Writers of Pro Football Prospectus 2008

Most Recent FO Features

PrescottDak15.jpg

» OFI: Blowouts, Upsets, and Narrow Escapes

The College Football Playoff field narrowed on Saturday. Some teams got upset, some barely escaped an upset, and a few had big record-setting blowouts.

11 Nov 2008

Week 10 DVOA Ratings

by Aaron Schatz

Parity is nothing new in the NFL. What's new in 2008 is that the parity doesn't revolve around .500.

Right now, 16 different teams are either 6-3 or 5-4. That's half the league! With so many teams hovering around "kinda good," only three teams are better than 6-3, and only three teams are 4-5. By comparison, last year at this point nine different teams were 4-5.

To show you the difference in 2008, I put together this neat little graph, showing how many teams had X number of wins in each of the last five seasons. Check out how much bigger the green bars (5-4 and 6-3) are in 2008.

With so many teams hovering around 5-4 and 6-3, the Football Outsiders DVOA ratings end up looking a lot different from the standings. The top three remain the same as last week. Sure, it seems a bit strange to have the Eagles second at 5-4, but on Sunday night they looked almost as good as the Giants -- and that's what being ranked second in DVOA makes them, almost as good as the Giants. The Titans still rank third. They have not lost a game, but they've played an easy schedule and four of their wins have come by one score or less. Unlike undefeated midseason teams of the past (2007 Patriots, the 1998 Broncos, 1985 Bears, etc.) the Titans have not yet had a dominating win over a good opponent. A big reason why the Titans are undefeated without blowing the league away is that they are extremely consistent. Right now the Titans rank first in VARIANCE.

Below the top three, things begin to diverge from conventional wisdom. The Packers may only be 4-5, but they rank seventh in DVOA, basically tied with the 7-2 Panthers. (They lost to the Vikings, and moved up three spots anyway because the system thinks they outplayed Minnesota by a small amount.) The Patriots may be 6-3, thanks in part to an easy schedule, but they're all the way down at 20th -- although as of this week they are no longer the lowest-ranked AFC East team. That honor goes to Buffalo, one of three teams that rank below 20th despite a winning record.

What do these topsy-turvy rankings mean in the long run? One thing: get ready for upsets. It is very likely that some lower-ranked teams will make the postseason. Wild card teams are likely to be underrated because their win-loss records do not match their DVOA ratings. According to our DVOA playoff odds report, the Titans now have an 88 percent chance of winning the number one seed in the AFC -- yet according to the DVOA ratings, they are only slightly better than the generally ignored 6-3 Ravens, a team with similar strengths and weaknesses.

By the way, the playoff odds report now gives the chances of the Titans going undefeated (4.8 percent) and the Lions going winless (9.2 percent).

* * * * *

The addition of the "undefeated watch" is just one of the new items at Football Outsiders this week. Even more important is the expansion of our weekly defense vs. receivers table.

We know people use the defense vs. receivers table to help with fantasy matchups, and our own Bill Barnwell uses it each week in his ESPN fantasy matchups column. The problem? DVOA wasn't created to measure fantasy football value. The DVOA ratings measure value per play without telling us if a defense forces its opponents to alter their offensive game plan, avoiding or favoring certain receivers. They also measure certain defensive plays that don't mean anything from a fantasy football perspective, like interceptions.

Starting this week, the defense vs. receivers table will also list average passes per game and average yards allowed per game against the five "receiver types." These numbers are adjusted for opponent strength. The Cardinals throw to their number-two receiver (Anquan Boldin) 34 percent more often than the average offense, and he gains an average of 1.8 yards per play more than the average number-two receiver. Therefore, if the Cardinals throw to Boldin eight times for 60 yards, it won't count as much against the defense as if would if that same defense played the Giants and Eli Manning threw to Amani Toomer eight times for 60 yards.

With the new numbers, you can see certain places where a good DVOA does not mean that a defense shuts down a wide receiver for fantasy purposes A good example is Chicago, which ranks 13th in DVOA against number-one receivers. Most Tampa-2 style defenses tend to face more passes to the opposition's top receiver, and the Bears are a clear example. Their adjusted number is 12.3 passes per game -- second to Washington -- which ends up with 80 (adjusted) receiving yards allowed per game, fifth in the NFL. It turns out the Bears are a good matchup for your receiver, even if that receiver is going to be the target on a good number of incomplete passes.

