his lengthy punt return in the first quarter, Moore benefited from a totally blown coverage to pick up an 80-yard score. And even outside of his long touchdown, he was 5-of-6 for 69 yards, including a crucial second down conversion in overtime that helped set up Garrett Hartley's game-winning field goal. The problem Moore has isn't talent, but availability: Moore's struggled with injuries as a pro, and in such a dynamic offense, it's easy to get lost in the hunt for targets when Drew Brees develops a rapport with one of your competitors while you're gone. While some quarterbacks have a particular receiver they fall for, Brees's favorite guy is the open one.
3.	Tony Gonzalez	ATL	8	9	110	13.8	1	68
Well, he sure isn't done yet. Gonzalez pulled all kinds of tricks out of his bag against the Saints. He started off most frequently against Tracy Porter, who was in coverage on Gonzalez's touchdown catch. When Porter went down, the Saints were forced to use Roman Harper against Gonzalez, and Harper had no hope, with Gonzalez using his body to shield Harper from throws that ended up being easy catches.
4.	Roy Williams	DAL	5	6	117	23.4	2	67
For one week, Jerry Jones was a genius. With the porous Texans secondary seemingly giving him the run of their backfield, Williams downright dominated at times on Sunday night. On both of his touchdown catches, Williams was able to get a free release at the line of scrimmage and then use his underrated route-running skills to get downfield for touchdowns on go routes. His only incompletion on a 5-of-6 day came on a second-and-20 throw. Don't expect this to continue, since most teams have better cornerbacks than Houston and can just bump Williams off of his routes at the line of scrimmage. But hey, it was nice while it lasted, right?
5.	Austin Collie	IND	12	16	171	14.2	2	64
Despite not starting any of the Colts' first three games, Collie has caught 27 passes for 359 yards. In the DVOA Era (1993-2010), that's a record start to the season for a receiver who hasn't started a game. In 1994, both Mike Pritchard and Glyn Milburn of the Broncos had 19 catches through three games without starting, although Milburn was a hybrid running back/receiver. Pritchard had a era-best 271 yards, which paced the rest of the opposition by 49. And Collie's 88 yards ahead of him. Suffice to say that the Colts' slot receiver is off to an impressive start.

Least valuable wide receiver or tight end
Rk	Player	Team	Rec	Att	Yds	Avg	TD	Total DYAR
1.	Chad Ochocinco	CIN	4	12	34	8.5	0	-45
Thrown passes on five third downs, Ochocinco was only able to convert one of them into a new series. Two of his catches were quick hitches. Will he finish with fewer DYAR than Terrell Owens? It's going to be an exciting race to the bottom.

Football Outisders+ Perks & Features image

Comments

237 comments, Last at 30 Sep 2010, 6:51pm

#216 by onetime91 // Sep 30, 2010 - 7:21am

I'm looking forward to the game where Vick performs just an average game. He'll probably be ranked 42nd behind a handful of backups who didn't get on the field.

#219 by nat // Sep 30, 2010 - 9:29am

1) Don't obsess about rankings. Look at the actual value of DYAR. Vick's 126 is pretty good. He's closer to ranking 3rd than he is to being average.
2) All the QBs listed ahead of him had better completion percentages, when you factor in sacks. Completion percentage isn't everything, but it matters a lot.
3) Sacks are worse than incompletions. Vick was sacked three times. This counts against his passing DYAR.
4) DYAR is a cumulative stat. Vick was much better per-play than Orton, for example. But Orton threw almost twice as many passes, and so produced more value in total. Vick played better, but Orton was in more plays.

Personally, I would like FO to include single game DVOA for the QBs here, too, so we could talk about per-play quality in parallel with total production. But that's not what this list is trying to convey, so they don't.

While some FO commentators may act like somewhat irrational Vick-deniers, the stats themselves are unbiased. This last week, Vick was quite good, but by no means the best or most productive QB in the league. Relax and enjoy the ride.

#225 by onetime91 // Sep 30, 2010 - 2:43pm

Sacks aren't always the quarterback's fault. The Eagles offensive line is historically bad right now. Is it really fair to count sacks more negatively than incompletions? Also, if they really want an accurate analysis of quarterback performance, they should incorporate dropped passes into the equation. Vick had atleast 4 dropped balls. The stat as it is now is quite flawed.

#231 by nat // Sep 30, 2010 - 3:27pm

Go read "About: Our New Stats Explained" if you haven't already. It'll help you understand what's going on here. They explicitly discuss the "quality of teammates" issue with DVOA and how to think about it.

A QB's DYAR and DVOA are about that QB, with that line, with that running threat for play action, with those receivers. You are free to look at DVOA and say "that's because the rest of his team sucks!" if you want. It might even be true to some extent. But don't expect an objective stat to incorporate your subjective ideas about who to blame for incompletions and sacks.

#233 by horn // Sep 30, 2010 - 3:35pm

Fair enough, but sacks can be objectively assigned in many cases, just like INTs can be, just not always. You have heard of a 'blown assignment,' since you peruse this site I assume...

#234 by nat // Sep 30, 2010 - 4:04pm

"Objectively"
You keep using that word. I do not think it means what you think it means.

#235 by chemical burn // Sep 30, 2010 - 4:45pm

Listen, I'm speaking positively of Vick's performance as much as anybody here... but at least 3 of those drops were primarily on him. Objectively speaking.

Vick played well behind a bad line and surprised in many ways... but be real, he also frequently displayed some of the terrible mechanics and poor accuracy that caused FO's numbers to be down on him for his entire career.

Also, it is fair and logical to count sacks more negatively than incompletions. It is not the entire end of the story, but I completely trust numbers that penalize more heavily for sacks than incompletions.

#237 by t.d. // Sep 30, 2010 - 6:51pm

Sacks actually are much more tied to quarterbacks than is commonly conceived. Both FO and PFR have done research indicating this

#215 by onetime91 // Sep 30, 2010 - 7:19am

3rd down conversions shouldn't add anything to a player's value. They are no different than a 1st or 2nd down conversion. This site needs to read some baseball prospectus, it will likely help them. Judging a quarterback on 3rd down conversions is unbelievably inaccurate. It's like judging a basketball player on his last shot of the game or a baseball player on clutch hitting. It's stupid, a small sample size and has no sustainability.

#217 by Thomas_beardown // Sep 30, 2010 - 7:28am

Did you even read the blurb about Sanchez?

They're making fun of someone else calling him clutch.

#218 by ammek // Sep 30, 2010 - 7:52am

Also, in the introduction to this year's Almanac, it is explained that, for the past three years, for offenses, third-down DVOA has been more predictive than first- or second-down DVOA. It remains to be seen whether this is a blip, but the argument is less concrete than you make out.

#222 by commissionerleaf // Sep 30, 2010 - 12:50pm

3rd downs aren't "clutch", in that sense. They're different and important because they are a pure test of passing for quarterbacks, against a defense that mostly knows they are throwing and schemes against that. On first and second down, defenses are less able to play the pass. But on third down, especially third and long, a lot of the trickery (play-action, etc.) isn't possible because that stretch play isn't going to get you eight yards.

[Also, with one and only one play to get any particular yardage, passing is a good idea because you can pick a receiver to throw to who will get the yardage you need].

Hence, third downs are a good test of quarterbacks as quarterbacks, facing blitzes or heavy coverage, and success in that context is both important and predictive.

#206 by Veloso // Sep 29, 2010 - 5:17pm

"And even outside of [Lance Moore's] long touchdown, he was 5-of-6 for 69 yards, including a crucial second down conversion in overtime that helped set up Garrett Hartley's game-winning field goal."

Wow, way to stick it to us Saints fans there.

#190 by jeff (not verified) // Sep 29, 2010 - 4:28am

You should correct your offensive line stats for the Eagles. THe NFL revised the sack number from the Detroit game from 6 to 5.

#173 by the K (not verified) // Sep 28, 2010 - 9:48pm

Williams downright dominated at times on Sunday night

He didn't look too bad Sunday afternoon either, in the game that had a 1 ET 10 PT kickoff time.

#165 by TruFalcons (not verified) // Sep 28, 2010 - 9:07pm

Vick hasn't had much time with those receivers either. As he works with them he'll likely perform even better. He came in with no gameplan against the Packers and was dominant - yet if anything gets discredited for that because apparently the Packers are the only ones allowed to implement a gameplan.

#167 by halfjumpsuit // Sep 28, 2010 - 9:10pm

They were all there last year, and he got at least some reps with them for Wildcat plays. It's not like his next TD to Celek will be his first.

#232 by horn // Sep 30, 2010 - 3:33pm

I'm not sure if this is a joke, or a troll, but Vick had exactly one passing TD in the regular season last year - to Celek.

#157 by F.Leghorn (not verified) // Sep 28, 2010 - 8:19pm

Who knew so many people loved Michael Vick so deeply? The pages here are drenched with tears of Vick devotees who feel he just isn't getting his due.
It is funny how a couple of decent games can change people's perceptions totally about a player. It's what makes sports, and celebrity, feel rather ludicrous.