Here are some interesting tidbits from the first run of this table (all numbers are adjusted for opposing offenses):

  • Even though Washington faces more passes to WR1 than any other defense, the Redskins allow just 50 passing yards per game, below the NFL average. Only Tampa Bay and Carolina face fewer passes per game to the opposing WR2.
  • Minnesota, Philadelphia, and New Orleans all rank in the top six in DVOA against WR2 (yes, the Saints, really, not a misprint) but also all rank in the top six in passes per game to WR2.
  • On the flip side of that, St. Louis has the worst DVOA against WR3 but faces just 5.1 passes per game to "other" wide receivers, well below the NFL average of 7.3.
  • In general, the best teams face the most passes to WR3 because of opponents trying to come back late in games -- but Jacksonville is actually one of four teams that faces more than 10 passes per game to WR3. (The others are the Giants, Bears, and Steelers.)
  • When do you sit your studs? The defense allowing the fewest yards per game to opposing WR1 is Philadelphia, followed by Green Bay and -- a real surprise -- Houston.
  • Keep your tight ends away from the 49ers, who give up just 18 yards per game to tight ends. Play them against the Chargers, who give up a league-leading 83 yards per game to tight ends. They face an average of 11.2 passes per game to tight ends, two more than any other defense.
  • Cleveland may have a poor defense this season, but the Browns excel in one area: preventing success by running backs in the passing game. The Browns rank number one in DVOA against running backs, passes per game allowed to running backs, and receiving yards per game allowed to running backs.

Do note that these defense vs. receivers numbers count pass interference, and the totals for "other wide receivers," "tight ends," and "running backs" include multiple players for each offense. (For example, that number gives you yards per game allowed to all tight ends, not just the other team's starting tight end.) These numbers also count Dallas Clark as a WR3, although I may change him to a tight end in future weeks now that we're running these numbers as well as DVOA.

* * * * *

The other piece of news this week regards the Football Outsiders Premium section. Beginning this week, we're offering a "Half Season Special" subscription to Football Outsiders Premium. For $25, or just over half the normal subscription fee, you will get access to Premium through February 1, 2009, the day of Super Bowl XLIII.

If you've been wavering on subscribing to Premium, this is your chance to check out the DVOA Matchup view as the games become more important to who makes the postseason. This is your chance to get instant advice as your fantasy league enters the playoffs. This is your chance to see how successful our new picks against the spread will be during the second half of the year, now that we have more information to go on. So far this year, our picks against the spread are 78-61-5, and the "green" picks -- our best bets -- are 8-3.

(If you only recently bought Premium and you're now getting buyer's remorse, don't worry -- your standard subscription still goes all the way through the middle of next season, and it is still a better value when it comes to getting bang for your buck.)

The new Premium "Half Season Special" is the first item to be added since we moved our store over from our old server to our new one this weekend. This move should solve the problem we were having with certain subscribers not being able to access Premium. It's important to know that the way to access products in the store has changed a little bit. You'll need to log in to the website, then look for the "My Files" link in the left-hand column. If you've been accessing Football Outsiders Premium or other products with an e-mail link from the old store, that won't work anymore.

* * * * *

These are the Football Outsiders team efficiency ratings through ten weeks of 2008, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)

OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. These ratings also include opponent adjustments. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. WEIGHTED DVOA is adjusted so that earlier games in the season become gradually less important. It better reflects how the team is playing right now.

As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.

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

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

All stats pages should be updated in the next few minutes. Remember that as of this week, opponent adjustments are at full strength.