#170 by chemical burn // Sep 28, 2010 - 9:34pm

I hate Michael Vick. You know what else I hate? Bad analysis. Unfortunately, Vick now QB's for the only team I give a shit about. And much more surprisingly, FO (in my opinion the gold standard for football writing) has only offered bad analysis of his performance as the QB.

I wish Vick weren't playing football. I would probably prefer if he were still in jail. I don't believe he has changed as a human being and I have my suspicions about what happened with the shooting at his birthday party this summer. But these guys aren't my friends, a bunch of them are probably terrible people and I'll never have the slightest inkling of it, it's morally pointless for me to take a stand and stop watching football because he's the Eagles QB.

For 2 and a half games, the guy has been playing and FO has written about him extensively. I expect more from their coverage because, with exception of Doug Farrar's writing, it has been awful. If you asked me if I would mind if a giant safe fell on Vick tomorrow and he died, I would say "Not at all." If you asked me if he was playing in the same fashion as 2004, I would say "So far, not even slightly. I am deeply surprised."

#205 by tunesmith // Sep 29, 2010 - 3:47pm

Nah. I think the point is more that saying "Vick is playing really good football, but he kills dogs" would be more honest than to highlight all these clues and indicators that his football is theoretically not as good as it appears to be.

#148 by bengal85 // Sep 28, 2010 - 7:01pm

via ChemicalBurn:

"The knee-jerk, "don't believe the hype - and Vick is ALL hype" response is frustrating because for the moment Vick is playing like a completely different player - the questions are why and how and if it is sustainable. To answer those questions with a simple, "The opposition blows and he's not playing good, anyway" is just so reductive and simplistic - it's like a mirror image of Sportscenter's "The King is back, baby!" analysis. I just expect better from FO. Except if it Barnwell writing."

This is EXACTLY WHAT I WAS SAYING. I cannot agree more.

And, for Peterson, again, without getting too specific, the number of catches were way, way too low, based on a regression of his previous years, completely failing to account for
-The departure of Chester Taylor,
-Peterson's expanded role on 3rd down (both public knowledge, might I add)

The number came out as less than he had last year, despite the fact that between him and Taylor, 80+ catches were made.

Anyway, as someone else stated, common sense rules the day. We are picking nits, as they say. Still a phenomenal product on draft day.

#138 by Jim in VT (not verified) // Sep 28, 2010 - 6:41pm

The Batch attack is only partially well founded. Reporting from Pittsburgh outlets like Ed Bouchette of the P-G revealed that Batch read safety Grimm's tendency to turn his back, and that Batch told wallace in the huddle that if Grimm turned his back on a fly/post pattern that he would try to drop it over the top. Grimm did, and Batch did.

The second TD on the Talib deflection was lucky, but the first was, as Branch Rickey would say, the residue of design.

#142 by Mountain Time … // Sep 28, 2010 - 6:45pm

That is very interesting, thanks for posting! Do you happen to have a link?

#156 by Jerry // Sep 28, 2010 - 8:01pm

Peter King had it in yesterday's MMQB.

#158 by Rocco // Sep 28, 2010 - 8:20pm

Who knew that going after a guy who was a try-really-hard linebacker in college posing as a safety in the NFL would yield positive results?

#133 by 1618 (not verified) // Sep 28, 2010 - 6:25pm

This is really sad. I appreciate the articles, and at their relative price who wouldn't, but I think I'm done here. Good luck you guys =)

#112 by onetime91 // Sep 28, 2010 - 4:55pm

Nat, you hit it right on. The fact that Mark Sanchez is higher than Vick is disgusting. He literally did nothing better than him at all except convert an unsustainable amount of 3rd downs.

#147 by bravehoptoad // Sep 28, 2010 - 7:01pm

Hmm...where to put the "unsustainable" variable in a one-game performance statistic? Because you're right on. I hate those quarterbacks that convert an ungodly amount of 3rd downs. That's why I love Alex Smith!

#214 by onetime91 // Sep 30, 2010 - 7:12am

Start reading more articles about statistics in any sport and you'll begin understand what's unsustainable. You seem to be out of your element.

#152 by Mountain Time … // Sep 28, 2010 - 7:24pm

I don't think the word "literally" means what you think it means.

#153 by nat // Sep 28, 2010 - 7:32pm

Thanks. But I can't agree back at you.

If Sanchez had a higher YAR, then he did produce more (relative to a replacement, throwing the same throws, yada yada yada...) But the difference in production (29 YAR) isn't very big, and could easily be due to Vicks three sacks, or the particular downs and distances they threw in. Their results in conventional stats were pretty close.

My beef with Bill's Vick comments was not with the stats. It was with his poor analysis. It was sufficiently below FO standards of objectivity as to simulate trolling. I wonder if each year the FO team decides on a topic to make into a controversy, and intentionally ignores or misuses their own stats to generate traffic.

It's a shame, really. I actually like opinionated writing.... in its place, if it's well done. This is neither.

#106 by nat // Sep 28, 2010 - 4:41pm

Bill, quit with the Vick trolling. It's neither seemly nor insightful.

The man had 112 YAR on 33 drop backs. He had a 27 yard P.I. in his favor that doesn't get included in his numbers, just like Henne did. If he was at times "wildly ineffective", his numbers show him to be at times wildly effective too, with the balance in his favor.

His VOA and YAR are not "slightly above average at best". They are 24.6% and 223 (in 10 quarters), compared to medians for the top 32 QBs of 5.6% and 77. We routinely bitch-slap FO newbies for using rankings without looking at the actual numbers. Consider yourself slapped, B.

#121 by chemical burn // Sep 28, 2010 - 5:22pm

What's frustrating is that Vick is clearly playing DIFFERENTLY than he played before, even if he is not suddenly a Top 10 pure passer, guy who sets his feet on every pass, etc., whatever measure FO is determined to point out he's failing to live up to.

Just look at his DVOA as a passer from previous years - almost every year it is downright awful. Now, of course opponent adjustment will drag him down a bit, but in two games he is playing very different than he ever did in the pass: significantly higher VOA as a passer, fewer runs, making different reads.

The knee-jerk, "don't believe the hype - and Vick is ALL hype" response is frustrating because for the moment Vick is playing like a completely different player - the questions are why and how and if it is sustainable. To answer those questions with a simple, "The opposition blows and he's not playing good, anyway" is just so reductive and simplistic - it's like a mirror image of Sportscenter's "The King is back, baby!" analysis. I just expect better from FO. Except if it Barnwell writing.

#127 by dmb // Sep 28, 2010 - 5:47pm

"Quality of opponent" may be a simple argument, but that doesn't make it an inaccurate one. Perhaps the most memorable example is Derek Anderson's 2007 season, where a case could be made that a really bad QB looked like a really good one due mostly (if not entirely) to playing a stretch of cupcake pass defenses. If playing bad teams could make Derek Anderson look that good, then it's entirely plausible for, say, a slightly below-average player to look elite when playing two games against horrendous defenses.

That's not to say that I necessarily agree with all of Bill's analysis, or even that this happens to be the case for Vick. It's just worth noting that opponent quality can be a big, big, BIG deal.

#154 by chemical burn // Sep 28, 2010 - 7:46pm

Yeah, but by FO's own numbers he had his best game against the #5 ranked defense. And don't give me a "they didn't plan for Vick." If they didn't, they're stupid because Reid said all off-season that Vick would see the field on 12-15 plays a game. He even started the game and played on the very first snap. The Packers should have had something resembling a plan for Vick. The Eagles played a good defense, a mediocre defense and a bad defense - that is not 3 cupcakes...

#178 by dmb // Sep 28, 2010 - 10:33pm

*A team with only two healthy defensive linemen is definitely not a team that's fully prepared for Vick.

*Coaches say a lot of things about what they will or won't do, and though some of those things are true, it would be pretty foolish for anyone to take public proclamations from NFL head coaches at face value.

*Considering that Vick is currently being used in a significantly different role than he was in the special "Vick package," I think it's reasonable to suggest that GB's preparations for Vick were designed with the latter in mind, and might not be as useful once he was placed in the former.

*I'm unsure which defense you'd classify as "mediocre," since JAX and DET have both started out pretty terribly in 2010, were both atrocious last year, and both have secondary personnel that so far seem ... underwhelming.

I'm actually pretty agnostic about Vick. I definitely agree with you that he looks like a different player than he was in Atlanta, and I came away quite impressed with his work against Green Bay. I have no doubt that he's capable of looking incredible again this week against a Redskins' defense that could very easily turn out to be the league's worst this season. And even once he faces a tougher slate, it wouldn't surprise me if he still turns in respectable (or even some excellent) performances. But given the circumstances, I think it's awfully early to declare that Vick's proven himself to be a completely new and elite player.