TEAM TOTAL
DVOA
LAST
WEEK
WEIGHTED
DVOA
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 NYG 39.0% 1 38.4% 1 8-1 23.9% 2 -10.7% 8 4.3% 4
2 PHI 33.5% 2 34.4% 2 5-4 18.1% 7 -14.6% 6 0.8% 16
3 TEN 28.9% 3 30.2% 3 9-0 8.2% 12 -21.7% 3 -1.0% 22
4 BAL 26.4% 5 26.3% 4 6-3 5.3% 15 -22.5% 2 -1.3% 23
5 ARI 18.7% 4 17.0% 6 6-3 25.1% 1 3.1% 15 -3.3% 27
6 PIT 17.0% 6 16.2% 8 6-3 -7.7% 26 -24.2% 1 0.5% 18
7 GB 16.5% 10 16.6% 7 4-5 7.5% 13 -5.9% 9 3.0% 8
8 CAR 16.5% 7 17.8% 5 7-2 2.3% 20 -12.0% 7 2.1% 12
9 TB 14.7% 9 13.0% 11 6-3 -1.3% 21 -16.6% 5 -0.6% 20
10 ATL 14.5% 13 16.1% 9 6-3 18.7% 4 7.0% 18 2.7% 9
11 IND 13.0% 14 14.5% 10 5-4 18.3% 6 4.0% 16 -1.4% 24
12 WAS 12.6% 8 12.2% 12 6-3 14.9% 10 -1.2% 12 -3.5% 28
13 NYJ 10.7% 18 11.5% 13 6-3 3.0% 18 -5.4% 11 2.4% 10
14 CHI 9.5% 11 8.1% 15 5-4 4.9% 16 -5.6% 10 -0.9% 21
15 MIA 7.7% 12 10.1% 14 5-4 15.3% 9 -0.5% 13 -8.2% 31
16 MIN 1.9% 16 1.3% 16 5-4 -4.1% 24 -17.9% 4 -11.9% 32
TEAM TOTAL
DVOA
LAST
WEEK
WEIGHTED
DVOA
RANK W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
17 NO 0.7% 15 0.4% 17 4-5 20.0% 3 19.1% 26 -0.1% 19
18 SD -0.4% 17 -2.0% 18 4-5 18.5% 5 21.1% 29 2.3% 11
19 JAC -3.5% 21 -3.2% 19 4-5 8.3% 11 14.9% 24 3.0% 7
20 NE -5.2% 23 -6.5% 20 6-3 2.6% 19 11.0% 21 3.2% 6
21 DEN -6.5% 20 -8.2% 21 5-4 15.8% 8 20.3% 27 -2.0% 25
22 BUF -7.7% 19 -10.7% 23 5-4 -2.7% 22 9.3% 20 4.4% 3
23 DAL -11.2% 24 -16.1% 27 5-4 3.8% 17 8.6% 19 -6.4% 30
24 CLE -11.3% 25 -9.6% 22 3-6 -3.5% 23 16.6% 25 8.8% 1
25 SF -14.2% 26 -14.1% 26 2-7 -18.6% 29 1.8% 14 6.3% 2
26 HOU -14.9% 22 -14.0% 25 3-6 6.5% 14 25.1% 30 3.7% 5
27 SEA -16.7% 27 -13.8% 24 2-7 -4.6% 25 13.5% 22 1.3% 13
28 OAK -31.0% 28 -31.8% 28 2-7 -27.5% 32 4.4% 17 1.0% 15
29 KC -34.9% 32 -32.5% 29 1-8 -8.8% 27 21.0% 28 -5.2% 29
30 CIN -36.1% 30 -35.9% 30 1-8 -18.9% 30 13.9% 23 -3.2% 26
31 DET -42.7% 29 -42.7% 31 0-9 -13.5% 28 30.3% 32 1.1% 14
32 STL -48.0% 31 -46.3% 32 2-7 -21.9% 31 26.8% 31 0.7% 17

  • NON-ADJUSTED TOTAL DVOA gives performance without adjustments for schedule strength, fumble recovery luck, and weather/altitude on special teams.
  • ESTIMATED WINS uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close.  It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles.  Teams that have had their bye week are projected as if they had played one game per week.
  • PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • VARIANCE measures the statistical variance of the team's weekly DVOA performance.  Teams are ranked from most consistent (#1, lowest variance) to least consistent (#32, highest variance).


TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR RANK
1 NYG 39.0% 8-1 48.7% 6.9 1 -7.4% 30 16.4% 1 23.6% 28
2 PHI 33.5% 5-4 34.9% 6.6 3 0.2% 18 6.4% 8 10.5% 10
3 TEN 28.9% 9-0 29.4% 6.7 2 -2.2% 23 -5.3% 24 4.1% 1
4 BAL 26.4% 6-3 30.1% 6.4 4 -3.8% 24 8.5% 6 19.0% 20
5 ARI 18.7% 6-3 21.8% 5.6 8 -4.8% 25 -2.0% 18 12.3% 11
6 PIT 17.0% 6-3 16.3% 5.8 5 5.9% 5 -1.5% 17 4.8% 2
7 GB 16.5% 4-5 11.4% 5.5 9 0.4% 17 -4.2% 22 8.5% 8
8 CAR 16.5% 7-2 12.5% 5.7 6 -0.6% 20 6.0% 9 21.4% 24
9 TB 14.7% 6-3 19.2% 5.1 13 -1.2% 21 -6.8% 27 19.6% 21
10 ATL 14.5% 6-3 14.7% 5.2 12 -1.7% 22 -3.5% 21 23.3% 27
11 IND 13.0% 5-4 6.3% 5.7 7 7.7% 2 -13.3% 30 22.3% 25
12 WAS 12.6% 6-3 13.0% 5.4 11 -0.4% 19 3.5% 12 7.0% 5
13 NYJ 10.7% 6-3 22.2% 5.1 14 -13.7% 32 -2.3% 20 26.5% 30
14 CHI 9.5% 5-4 5.8% 5.5 10 3.8% 9 -5.1% 23 5.1% 3
15 MIA 7.7% 5-4 16.4% 4.9 16 0.5% 16 -21.7% 32 17.0% 17
16 MIN 1.9% 5-4 -2.2% 4.5 18 4.4% 6 8.4% 7 7.1% 7
TEAM TOTAL
DVOA
W-L NON-ADJ
TOT VOA
ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR RANK
17 NO 0.7% 4-5 3.4% 4.9 15 0.8% 15 -1.0% 16 6.8% 4
18 SD -0.4% 4-5 5.2% 4.8 17 -5.0% 26 -2.2% 19 13.3% 12
19 JAC -3.5% 4-5 4.7% 4.5 19 -6.0% 27 13.6% 4 7.0% 6
20 NE -5.2% 6-3 2.9% 4.4 20 -8.0% 31 -0.2% 14 23.3% 26
21 DEN -6.5% 5-4 -1.9% 3.8 23 -6.3% 28 -5.4% 25 17.5% 18
22 BUF -7.7% 5-4 -3.1% 3.9 22 -6.8% 29 -8.9% 28 15.3% 15
23 DAL -11.2% 5-4 -10.8% 3.5 24 4.0% 8 16.3% 2 27.2% 31
24 CLE -11.3% 3-6 -15.6% 4.0 21 6.4% 4 5.6% 10 20.5% 22
25 SF -14.2% 2-7 -18.4% 3.4 25 2.9% 11 -14.0% 31 8.7% 9
26 HOU -14.9% 3-6 -17.8% 2.9 27 1.3% 14 3.7% 11 14.1% 14
27 SEA -16.7% 2-7 -19.2% 3.4 26 2.7% 12 -0.6% 15 18.1% 19
28 OAK -31.0% 2-7 -33.7% 2.3 28 1.9% 13 -6.6% 26 29.7% 32
29 KC -34.9% 1-8 -32.2% 2.2 29 4.2% 7 -12.2% 29 20.7% 23
30 CIN -36.1% 1-8 -43.4% 1.7 31 8.1% 1 9.4% 5 13.5% 13
31 DET -42.7% 0-9 -42.4% 1.8 30 3.2% 10 15.4% 3 16.4% 16
32 STL -48.0% 2-7 -52.9% 1.2 32 7.4% 3 0.9% 13 24.8% 29

Posted by: Aaron Schatz on 11 Nov 2008

153 comments, Last at 17 Nov 2008, 12:32pm by Mig

Comments

1
by panthersnbraves :: Tue, 11/11/2008 - 5:31pm

8-10 and 15 means a fight to the finish, right?

3
by Dales :: Tue, 11/11/2008 - 5:48pm

"but on Sunday night they looked almost as good as the Giants"

Actually, no, they didn't. They looked extremely fortunate to not be getting blown out, needing the Giants to be very sloppy with fumbles and with untimely penalties to keep the score close.

ETA- and going in to the game, I thought the Eagles were probably nearly as good, if not better, and expected them to win. However, they did not look (IMO) anywhere near "almost as good" on Sunday night.