#201 by chemical burn // Sep 29, 2010 - 12:44pm

I'm just saying GB didn't have an excuse of "we never expected to see Vick on the field!" In the way that most teams probably aren't game planning even in any capacity for Kellen Clemens (or Mark Burnell?) when they play the Jets. They knew he would see the field. They should have had a plan for what to do when Vick was on the field - even last year, he wasn't doing just gimmick plays. And he played plenty last year, so it's not just Reid talking - it's Reid talking combined with past evidence.

Also, I am classifying Detroit as mediocre according to FO's numbers. I could have also said "slightly below average." I agree they're underwhelming, but others have pointed out that according to FO's numbers, Vick has played a more difficult slate of defenses than Peyton Manning, who truly has gone up against cupcakes. There's got to be an acknowledgment of some kind of a scale between "sorta bad" and "truly putrid" if we're going to use weak opponents to disregard Vick's performance.

And I've never said "elite" in describing Vick. In fact, I think he'll probably end up ranked somewhere around 16 or 17 in passing DVOA if the Eagles are lucky, lucky, lucky. Just go back and look at his DVOA with the Falcons - he's always hovering around 37th in the league with a HARD negative DVOA. Also, watch how he played for those Falcons teams - he's a DIFFERENT player now. I'm not sure he has enough accuracy (and although we haven't seen problems with it yet, decision-making ability) to be an elite passer. But being an ok passer combined with what he brings to the running game (not just his own scrambles, but the effect it has on McCoy's production) make him a hugely valuable player.

If he can sustain it. If he can even be mediocre against a good defense. They jury is obviously out on those things.

#104 by TruFalcons (not verified) // Sep 28, 2010 - 4:36pm

Can one of the FO staff please explain why Lesean Mccoy's DVOA has gone from -4% to 47%?
Obviously Vick is benefitting from this massive, unprecedented improvement in the performance of Philly's running back.

#115 by t.d. // Sep 28, 2010 - 4:58pm

lol

#102 by onetime91 // Sep 28, 2010 - 4:31pm

This is brutal. Vick was wildly ineffective at times? Considering he had about 4 dropped passes, scored a rushing touchdown and had a 115+ passer rating anyway, I'd say you are out of your mind. Also, I like how you mention Henne's 27 yard pass interference but don't mention Vick's 25 yard pass interference call on his throw to Jeremy Maclin. Bias is an ugly thing.

#146 by bravehoptoad // Sep 28, 2010 - 6:56pm

On Philly's first four drives, Vick had the 61-yard touchdown pass to DeSean Jackson ... and otherwise went 2-of-8 for six yards with a sack. After an effective two-minute drill to get an 11-point lead, Vick led two possessions that combined for seven plays by going 2-of-4 for 22 yards and a sack, with both completions coming on third-and-long and ending up short of the sticks, leading to punts.

What part of that don't you understand? You do know what "at times" means, don't you?

Are the stats wrong that he's quoting? Otherwise, I have to say they're more convincing than "you are out of your mind."

#149 by bengal85 // Sep 28, 2010 - 7:07pm

I would say that we could pick any QB, including Rodgers, Manning, etc., and point out a couple unsuccessful drives in each game.
And casually saying in passing, "oh yeah, he did throw a 61 yarder to Jackson, and a 40 yarder to Maclin on a rope, and a 2 minute drill for a td, and for good measure turned a busted play in the redzone into a running td... BUT HE FOLLOWED EACH OF THOSE DRIVES WITH INCOMPLETIONS AND FAILED 3RD DOWN CONVERSIONS, SO CLEARLY HE'S NOT PLAYING WELL YADA YADA YADA..." I mean, aren't we kind of glossing over the lead? Kind of misssing the point, don't you think???

#210 by bravehoptoad // Sep 29, 2010 - 8:37pm

Yeah, I can see that argument: every QB is wildly ineffective during some stretch of any game. I'm not quite sure I buy it...here in Northern California it's been such a long time since I've gotten to see a good QB play regularly that I can't remember what it looks like.

That's a bit more intelligent than a few wild stats followed by "you're out of your mind," though.

#213 by onetime91 // Sep 30, 2010 - 7:09am

Yes, I do understand what "at times" means. He wasn't ineffective at any point in the football game. The incompletions he threw were more dropped passes than inaccurate throws.

#230 by horn // Sep 30, 2010 - 3:26pm

Yes, the part where he says Vick's completion ended up short of the stick, leading to punts. DeSean Jax had a wildly *clear* first down if he ran forward, he tried to outrun 11 guys to the corner when all he needed was 1 yd and he blew it. [or the part where the porous OL led to a sack on Vick somehow being a condemnation/negative for Vick.]

Most of us come to this site for the opposite of the meaningless box score stat-to-verbiage that was given in the OP about Vick. You can tell in various articles on the site that the game in question [or at least the part referenced] wasn't watched by the person writing it. Tanier did the play-by-play of Vick's performance, what was given here was a poor bastardization of that/or reading the box score.

There's been some anti-Vick bias at FO these past few weeks for sure. Frankly, I'dve kept McNabb for a year -- nobody asked me, but Vick has far and away been one of the top QBs in the NFL for 3 weeks -- and he did play the majority of the snaps v GB and started the game, in fact...their pass D is pretty damn good.

#101 by rk (not verified) // Sep 28, 2010 - 4:30pm

Didn't Collie start in place of Pierre Garcon on Sunday?

#97 by CoachDave // Sep 28, 2010 - 4:25pm

KUBIAK for me has been worth every penny for it's ability to identify horribly overvalued and/or undervalued players and to think situationally about player/team changes that can and likely will impact positional performance.

Knowing Martz was going to use Forte in the Marshall Faulk manner and thinking about Santana Moss with a "real" QB like McNabb vs. Campbell were instrumental in me getting steals in later rounds...both picks I attribute to thinking that was catalyzed from KUBIAK-based analysis. Yes, I had to execute the picks and of course get lucky in draft order, etc., but the initial thinking came from this site and KUBIAK.

Also, the positional comparisons I've also found helpful in determining when to draft certain positions. Knowing the differential performance distribution of TEs, RBs and WRs this year was extremely helpful when thinking about when to draft for what position. Watching people burn TE picks too early while I bolstered my bench with top-tier RBs and WRs was also rooted in this site, KUBIAK and PFP.

My first year using KUBIAK, I "slaved" to it and the results were terrible. But now I take a holistic (and likely more common sense) approach and I couldn't be happier with the results.

But that being said...can someone explain to me why their are so many posters who seem to be hell intent on doing nothing but commenting on why this site/statistics suck?

Has FO.com reached the heights of popularity where now the "haters" are an unfortunate by-product? This is one man's opinion, but this year these threads have gone from somewhat informative/substantiative to 60% drivel and quite frankly it's really curtailing my visits.

#111 by Eddo // Sep 28, 2010 - 4:51pm

This lines up pretty well with my three-year-running KUBIAK experience. It's interesting, KUBIAK is what led me to take a TE early in two of my drafts - the TE named Antonio Gates. KUBIAK suggested this year that there was Gates on tier one, then nearly every other TE somewhere between tiers two and three.

So far, it's paid off; Gates has been key in winning two close games for me.

#144 by bravehoptoad // Sep 28, 2010 - 6:52pm

I've been coming here since the site was about three months old, and the "haters" tend to come in waves. We'll have whole months without any good "I hate stats" threads. Usually I think we get a rush because we're linked on someone's site, they come and check FO out, and are horrified at how their team is evaluated here. Most often they leave because the site is proven correct (FOMBC and all), but occasionally we get converts who become interesting members. I haven't seen Chris (known as C since FO got a new site) around this year, but he was one of those.

Less often we get angry gamblers; they're usually flying solo, though. One odd guy stuck around for an entire season and ragged on the site relentlessly, driving me (at least) crazy with how he would say he knew all along xxx event would happen, but would never make predictions. He was the Common Sense advocate. Even he eventually burned out.

Without doubt, the people who hang around the longest are the ones who like the site and believe in it. They're the best reason to come here -- even moreso that the FO stats and articles, for me.

Conclusion: the guys ragging on the site will burn out, and the interesting posters will stick around. Stay and be one of them!

#151 by Eddo // Sep 28, 2010 - 7:18pm

"One odd guy stuck around for an entire season and ragged on the site relentlessly, driving me (at least) crazy with how he would say he knew all along xxx event would happen, but would never make predictions. He was the Common Sense advocate. Even he eventually burned out."

I remember him! He kept calling you "Mr. Hoptoad".

This is a great post, Mr. Hoptoad (:P). I think we get several kinds of louder dissenting commenters:

- People from team sites whose teams have recently gotten unfavorable ratings from FO; these people usually stay for less than a week, commenting in the thread that was linked on their forums, then getting bored of FO.

- Angry gamblers (good term). They tend to stick around a bit longer, but strangely (haha) never post anything they "knew would happen" before it actually happens.