5
by joon :: Tue, 11/11/2008 - 5:58pm

well, sort of. the giants were definitely better in terms of yards and first downs, but the eagles did much better on kickoffs, performed better in the red zone, and fumbled fewer times (3 to 1, even though both teams lost 1 fumble). i know subjectively it felt like philly was lucky to be close because it seemed like the giants were getting 6+ yards every carry, but they did play pretty well against an obviously elite team.

21
by Dales :: Tue, 11/11/2008 - 6:48pm

"but the eagles did much better on kickoffs"

And the Giants did better in the punt game. Overall, despite the kickoff return yards advantage, the Eagles still were outgained by the Giants-- and the only reason it was close was because of a rather untimely hold by the Giants on the Hixon TD that was called back. In fact, I specifically mentioned how the Eagles benefited from very untimely (from the Giants perspective) penalties for this very reason.

"performed better in the red zone"

Both had four touchdowns after making it to the red zone.

The Giants got to the red zone two more times, and came away with 6 more red zone points than the Eagles did.

"and fumbled fewer times"

True. So basically you are taking what I said, which was that the Eagles were fortunate that the Giants were being sloppy and keeping them in it, and using that to say that the Eagles were in it, if I am following you.

"i know subjectively it felt like philly was lucky to be close because it seemed like the giants were getting 6+ yards every carry"

Objectively, they were lucky to be close because the Giants were getting five yards a carry (ETA- while carrying a lot). They were lucky to be close because the Giants were converting third downs at nearly twice the rate. They were lucky to be close because Kevin Boss dropped a touchdown.

The Eagles did well to keep the game close, and to actually be ahead for a while, but that was mostly because of Giants' problems than great play by the Eagles, excepting for the opening Giants drive and the last touchdown drive by the Eagles.

I went in to the game thinking the two teams were very close, and with it being at home I thought the Eagles would win. What I saw was a Giants team handling the Eagles with relative ease, but failing to put them away through their own failings rather than exceptional play by the Eagles. It was a game that was not as close as the score. All in my opinion.

23
by Scott P. (not verified) :: Tue, 11/11/2008 - 6:57pm

The Eagles did well to keep the game close, and to actually be ahead for a while, but that was mostly because of Giants' problems than great play by the Eagles, excepting for the opening Giants drive and the last touchdown drive by the Eagles.

I don't understand the arguments. You're right the Eagles weren't great, but you are also correct the the Giants had lots of problems. Those two things suggest the teams are fairly close. The Giants don't get to discount their problems.

26
by Dales :: Tue, 11/11/2008 - 7:22pm

What I am saying is that the Giants lost 5 points more due to problems the Eagles did not force, and that is why the Eagles were fortunate to be so close.

I look at it this way- the Eli interception and the fumbled kickoff pretty much offset; both led to touchdowns for the other team. The Jacobs fumble and the McNabb interception come out to the Eagles advantage by four points, and deservedly so. While Jacobs made a bad decision in being a Flying Wallenda, it was a great hit by the Eagles to force the fumble.

The 5 points the Giants lost through their own unforced errors were 4 points when Boss dropped a touchdown, and 1 point from them going for two which they would not had done had they gotten the TD on the Boss play.

In other words, the game was much closer to being a 10 point Giants win than to being a tie. Again, all just my opinion.

35
by phillyangst :: Tue, 11/11/2008 - 8:11pm

All of what Dales has pointed out plus the "Magic Right Achilles" play favoring the Jints and DVOA [still] loves Philadelphia!!!

40
by Dales :: Tue, 11/11/2008 - 8:30pm

"Magic Right Achilles"

Awesome. I'd go with Achilles Last Stand though. :-D

47
by MTR (not verified) :: Tue, 11/11/2008 - 8:39pm

I believe the DVOA response would be that those unforced errors count quite a lot. Those are exactly the type of things bad teams do often and good teams do rarely.

54
by Dales :: Tue, 11/11/2008 - 9:18pm

And I believe the response to that is that the Eagles (ETA- recent, three and a half year) ability to underperform DVOA is becoming difficult to dismiss as a fluke. DVOA apparently does not understand that while teams similar to the Eagles will score touchdowns x percent of the times in the two minute drill, the Eagles won't, and that this is repeatable.

Anyhow, that game Sunday was a perfect example of a game that wasn't as close as the score and wasn't as close as DVOA seems to think it was. If those two teams played 100 games with each team performing at the same basic level of performance as they did on Sunday, the Giants would win a substantial majority of them.

60
by Joseph :: Tue, 11/11/2008 - 10:03pm

I will submit (as someone who doesn't follow the Eagles) this hypothesis for DVOA's love of Philly:
As many Eagles fans note, they seem to regularly fail at 3rd & short in the red zone, esp. when running. They also have bad clock management, which leads to longer FG's at the end of halves. Based on commenters here, I also get the impression that they seem to mess up key plays in the game. As I don't follow the Eagles, I will not assign blame for these regular errors that definitely affect wins and losses.
My hypothesis is: DVOA notices 6 to 8 successful plays in the same drive that move the chains and result in a 50 yd. drive. This puts them in FG range, which Akers converts. Fans notice that WR X dropped the 3rd & 5 pass that would have given them a first down in the red zone (or the failed 3rd down plunge into the line, route run 2 yds. short of the marker, etc). DVOA thinks that the Eagles are a regularly successful team that messes up 15-20% of the time. Fans see that they make the same mistakes game in and game out, and are frustrated at a team that could be so much better. The computer sees a team that makes big plays for TD's and says, "They can't keep that up. One successful play won't result in a TD every time. A team like the Eagles is much better--they consistently make good plays." Eagles fans know that those successful plays too often end up short of the end zone, and wonder why a team that is so successful between the 20's can't punch it in a little more often.

Just my hypothesis. I would love to see the Eagles Red Zone Offensive DVOA, Power Run %, and other things that Eagles fans criticize the team for.

61
by Josh :: Tue, 11/11/2008 - 11:10pm

You can find that in the Premium Stats section, I think.

81
by Alex51 :: Wed, 11/12/2008 - 5:51am

the Eagles (ETA- recent, three and a half year) ability to underperform DVOA is becoming difficult to dismiss as a fluke.

Really? That's funny, because their 3 year streak of overperforming DVOA from 2002-2004 seems like a bit of a fluke in retrospect. So, three and a half years can't be dismissed as a fluke, but three years can? That extra half year must mean an awful lot.

Personally, I think they're both flukes. Over the entire Andy Reid era, the Eagles have gained 96.6 estimated wins, while winning 93 actual games. That's an error of less than 4%. Sorry, but a 3.6 game difference over nine and a half years is not proof of some missing variable in DVOA. It's not even an outlier.

89
by billsfan :: Wed, 11/12/2008 - 10:25am

The Eagles are clearly ranked too high because they lost too many close games to better teams that weren't actually as close as a subjective measure like the score would seem to indicate. The New York Football Giants are way better than this. Andy Reid couldn't coach his way out of a cardboard box, even with an extra timeout. Lucky plays are only lucky for teams not named The New York Football Giants. See also Catch, Helmet.

101
by Andrew B :: Wed, 11/12/2008 - 12:55pm

Dales:

Did we watch the same game?

Eagles Total Yards - 541 Yards
Giants Total Yards - 567 Yards

Eagles Redzone - 4 Touchdowns, 1 Field Goal
Giants Redzone - 4 Touchdowns, 3 Field Goals

Eagles - 3 forced fumbles, 1 interception on defense
Giants - 1 forced fumble, 1 interception on defense

Eagles - 5 tackles for loss, 1 sack, 3 Quarterback hits
Giants - 2 tackles for loss, 0 sacks, 2 Quarterback hits

Eagles Drives over 4 Plays - 3 Touchdowns, 1 Field Goal, 1 Downs
Giants Drives over 4 Plays - 4 Touchdowns, 1 Field Goal, 1 Punt

Eagles Drives 4 plays or less - 1 Touchdown, 5 Punts, 1 Interception, 1 Fumble
Giants Drives 4 plays or less - 2 Field Goals, 2 Punts, 1 Interception, 1 Fumble, 1 End of Game

The last lines emphasize the actual difference in the game. Both teams had 5 three and out drives. But the Giants got 6 points off of two drives that went three and out because they were in the Field Goal range when they got the ball. The Eagles went three and out elsewhere on the field and could not score.