- People who are generally skeptical of the numbers, but have some knowledge of statistics. These commenters can be quite persistent, and though they raise legitimate concerns, they're often a bit too pedantic and miss the forest for the trees, if you will.

- People who are intelligent about football, but are a bit too passionate about their fandom and/or judgement of individual players. They ultimately wind up contributing the most, since when they discuss teams and players they aren't passionate about, they're quite rational and knowledgeable.

- Raiderjoe.

#187 by BigCheese // Sep 29, 2010 - 2:27am

"- People who are intelligent about football, but are a bit too passionate about their fandom and/or judgement of individual players. They ultimately wind up contributing the most, since when they discuss teams and players they aren't passionate about, they're quite rational and knowledgeable.

- Raiderjoe."

Is the addition of copious ammounts of Sierra Nevada really enough to split them up into two distinct categories?

- Alvaro

#191 by Bowl Game Anomaly // Sep 29, 2010 - 6:25am

Yes. That, and the brilliantly misspelled posts. Raiderjoe is his own category.

#211 by bravehoptoad // Sep 29, 2010 - 8:39pm

Nice categories. Seeing Raiderjoe in his own category gives me a little pang.

We ought to gather a "best of" FO posters.

Pat -- who comes closest to earning the label "stat bigot" of anyone I've read, but who still wrote some of my favorite posts. Most famous on this site, probably, for inventing Robo Punter.

STEVEN YANG -- WHO LOVED TO WRITE LONG PREDICTIONS OF TEAM STRENGTH BASED ON THREE OR FOUR PLAYERS ON THE TEAM, ENDING WITH A SEVERAL SENTENCE SUMMARY, ALL IN CAPS.

Chris, or C -- famous for his passions about specific players: fanatical hatred of Tony Dungy (why???) and Campbell...I think he's winning that one. Also his love of Eli Manning.

zlionsfan -- inventor of the famous template

Several fans most notable for their intelligent fandom of one team -- Bobman our Colts guy; Will Allen our libertarian/Vikings guy, Karl Cuba our 49er guy over in Britain, etc.

Various lawyers who pop up on the legal Extra Points...can't recall one in particular.

Many cool stats guys, at least one of which I can recall being recruited to the Outsiders (Sean McCormick, if memory serves.)

Good site, isn't it?

P.S. That particular gambler...Rich C., wasn't it? Rick C? Something like that? Weird how he gave a big long teary farewell post.

#175 by greybeard // Sep 28, 2010 - 9:55pm

You have been here so long yet you still don't know DVAO and DYAR are not stats.

#209 by bravehoptoad // Sep 29, 2010 - 8:30pm

DVAO, no. DVAO is the Delaware Academy of Osseointegration, according to Google.

DVOA and DYAR are both strangely listed under "Our New Stats Explained" up at the top of the page. You might want to check it out. It's interesting reading.

#212 by greybeard // Sep 30, 2010 - 12:28am

Great, you caught me misspelling. Good for you.

You do not know how DVOA and DYAR calculated. You cannot reproduce numbers, you cannot modify it and find where it fails where it excels.

They both are interpretation of data, based on a lot of false assumptions, you can call them statistics if you want, but they are only statistics in the sense of interpretation of data. Someone arguing against them would not be "I hate stats" but would be arguing about how the data is interpreted.

DVOA is based on the assumption that one can calculate what a league wide average defense is, in a league where all teams do not play each other, where the performance of teams vary wildly from week to week.

It is based on a concept of success, such as 45% of yards on 1st down is success. How do we know success is not actually 4.2 yards, or 4.8 yards, or even sometimes 2 yards if it opens the way for play action. Furthermore, I bet most OCs would consider 4.5 yards on 1st and 10 success on a running play and not success on a passing play. These "stats" are based on very subjective way of interpreting data. There are many subjective interpretations of data that is not statistics.

If I have to call DVOA or DYAR anything, I would call them metrics. Just like simple rating system, DVOA is one of the metrics that give you insight on what happened, because you cannot watch every single game. You on the other hand, seem to believe it like a religion, and defend it like it is science.

It is funny that your proof of DVOA and DYAR are stats is that "FO says so". This is like an evangelical arguing that universe must be created in six days and then using bible as his proof.

#221 by commissionerleaf // Sep 30, 2010 - 12:44pm

What, pray tell, is the difference between a "statistic" and a "metric".

#223 by dryheat // Sep 30, 2010 - 12:53pm

a metric is a tool to measure something. a statistic is a tool to report something. not all statistics are necessarily metrics.

Having said that, it doesn't make the post any less confusing.

#224 by greybeard // Sep 30, 2010 - 1:36pm

Here is the bottom line. DVOA and DYAR are very subjective interpretation of data. People who suggests that they are broken or they have deficiencies are not people of kind "I hate stats". Conflating criticism of DVOA and DYAR with a general rejection of statistics shows a basic lack of understanding what they are, regardless of how one categorizes DVOA and DYAR.

WRT statistics and metrics, metrics is a measurement of something and you can have a metric of anything, like how attractive you find someone on the scale of 10, that does not make it a statistics.

#236 by commissionerleaf // Sep 30, 2010 - 5:03pm

Fair enough, but all metrics, and hence DVOA, are statistics.

#95 by Jim Glass (not verified) // Sep 28, 2010 - 4:20pm

I don't recall ever seeing three aborted snaps for one quarterback in one game

John Unitas fumbled four times in his first game with the Colts.

Some players get better, some don't.

#109 by billsfan // Sep 28, 2010 - 4:47pm

Didn't Olin Kreutz and Rex Grossman team up for multiple aborted snaps in a single game a couple of years ago?

(I also like the Eagles)

#122 by Lou // Sep 28, 2010 - 5:23pm

I know they fumbled at least twice in the same game on multiple occasions

#179 by Marko // Sep 28, 2010 - 10:36pm

That is correct. It definitely happened in more than just one game.

#92 by BJR // Sep 28, 2010 - 4:17pm

No comment on the dreadful performances of David Garrard and Carson Palmer? The struggles of these two previously highly regarded QBs are something I would like to read about.

#174 by cbirkemeier // Sep 28, 2010 - 9:50pm

Carson Palmer was saved by the Panthers inability to catch the ball. After the first quarter, he was 3-for-9 with 2 interceptions and at least 3 others that hit a Panthers player. I remember pointing out to a friend that he had hit a Panthers player with more than half his passes.

I would hold Palmer responsible for Ochocinco's horrible game, although Ochocinco did have one drop that was completely on him.

#87 by Jeff Fogle // Sep 28, 2010 - 4:07pm

This Week:
"Flacco had been derided for a rough start based on two road games against the Jets and Bengals, who are very likely to rate among the league's best pass defenses this year."

Yes, derided by YOU!

Last Week:
"The Bengals have a good secondary, but the Flacco that showed up this Sunday appeared to have mistaken them for the Jets once more. He had one of the worst first halves of football you will ever see a NFL quarterback have: 5-of-17 for 23 yards, with one first down and one interception. That's Derek Anderson on a day where even Ken Whisenhunt would wish for Matt Leinart. He put together one drive after the break, picking up two first downs and then finishing with a gorgeous 31-yard touch pass to Derrick Mason for a touchdown, but then it was back to more of the same. After that drive, he went 8-of-17 for 62 yards with two interceptions, another pick on a Hail Mary, a sack, and just three first downs. While the Bengals were able to get effective pressure on Flacco -- and T.J. Houshmandzadeh dropped a big pass downfield -- Flacco also missed open receivers, both with his throws and with his eyes."

Can we just agree that Flacco has shown pretty extreme tendencies...he's GREAT vs. crappy teams...but struggles vs. good teams, particularly late in close games? And that those struggles in close games haven't been "random" in a "guts and stomps" context, but are an actual reflection of his actual tendencies to have serious troubles vs. quality? And, therefore, DVOA'S ranking of Baltimore as the best team in the NFL last year was NUTS?

When Flacco gets the hang of what to do it in close games vs. quality, the Ravens could very well develop into that level of team in a heartbeat. They weren't there last year, even though you left the DVOA chart up on the home page all summer long showing that they were...by that metric anyway.

#136 by ec (not verified) // Sep 28, 2010 - 6:33pm

Thanks for writing that. I was thinking the same thing. Also, I'm not sure why he claims that Cinci and NYJ are two of the best pass defenses that Flacco will see this year, since Cinci is ranked 7th and the Jets 24th. Perhaps the Steelers are favored--something he's surprised about--since they are ranked 3rd. If Flacco looks worse against a superior pass defense, then maybe it doesn't matter that Batch's two long TD passes involved a little luck.

#83 by bengal85 // Sep 28, 2010 - 3:51pm

I love this site, and I think you guys are great. In Fact, I bought KUBIAK for the 1st time this year, and I am starting out the gate with the most consistently dominant team I've ever had so far. Praise all around.