If the officials had ruled initially that Jacobs fumbled on the goal line and given the ball to the Eagles, who had recovered it, we'd be having a different conversation, wouldn't we? The replay wasn't conclusive enough to show anything either way, so the initial ruling was what mattered. Similarly, if Stewart Bradley had not gotten a facemask penalty for clotheslining Jacobs and tackling him by his shoulder pad the drive before, we might be having a different conversation too. Of course sometimes you have to overcome calls to win a close game, but sometimes also, you win a close game on calls.

There aren't any statistics that will show this game as a domination by the Giants because it was not.

The Original Andrew

112
by Kurt :: Wed, 11/12/2008 - 2:44pm

Including kickoff returns in Total Yards is certainly some creative accounting. Kudos. I certainly hope that every future Giants opponent gets to rack up a couple hundred yards by virtue of returning eight kickoffs.

127
by Andrew B :: Wed, 11/12/2008 - 6:23pm

The two teams had an equal number of possessions. I was looking at total yardage up and down the field from the time the team took possession to the time when they either scored or gave up possession.

The Giants had 130+ fewer kick return yards on just 2 fewer kick returns. At 8 points per 100 yards, that's the equivalent of giving up 10 points of field position.

The teams were even in passing and interception return yards.

Essentially, the Eagles made up for their lesser performance in rushing and punt returns with a superior performance in kick returns, giving them consistently better field position after a kickoff. Which is also why, when they had the ball with two minutes left in the game, they were only 57 yards of offense away from winning the game despite the Giants supposedly "dominating" them.

Say the Eagles had scored on their final drive with minimal time left on the clock. The Eagles would then be 2 points and about 25 yards ahead on total combined yards from scrimmage, instead of being 6 points and 26 yards behind.

The Original Andrew

117
by Quincy :: Wed, 11/12/2008 - 4:06pm

I agree with Kurt that you've done some great cherrypicking. One could make an equally valid argument that the following stats suggest the game wasn't as close as the score:

Total Yards NYG 401 PHI 300
Time of Possession NYG 39:10 PHI 20:50
First Downs NYG 26 PHI 17
Points off of turnovers NYG 6 PHI 14

I'll admit my bias but I saw the same thing as Dales. Going in, I fully expected the Giants to lose that game. Instead my only concern was that they would blow a game they had no business losing. It sure didn't look as close as the Pittsburgh game to me. Things could change come playoff time, but from what I saw Sunday I don't think the two teams are that close.

118
by Quincy :: Wed, 11/12/2008 - 4:09pm

Also your argument that the difference was the 6 points the Giants got as a result of drives less than 4 plays is completely undermined by your own stats which indicate that the Eagles got 7 points the same way.

128
by Andrew B :: Wed, 11/12/2008 - 6:29pm

"Also your argument that the difference was the 6 points the Giants got as a result of drives less than 4 plays is completely undermined by your own stats which indicate that the Eagles got 7 points the same way."

Not at all. The Eagles defense forced the Giants to kick by preventing them from scoring a touchdown after both turnovers. The Giants didn't voluntarily kick two under 30 yard field goals after a turnover, they were forced into it by an inability to score a touchdown. They just happened to be in field goal range when stopped because that is where the turnover gave them the ball to begin with. Had the turnover ocurred elsewhere on the field, the same set of plays by the Giants would have resulted in zero points and a punt. So the points should be credited to the field position of the turnover, and not to some skill by the Giants offense.

The Giants defense coughed up a touchdown in 2 plays to the Eagles when placed in a similar position. We have no reason to believe the Giants would have stopped that particular Eagles drive simply because they were unable to stop it on the field. The Giants had an opportunity to hold the Eagles to a field goal, and they failed miserably.

The Original Andrew

142
by Dales :: Thu, 11/13/2008 - 12:32pm

"Did we watch the same game?"

Apparently, we did not.

I saw a game where:
1) When neither team was making mistakes, the Giants were consistently moving the ball and the Eagles were not
2) When neither team was making mistakes, the Giants were pulling away and the Eagles fading

(Yeah, yeah, I know, you can't take away mistakes because it is a sign of a good team to not make them etc.)