BUT...

these statistics are not fool proof. As an example, let me point out the prediction on Adrian Peterson. Without getting too specific, as I know that is frowned upon, you all had him WAY undervalued, and specifically as a pass catching back. I adjusted my KUBIAK accordingly, and so far, it seems my projections are more accurate.

Likewise, I understand the value of charting a QB's relative performance, and I completely understand the value of looking at Vicks production through this prism of DYAR, but I have to say that I think to predict a regression to 2007 Vick production based on these paradigms is a mistake, for 2 reasons.

1. While you all are rightfully pointing out the woeful caliber of the pass defenses Vick has faced, you are not giving any credence to the fact that before week 2, Vick was taking 2nd team snaps, not preparing as a full time starter. Thus, it is reasonable to assume that his performance SHOULD improve as the season goes on and he and the offense gel better.

2. Using ATL Vick as a baseline to compare to is simply a faulty premise, because the situation in Philly is simply way, way better for a QB. Other than the guy being named the same and wearing the same jersey, nothing about the two situations is the same.
-He is changed in his preparation and practice methods/attitude, by all accounts. While I'm sure this will get snickers, I don't think it should, as this was a major factor in his mechanical inaccuracy in the past.
-Related to this 1st point, the coaching staff/ playcalling / protection schemes are going to be better. People forget Vick had basically a new coordinator every year since Dan Reeves was fired. NO QB is going to put good metrics in that setting.
-The Skill positions around him are better than all his years in ATL combined.

When you take all this together, I am skeptical that you all are calling for Vick becoming a retread of his past incarnation. He may not become a coach on the field like Peyton Manning, but I bet as a passer he approximates 75-80% of what McNaab did on a metrics based level, and when you throw in what he can do on the ground, I think he's going to end up the year as a top 5 QB, in real terms.

my two cents.

#126 by dbostedo // Sep 28, 2010 - 5:39pm

"As an example, let me point out the prediction on Adrian Peterson. Without getting too specific, as I know that is frowned upon, you all had him WAY undervalued, and specifically as a pass catching back."

For my PPR league, he came out near the top. What more do you want? What else did you change?

And no, KUBIAK will never be perfect. It will just likely be more correct ON AVERAGE than other predictions. It's simply probabilities of how well each player will do - not an absolute. Especially since they are largely guessing at how the actual usage numbers will work out.

#129 by Thomas_beardown // Sep 28, 2010 - 6:01pm

these statistics are not fool proof.

No has ever claimed this, or anything close to it.

#79 by WD (not verified) // Sep 28, 2010 - 3:46pm

"our article comparing JaMarcus Russell and Mark Sanchez's rookie seasons last year"

Too bad you didn't actually compare their rookie seasons. Way to re-write history, Bill!

#78 by bigtencrazy (not verified) // Sep 28, 2010 - 3:42pm

Nick Collins hands are so ridiculously bad it's ridiculous.

#73 by PTORaven // Sep 28, 2010 - 3:28pm

Not that Boldin's numbers weren't already impressive, but one of his three misses was a 55 yard bomb that pretty clearly should have been a PI call on the Browns

#68 by Carlos // Sep 28, 2010 - 3:24pm

How's that Garrard KUBIAK prediction working out?

I know I'm glad I loaded up on RB+WR and drafted Kolb late and Garrard in the last round! Starting Kolb in week 1 and Garrard in wks 2 and 3 have netted me... wait for it... exactly zero points thru 3 games.

#77 by Brik City (not verified) // Sep 28, 2010 - 3:39pm

Kubiak is a joke. The only stats you should use from this site are the team-based ones. Their team stats are very good. They have no idea how to separate the different facets of football into individual player performance statistics, though. This has been clear for years.

#84 by Lou // Sep 28, 2010 - 3:53pm

I've used Kubkiak for a few years now, and i find it generally pretty useful. I'm 3-0 in one league and 2-1 and 1st in scoring in the other. Its not perfect, but what is? It does seem a bit odd that it seems to have missed on value qbs this year. That was the thing i thought it was best at in the past. It turned me on to favre a few years ago when he was with the packers and everyone was down on him, rivers in his breakout year, and garrard in his. Of course i generally pick those guys to be a backup and then made a trade when they turned out good. This year i drafted Cassell and McNabb for that reason, and don't see much trade value yet.

#91 by Chris M (not verified) // Sep 28, 2010 - 4:12pm

They also had Orton as a top 10 guy this year, and that's worked out pretty well. Cassel's definitely been a disappointment, although he had a good fantasy game this week. Jury's out on McNabb.

#107 by billsfan // Sep 28, 2010 - 4:44pm

They freely admit that the Garrard projection should be taken with a grain of salt... something along the lines of "the only person who likes David Garrard more than KUBIAK is his mother."

KUBIAK sucks at predicting concussions, too.

(I also like the Eagles)

#169 by Basilicus // Sep 28, 2010 - 9:29pm

Something I like about this site is that the writers will freely disagree with their own statistics sometimes. If you're just skimming predictive statistics then, yeah, you're going to get screwed, which seems like the most obvious statement I've ever typed.

If you slow down and read the comments on the guys you're interested in, you'll find sometimes that a writer will say, "KUBIAK may've rated this player too high because of the following numbers that get plugged in," and, "KUBIAK isn't taking Factor X into account."

READ THE WRITING or you won't be able to put the numbers into proper context.

#128 by Thomas_beardown // Sep 28, 2010 - 5:57pm

Are you really blaming KUBIAK for 1) not predicting Kolb would get injured and 2) making you miss on your last pick in the draft?

#155 by Foosballs (not verified) // Sep 28, 2010 - 7:47pm

I don't think the point is "blaming" kubiak for anything. The point is garrard has been atrocious and kubiak had him as great value.

The comment bat concusions belongs on fox sports.

#160 by billsfan // Sep 28, 2010 - 8:34pm

There's sarcasm on the internet?

(I also like the Eagles)

#229 by horn // Sep 30, 2010 - 3:18pm

Not to defend Kubiak, but blaming it for a concussion Kolb got in h1 of Week 1 seems far-fetched. Perhaps you just suck at drafting.

#63 by Boots Day (not verified) // Sep 28, 2010 - 3:17pm

Is there a reason you don't just go ahead and credit a QB for yards gained on DPI, rather than keep apologizing for not including it?

#186 by Dave Bernreuther // Sep 29, 2010 - 2:24am

I can't speak for them, but I can say that I think it would be really unfair.

I've been seeing a LOT of DPI lately that reward terrible throws. It's far, far too easy for a receiver to slow down waiting for a badly underthrown ball, get run into by a DB who is a bit slower to recognize the trajectory, and collect a whole bunch of free yards.

On the one hand, the DB doesn't deserve credit for breaking up a pass on which he was clearly beaten. But on the other, the QB shouldn't get any credit for the fact that he underthrew an open receiver by ten yards.

Not all DPI are like this, of course.

#85 by Led // Sep 28, 2010 - 3:57pm

I'm guessing that it's too random and has proven not to improve the predictive value of the stat.

#61 by crack (not verified) // Sep 28, 2010 - 3:15pm

DYAR confuses me. D implies a D based adjustment, but a guy doing poorly in an important situation when the D would be keyed to what he is trying to do gives him a horrible result.

It's an oddly leveraged stat. How many replacement level RBs would be able to consistently get a yard from the 1?

#89 by Chris M (not verified) // Sep 28, 2010 - 4:10pm

They're adjusting for the quality of the defense. It's easier to get that yard against the Lions than the Ravens.

#123 by dbt // Sep 28, 2010 - 5:23pm

Should probably pick a better example, as the Lions DLine is pretty stout. Ask Lovie Smith. :)

#117 by dmb // Sep 28, 2010 - 5:09pm

DVOA (and DYAR) also adjust for down and distance, as well as game situation (field position, time remaining, score differential, etc.). I guess you could say that's part of the "A" in DVOA, or the "R" in DYAR -- they're taking a look at the outcome of each play in comparison to the average (or replacement-level) result of plays run in similar situations.

Chris M is correct, that the "defensive" adjustment refers to controlling for quality of opponent.

#197 by crack (not verified) // Sep 29, 2010 - 10:22am

It can't be replacement level. There is no way that not getting a TD from the 1 is worth -10 made up fake yards compared to replacement level.

What replacement level back, which would be approximately the 65th best back in the league, would be expected to make a yard in that situation with any regularity? Leverage index is overwhelming replacement level.

#198 by nat // Sep 29, 2010 - 10:36am

Think of it this way. Imagine a really mundane replacement back. He might get the ball into the end zone, for the sake of argument, one-seventh of the time. That's pretty bad. But it still averages to one point per attempt.

So each failed attempt to score from the 1 is worth -1 point compared to a replacement level back's per-play average. How many YAR-yards do you think equate to a point?

#199 by dryheat // Sep 29, 2010 - 11:14am

I'm not even sure it's relevant, since scoring a rushing touchdown from the 1 yard line is 99.99% on the offensive line. Behind a good offensive line, I'm willing to bet I score from the 1 at least 1/4 the time.

#56 by jmaron // Sep 28, 2010 - 3:10pm

I was really impressed with Rodgers last night. I really didn't expect him to be very good and thought he was overrated significantly last year. The reason I was negative was his tendency to hold onto the ball too long, but watching last night I saw several plays where he made great decisions to avoid a sack by either making a good move to get out of the pocket or just throwing the ball away.

In watching Rodgers I see I guy I would very rarely blitz. He gets the ball out as quick as anyone I've seen. I would do everything I could to keep in the pocket as well. He's an excellent runner and throws brilliantly on the run. The only weakness I see is when is trapped in the pocket and he doesn't see an open receiver. He just seems to slow down and almost freeze at times.

#62 by Dave Bernreuther // Sep 28, 2010 - 3:15pm

I actually thought he was a touch quick in throwing the ball away, as was Cutler. More Cutler, I think, but definitely Rodgers on the grounding (where, in my view, they called it based on the obviousness of him chucking it away, rather than on anything the receiver did or should've done... I agree about the route adjustment possibly being wrong, but Rodgers pretty obviously just chucked that one out of bounds.)

Better to be quick than slow, of course.

The only weakness I see is when is trapped in the pocket and he doesn't see an open receiver. He just seems to slow down and almost freeze at times.

He and Ben both do this sometimes. It's really odd how they can be so decisive sometimes and so indecisive others. Obviously it's a coverage thing... it's just funny that they're both so able to avoid sacks that 90% of other QBs would take, but then also take sacks that 75% of others would avoid. Rodgers seems to be improving in this regard, which makes him a serious weapon (as if he wasn't already).

#66 by jmaron // Sep 28, 2010 - 3:21pm

I think QB's like Roethlisberger and Rodgers just have trouble giving up on a play. I think Roethlisberger is overrated for that very reason. Thought the same of Culpepper. But it seems to me in some late games last year and the one game I've watched this year that Rodgers is evolving and getting rid of the ball. Certainly the sack rate plummeted in the second half of last year and I believe he's only been sacked 3 times this year.

That to me bodes very well for GB.

#54 by CraigInDC // Sep 28, 2010 - 3:09pm

Yes Bill, Charlie Batch all by himself is favored to beat Baltimore. Polamalu, Harrison, Woodley, Ward, and Mendenhall are all expected to stay on the sideline watching Batch cover Boldin, tackle Rice, and block Ngata. It should be an epic battle.

#82 by countertorque // Sep 28, 2010 - 3:50pm

#48 by Brik City (not verified) // Sep 28, 2010 - 3:04pm

Has DVOA for individual players shown any proven predictive capabilities? I know that team DVOA does, but has Vick's personal DVOA been proven to mean anything whatsoever? In my opinion, we should be looking at the Eagles offensive DVOA as a whole and judging Vick by that, since him drawing a spy or whatever solutions evidently render him useless immediately (weird that it hasnt happened yet) only serve to open up the rest of the offense.

#58 by shah8 (not verified) // Sep 28, 2010 - 3:12pm

Watch the TD run.

Spys don't work *that* well.

#113 by Brik City (not verified) // Sep 28, 2010 - 4:56pm

I was being sarcastic. I'm on your side here.

#46 by Sean McCormick // Sep 28, 2010 - 3:02pm

In Cutler's defense, the Packers were routinely sending heat while he was backpedaling out from under center in an empty set with only five-man protection. He had to move off his spot almost immediately on nearly every throw he made last night.

#39 by Eddo // Sep 28, 2010 - 2:49pm

Unrelated to this feature, I am sitting here watching some top-ten lists on NFL Network, and any time Aaron Schatz or Mike Tanier are shown, they're listed as a representative of "profootballoutsiders.com". Why is the "pro" in there? And why not just "Football Outsiders"?

#86 by Bobman // Sep 28, 2010 - 4:02pm

All part of their strategy to take over the world. Their subsidiaries NCAAfootball outsiders, HSfootball outsiders, Peeweefootball outsiders, Flagfootball outsiders, and of course Futbol outsiders have yet to be launched, but wait until they do.

I say inside ten years, they'll be running the world like the Stonecutters. I, for one, will sit back and await Steve Guttenberg's career resurgence.

#34 by Jonadan // Sep 28, 2010 - 2:45pm

I was going to start like this: "If DYAR continues to rate Detroit QBs over Brett Favre, I'm going to have to laugh. Last week was one thing, but this week?"

Then I reconsidered. Exactly what do we accomplish by trying out DYAR rankings across the leagues? I've read about the stat and everything, but it just seems to lead to some bizarre conclusions by using - if I understand it - an "average replacement", as it were, instead of the actual replacement.

Of course, my wanting to complain is tempered by the fact that this week's ranking more or less makes sense as a ranking no matter what it's based on.

#120 by johonny (not verified) // Sep 28, 2010 - 5:19pm

Don't read too much into the difference in such close one week ratings. The week to week variance of a player performance is greater than the difference between how the Lion QB and Favre played. However in no way are either playing well so far this year. Which is something pretty clear from the week to week performance tables. People seem transfixed on splitting hairs between 1 or 2 or the 27th and 28th best passer, rather than looking at the broader trend.

For the record Favre raw numbers say 61.9 %, 2 TD, 6 INT 6.2 YA
QB Detroit say 60.7 %, 3 TD, 5 INT 6.1 YA

Why would anyone be surprised to see the two grouped closely by a rating system?

#30 by Joseph // Sep 28, 2010 - 2:42pm

Does Moore's DYAR include his long punt return, or do you calculate DYAR for returns?

#27 by superbears // Sep 28, 2010 - 2:35pm

I really thought the Rodgers pass was intentional grounding, I will have to go back on the DVR but I believe there was pressure on him and he threw it away. I also agree that you could probably call holding on almost any play but I saw a lot of holding on the Packers line that wasn't called.

Finally to make this a total homer post, sure Cutler caught a few breaks and didn't look that great. I would like to looking at the charting data but I believe that is who he is, a QB who isn't consistant for a whole game but does catch on fire. Finally, sure they were a few missed interceptions but he is still cursed with having Devin Hester at WR who dropped at least two catches that I remember.

#33 by Eddo // Sep 28, 2010 - 2:45pm

Agreed on most counts. Peppers, in particular, clearly had the Packers offensive lineman intimidated.

Though I will defend Hester. Yes, he had the one terrible drop, but even ignoring the punt returns, he had a decent game. His route-running looks so much better than it ever has in the past; there was one play where he ran up about 10 yards, then made a ridiculous cut in and was wide open for a first down gain.

(As an aside, the most difficult part of seeing the game in person was having trouble telling Hester and Knox apart. They're both little, shifty guys, with a "3" as the ones digit of their numbers. So I apologize if I misidentify Knox as Hester or vice versa anywhere in this thread.)

#180 by Marko // Sep 28, 2010 - 10:39pm

You are not alone in having trouble telling Hester and Knox apart. Mike Tirico got them confused early in the game when he said a short pass was to Hester when it actually was to Knox. He never bothered to correct himself.

#25 by WeaponX (not verified) // Sep 28, 2010 - 2:33pm

QB Carolina should be credited with 1 less fumble (lost). QB Carolina put the football in Mike Goodson's belly, where somehow Goodson managed to completely "loose the handle" then fling the ball into the Bengals D. I'm not sure how that got credited to QB Carolina and it was certainly not a 3rd aborted snap.

#37 by Jonadan // Sep 28, 2010 - 2:48pm

Shouldn't that be "QB Panthers"? If we're getting tecmonichal, at least.

#24 by jhoff1 (not verified) // Sep 28, 2010 - 2:31pm

Now hold on a second. Yes Sanchez was pretty good on third downs. But its not like his overall accuracy was totally out of whack with his usual line. He threw 53%, his career line is 54 and for the season he's at 58. I'll admit to not being an expert on the typical dynamics of downs and accuracy. However, i'm pretty sure its inappropriate to suggest that his success this season is being driven by some luck about when his good tosses are coming in the series. As you point out, in his previous games he was at his normal level of third down conversion.
Bottom line: He's playing better than last year. And he's about to get another weapon. I'll take it. The jets can't be serious unless he's top 15, and it looks like he might be at that level.

#22 by t.d. // Sep 28, 2010 - 2:24pm

Calling Vick's performance against Green Bay 'mop up duty' is ridiculous. He was a fourth and one away from forcing overtime in that game. I also seem to recall reading somewhere that discounting late game performance in two touchdown spreads is inadvisable, because those plays are indicative of ability, too

#188 by dmstorm22 // Sep 29, 2010 - 3:38am

As shown by the Pats defensive performance in the second half of Week 1, that everyone chalked to "garbage time" but in reality was a harbinger of their problems for the last two weeks.

#14 by Jim Blob (not verified) // Sep 28, 2010 - 2:14pm

Cutler's game was underwhelming, in my opinion, but this seems like a stretch. The first interception on the final drive was on the play where Matthews pretty obviously roughed Cutler (leading with the crown of the helmet into the QB's facemask is pretty definitive).

******

Definitively is was actually Frank Zombo, not Matthews.

Emphasized

#17 by Eddo // Sep 28, 2010 - 2:16pm

Ah, my mistake. I misidentified the Packer.

#18 by TomC // Sep 28, 2010 - 2:22pm

There is no Clay Matthews, there is no A.J. Hawk, there is no Frank Zombo; there is only Ugly White Packers LB.

(And, in Eddo's defense, the ESPN team repeatedly assigned the penalty to Matthews, even doing an isolation replay on him, which showed him engaged in a block 10 yards away from Cutler.)

#32 by Eddo // Sep 28, 2010 - 2:42pm

Thanks for defending me, Tom, though it's really just my fault. I was there, so I wasn't misled by ESPN, and I just got my UWPLBs mixed up. :)

#59 by TomC // Sep 28, 2010 - 3:14pm

Lucky you. I drove by on my way home from work, and the tailgating was spilling all the way onto the grass bordering Lake Shore Drive. Must have been an awesome atmosphere.

One quick thing: Did I actually hear a "Kuuuuuuuuuhn" cheer when UWPFB carried the ball? Were there that many cheeseheads there? Me no likey.

#72 by Eddo // Sep 28, 2010 - 3:28pm

Yes, you did hear that correctly.

However, in my experience, this was probably the least-populated-by-Packer-fans game in this rivalry I can remember being to in my 10+ years of regularly going to games. (My dad would not shut up about this.) It seemed to us that there were less than half the Packer fans we expected.

#125 by TomC // Sep 28, 2010 - 5:30pm

Fair enough. Gotta give it to Packer fans, they do show up and support that team no matter where they're playing. (I could insert a gratuitous comment here about their probably being unemployed and not caring about their families which is why they have the free time and means to travel and buy tickets, but that would be beneath me.)

#134 by Eddo // Sep 28, 2010 - 6:31pm

Honestly, in my experience, the Packers don't get the highest visiting fans turnout at Bears games. I'd put it:

1. Eagles (really, and it's surprising to me)
2. Vikings
3. Packers
4. Steelers
5. Lions

#141 by thewedge // Sep 28, 2010 - 6:43pm

>1. Eagles (really, and it's surprising to me)

As a Philly fan who now resides in Chicago, this baffles me too. It's virtually impossibly to get tickets, and I rarely meet other Eagles fans around...maybe they live in the burbs or something.

#150 by Eddo // Sep 28, 2010 - 7:09pm

I wonder if the fact that the last two times the Eagles have been here were night games, which generally get a different crowd than standard day games. Those could be clouding my impressions a bit.

#200 by billsfan // Sep 29, 2010 - 12:31pm

McNabb is from Chicago.

(I also like the Eagles)

#57 by Dave Bernreuther // Sep 28, 2010 - 3:11pm

That was pretty funny. They just kept talking over it as if it was showing a completely different play, and never even acknowledged it until the actual rough was shown a few seconds later.

#12 by shah8 (not verified) // Sep 28, 2010 - 2:11pm

Oh, this is going to be entertaining. So, Vick is number 2 in passer rating and he still ranks a Chad Henne in DVOA. Also, an awful lot of confidence that he will have less success as "teams figure him out". "Mop up duty"? Then there's the fact that Desean Jackson isn't top 5 in DVOA. And of course, if these two characters pound the Hogs and take the fourth quarter off again, Vick will still be "worse than Chad Henne". Then the week after in San Fran. Of course, the three weeks after *that* would have doubters eating lots of crow, I suspect.

You know what?

I think the fundamental error in your thinking is that you guys aren't comprehending just how conservative this offense is, passing-wise. What Vick is doing right now is simply not easily stoppable, and they aren't showing much in the way of creative playcalls.

#20 by thewedge // Sep 28, 2010 - 2:22pm

>So, Vick is number 2 in passer rating and he still ranks a Chad Henne in DVOA.

But QB rating is a terrible stat, doesn't take into account fumbles, sacks, situation, opponent, or much of anything really.

#26 by SamWyatt (not verified) // Sep 28, 2010 - 2:34pm

I am not trying to be patronizing, but have you taken the time to understand how DVOA works? Or read the introductory explanation at the top of the page?

#31 by shah8 (not verified) // Sep 28, 2010 - 2:42pm

Sure, you can be patronizing as much as you want. It's not as if I'm not enjoying a feeling of superiority over people who mistake baroqueness for genuine and effective complexity.

I haven't bothered to learn the details.

1) Football is too noisy a game for statistics, period. All efforts at big-time statistics is simply an effort at mindwankery.

2) I routinely see losing quarterbacks way up high, and sheltered quarterbacks up there as well. There is routinely a bunch of players that are ranked too high and low in runningback stats and receiver stats. Of course, I realize these stats are pretty bad early in the season, but this is just really poor predictative efforts.

#42 by PatsFan // Sep 28, 2010 - 2:56pm

Hmmm. Perhaps a dose of the Ludovico treatment at the 2011 MIT SSAC is in order. :)

#45 by shah8 (not verified) // Sep 28, 2010 - 3:00pm

I do not support the torture of statisticians.

After all, who'd make a home for rescue mathheads?

#80 by Dean // Sep 28, 2010 - 3:47pm

Then why are you here?

#51 by thewedge // Sep 28, 2010 - 3:08pm

>Football is too noisy a game for statistics, period. All efforts at big-time statistics is simply an effort at mindwankery

Then why are you commenting at a site dedicated to statistical analysis of football?

>I routinely see losing quarterbacks way up high

It's entirely possible for a QB to outplay his counterpart on the field and still lose. Football is a team game. The designation "losing quarterback" seems meaningless.

>There is routinely a bunch of players that are ranked too high and low in runningback stats and receiver stats.

I'm unsure about what that statement means. If FO says "Player X is, according to our statistics, ranked #7 in the league", it's not valid to say "Player X is ranked too low" because Player X is ranked exactly correctly, according to the statistical measure in question. You could question the validity of the statistical system, but you should present good, and on this site preferably statistical, measures of why the statistical system is flawed. For instance, people above, and many other times, have commented that DVOA unduly rewards consistent gains over big plays, which may very well be a flaw in the system, or the comment questioning whether we should maybe look at the Eagles offensive DVOA as a whole, rather than simply Vick's individual DVOA are good topics of discussion. But simply saying "DeSean Jackson isn't in the top 5? this system is stupid" doesn't really get anyone anywhere.

#74 by SamWyatt (not verified) // Sep 28, 2010 - 3:31pm

Wow. Just, wow.

First, let us begin with how you didn't even bother to correctly read, or maybe you just did not comprehend, what I wrote before responding. I specifically told you I was not trying to be patronizing. I said it because enough frequent the site without bothering to see what DVOA actually means. It was a legitimate question. Just a little tip, but actually understanding what is said to you before respond goes quite a ways with credibility.

Second, admitting to us that you actually are petty, "enjoying a feeling of superiority over people," isn't going to make anyone want to actually take you seriously, or listen to you any more either.

Third, how does one set a line between what is 'baroque' and what would is 'genuine and effective' complexity? Seems arbitrary and self-serving.

Fourth, admitting that you are criticizing something that you haven't even taken the time to understand takes a pair made of brass, so kudos to you on that sir!

And lastly, coming to, and commenting on, the site while being actively close-minded about the entirety of its contents, while still using statistics such as passer rating to counter the statistic that you don't agree with, while stating that statistics such as those do not matter in professional football is the most impressive of all.

But thanks for the comment

#96 by shah8 (not verified) // Sep 28, 2010 - 4:24pm

Thing is...

"I'm not a racist/misogynist/etc, but" almost invariably leads to racist/misogynist/etc comments. In that sense, "Not to be patronizing, but" is taken in the exact same way, and I gave it all the due respect that deserves, which was a raspberry.

In any event, in the real world, there are tons of economists out there trying to sell these oh so complex models that, oh my goodness, tends to output mostly what people want to hear, same as any fortune-teller. Nobody in this busy world of ours has time to dig out the what the guts of the thing actually measures on the low-level. Especially when it comes to statistics, as in the end, statistics requires a pretty good grasp of episto philosophy to figure out what the assumptions are, what the desired question precisely is, and what answers actually answers that question. That's not that easy for an outsider to figure out, sometimes. It is much faster to simply eyeball the results and compare it to crude measures, like YPA/QB Ratings. Meteorologists, for example, will compare sophisticated hurricane models against someting called a Clipper model, which is just climatic info from previous hurricanes and not an actual model. DVOA is something that I actually compared with crude rankings. And crude rankings after Week 1 predicted Week 3 success better than DVOA. I also watched a bunch of games, and I can directly eyeball Henne to Vick to Rodgers to Cutler and get the sense that it's whack.

Is that arbitrary? Or good ole' skepticism.

As far as being in the right place? I sure am, if what matters is my own personal entertainment. It's not as if I've been verbally abusing people. This is supposed to be math, and not church as well!

#105 by Will Allen // Sep 28, 2010 - 4:38pm

I have some issues with how the individual stats on this site are labeled, with the caveat that using a very long descripion of a stat is a little unwieldy. I think the tems stats are much more useful. Do you not see some problems with your use of one week's results to draw conclusions about the predictive value of the team stats here?

#108 by shah8 (not verified) // Sep 28, 2010 - 4:46pm

Wait, did I ever talk about team stats? If I did, that wasn't intended.

I would not ever use any system for football, as it's far too calvinball for my tastes to try and predict stuff. Plus injuries, especially the ones people keep quiet about, skews everything.

I *do* think that towards the end of the season, that DVOA would have superior predictions compared to just eyeballing rough stats. It's just that there's a ton of noise early in the season, and some people are really inputting biases when they should be using their *humbled* common sense **cough**Vick**cough.

#124 by dbostedo // Sep 28, 2010 - 5:26pm

A couple of things :

1) "And crude rankings after Week 1 predicted Week 3 success better than DVOA."

Actually expanding on that and showing it with real numbers/information would be a great way to better make your point. And perhaps lead to a more intelligent discussion of how much value there is in early season DVOA.

2) This is a week by week ranking, not cumulative for the season, so I just wanted that to be clear. If you're using the above to compare "Henne to Vick to Rodgers to Cutler" it's for this week only. Some of your comments and others made it seem like this wasn't clear.

#162 by shah8 (not verified) // Sep 28, 2010 - 8:47pm

Isn't the D part of DVOA in effect cumulative?

#164 by Thomas_beardown // Sep 28, 2010 - 9:00pm

1) This is a list of DYAR, not DVOA.

2) Opponent adjustments are cumulative, but a players previous performance still has no effect on their week-to-week rankings. Look at the positional pages for that.

3) The D isn't in effect yet.

#166 by shah8 (not verified) // Sep 28, 2010 - 9:08pm

Alrighty then, how is average player benchmarked? Standard deviation over the median from week to week, or is that cumulative as well?

#168 by Thomas_beardown // Sep 28, 2010 - 9:19pm

Your question is a little confusing, but I'll try to answer what I think you are asking.

DYAR measures how successful a player is cumulatively, over a predefined "replacement-level player." This replacement-level player isn't actually a real player, but a baseline against which all players are measured.

So you take the success points earned on each play, subtract the baseline from each one, and add up all the results. Then you get this stat. On quick reads, it is simply the plays run that week. If you want to see every play ran for the year, check the positional ranking pages.

Everything is explained in far more detail here http://www.footballoutsiders.com/info/methods.

#177 by shah8 (not verified) // Sep 28, 2010 - 10:30pm

Thanks, but the parts that pertains to replacement value is not very transparent at all. I couldn't even judge whether replacement value is judged every week or from last year, or is it cumulative...

#182 by Thomas_beardown // Sep 28, 2010 - 10:49pm

Replacement value is just a baseline. So on 1st and 10 from the 50, if you run it's something like 3 yards. Anything over that you get is the value above replacement. FO doesn't make it crystal clear what the baseline is because that's how they make their money.

It is the same every week though.

#189 by Mountain Time … // Sep 29, 2010 - 3:38am

Pretty sure the baseline is calculated based on all of the DVOA data from 1993-2009.

#130 by SamWyatt (not verified) // Sep 28, 2010 - 6:13pm

So, you made a presumption, then when given a not only plausible, but actual reason for the statement, you stand by your presumption? And to top it off, you admit that I was actually correct when I asked the question?

Your second point is completely specious to what I was saying. In fact, DVOA isn't anywhere near as complicated a concept as you try to make it out to be.

And you missed what I was getting at with your last post. What I was saying is that life is too short to 'troll.' Constructive criticism is absolutely fantastic, and no model is perfect, many people here point out the legitimate flaws with the system, but that is not what you are doing, you're just throwing rocks.

Overall DVOA

Position

Line Stats

Other

Historical Lookups

Single Game DVOA & PGWE

Defense

Position

Stat Splits by

Team & Player Research

Betting

Week 3 Quick Reads

Comments

#216 by onetime91 // Sep 30, 2010 - 7:21am

#219 by nat // Sep 30, 2010 - 9:29am

#225 by onetime91 // Sep 30, 2010 - 2:43pm

#231 by nat // Sep 30, 2010 - 3:27pm

#233 by horn // Sep 30, 2010 - 3:35pm

#234 by nat // Sep 30, 2010 - 4:04pm

#235 by chemical burn // Sep 30, 2010 - 4:45pm

#237 by t.d. // Sep 30, 2010 - 6:51pm

#215 by onetime91 // Sep 30, 2010 - 7:19am

#217 by Thomas_beardown // Sep 30, 2010 - 7:28am

#218 by ammek // Sep 30, 2010 - 7:52am

#222 by commissionerleaf // Sep 30, 2010 - 12:50pm

#206 by Veloso // Sep 29, 2010 - 5:17pm

#190 by jeff (not verified) // Sep 29, 2010 - 4:28am

#173 by the K (not verified) // Sep 28, 2010 - 9:48pm

#165 by TruFalcons (not verified) // Sep 28, 2010 - 9:07pm

#167 by halfjumpsuit // Sep 28, 2010 - 9:10pm

#232 by horn // Sep 30, 2010 - 3:33pm

#157 by F.Leghorn (not verified) // Sep 28, 2010 - 8:19pm

#170 by chemical burn // Sep 28, 2010 - 9:34pm

#205 by tunesmith // Sep 29, 2010 - 3:47pm

#148 by bengal85 // Sep 28, 2010 - 7:01pm

#138 by Jim in VT (not verified) // Sep 28, 2010 - 6:41pm

#142 by Mountain Time … // Sep 28, 2010 - 6:45pm

#156 by Jerry // Sep 28, 2010 - 8:01pm

#158 by Rocco // Sep 28, 2010 - 8:20pm

#133 by 1618 (not verified) // Sep 28, 2010 - 6:25pm

#112 by onetime91 // Sep 28, 2010 - 4:55pm

#147 by bravehoptoad // Sep 28, 2010 - 7:01pm

#214 by onetime91 // Sep 30, 2010 - 7:12am

#152 by Mountain Time … // Sep 28, 2010 - 7:24pm

#153 by nat // Sep 28, 2010 - 7:32pm

#106 by nat // Sep 28, 2010 - 4:41pm

#121 by chemical burn // Sep 28, 2010 - 5:22pm

#127 by dmb // Sep 28, 2010 - 5:47pm

#154 by chemical burn // Sep 28, 2010 - 7:46pm

#178 by dmb // Sep 28, 2010 - 10:33pm

#201 by chemical burn // Sep 29, 2010 - 12:44pm

#104 by TruFalcons (not verified) // Sep 28, 2010 - 4:36pm

#115 by t.d. // Sep 28, 2010 - 4:58pm

#102 by onetime91 // Sep 28, 2010 - 4:31pm

#146 by bravehoptoad // Sep 28, 2010 - 6:56pm

#149 by bengal85 // Sep 28, 2010 - 7:07pm

#210 by bravehoptoad // Sep 29, 2010 - 8:37pm

#213 by onetime91 // Sep 30, 2010 - 7:09am

#230 by horn // Sep 30, 2010 - 3:26pm

#101 by rk (not verified) // Sep 28, 2010 - 4:30pm

#97 by CoachDave // Sep 28, 2010 - 4:25pm

#111 by Eddo // Sep 28, 2010 - 4:51pm

#144 by bravehoptoad // Sep 28, 2010 - 6:52pm

#151 by Eddo // Sep 28, 2010 - 7:18pm

#187 by BigCheese // Sep 29, 2010 - 2:27am

#191 by Bowl Game Anomaly // Sep 29, 2010 - 6:25am

#211 by bravehoptoad // Sep 29, 2010 - 8:39pm

#175 by greybeard // Sep 28, 2010 - 9:55pm

#209 by bravehoptoad // Sep 29, 2010 - 8:30pm

#212 by greybeard // Sep 30, 2010 - 12:28am

#221 by commissionerleaf // Sep 30, 2010 - 12:44pm

#223 by dryheat // Sep 30, 2010 - 12:53pm

#224 by greybeard // Sep 30, 2010 - 1:36pm

#236 by commissionerleaf // Sep 30, 2010 - 5:03pm

#95 by Jim Glass (not verified) // Sep 28, 2010 - 4:20pm

#109 by billsfan // Sep 28, 2010 - 4:47pm

#122 by Lou // Sep 28, 2010 - 5:23pm

#179 by Marko // Sep 28, 2010 - 10:36pm

#92 by BJR // Sep 28, 2010 - 4:17pm

#174 by cbirkemeier // Sep 28, 2010 - 9:50pm