3) But most significantly, consistently the Eagles were a play away from getting buried. For example, if not for the Jacobs fumble, they were likely to go down by 3 scores. Had Boss not dropped the ball in the end zone that hit him right in the hands, then the late rally by the Eagles would have been even more hopeless because they would have been down 3 scores rather than 2 with under 10 minutes to play. The Eagles scored a touchdown on 4th down, which was pretty much a do-or die play. The Eagles stopped the Giants on 2nd and 2 and then 3rd and 1 when a first down would have made it that they might not get the ball back and had left them with almost no time if they had.

Again, understand that this is nothing more than my impressions. However, it seemed to me that the Eagles were often a single play away from getting swamped, while the Giants were never in a similar position. That tells me that the game was closer to being a blowout than it was to the Eagles winning. It is to the Eagles credit that they kept from falling into the abyss, but the fact is that when a team continually is in a position where they have to convert make-or-break plays, then they are defying the odds if they continually do.

67
by phantomfacemask (not verified) :: Wed, 11/12/2008 - 12:03am

You can't cite untimely penalties against the Giants without mentioning the iffy (at best) facemask call against Gocong after a crucial stop on 3rd and 1. That call is the difference between a Giants punt and a Giants touchdown.

100
by Mark S. (not verified) :: Wed, 11/12/2008 - 12:32pm

Umm, what was iffy about that? It was unintentional on Gocong's part, but he did pull his head back by the mask.

103
by Andrew B :: Wed, 11/12/2008 - 1:05pm

Well, first, the penalty was on Bradley, not Gocong.

Second, the replay from the side of the field Bradley was on relative to Jacobs showed him clearly clotheslining Jacobs with his forearm under the chin (which caused the backwards movement of his head), and then grasping Jacobs by the left shoulder pad (his hand is clearly visible holding the jersey and pad while Jacobs is falling) and pulling him to the ground for no gain.

From the other side of the field (the initial camera angle), all you could see was Bradley's hand near Jacobs facemask, a jerking motion on his head, and Jacobs being pulled to the ground.

I suppose you can believe whatever you want to though. Bradley was clearly astonished that he got penalized.

As I understand it, a facemask penalty requires that you grasp the facemask with your hand. There was no evidence of that at all, and Jacobs head never turned as though his facemask was grasped.

The Original Andrew

111
by prs130 (not verified) :: Wed, 11/12/2008 - 2:36pm

yeah I saw Bradley taking great care to keep himself from grasping the facemask. It was a forearm to the mask followed by a grasp of the shoulder pad. Forgivable blown call if you're watching it full speed. Back, and to the left. Back, and to the left.

121
by Mark S. (not verified) :: Wed, 11/12/2008 - 4:28pm

I haven't check the replay from the other angle, but OK, say it wasn't a facemask. But the penalty for a clothesline in the NFL is also a 15 yard personal foul call. Just semantics, really, if the play occurred as you describe it: "showed him clearly clotheslining Jacobs with his forearm under the chin."

122
by Mark S. (not verified) :: Wed, 11/12/2008 - 4:32pm

Further explanation, from the Football Glossary:

"Clothesline
Definition: An illegal play in which a player strikes an opponent across the face or neck with an extended arm.
Examples: The penalty for a clothesline is 15 yards."

I'd say that pretty well describes Bradley's play, whether he got the facemask or not.

143
by Dales :: Thu, 11/13/2008 - 12:38pm

If it wasn't a facemask, it was a clothesline which is also a 15 yard penalty. Further, it did not lead to a touchdown, but rather a field goal.

74
by Michael19531 (not verified) :: Wed, 11/12/2008 - 1:42am

I agree with Dales 100%. Without looking at advanced or mainstream statistics, I believed that Philly could put 11 guys "In the box" and the Giants still could average 6 or 7 yards per running play. I would like to see the guys who run this website analyze how good this season's Giants o-line run blocking is compared to other o-lines from the past. I know that their data base only goes back to 1995, but from what I've seen from the Gmen's line thus far they compare very favorably with o-lines of the early '90s Cowboys, the '88 Bengals line led by Anthony Munoz and the '77-78 Patriots line led by John Hannah and Leon Gray.

87
by JSap :: Wed, 11/12/2008 - 9:56am

I agree that the Giants put up great rushing numbers in the game, but when they could have iced the game, they couldn't pick up 2 yards in two plays:

2-2-NYG 38 (4:37) 34-D.Ward left end to NYG 39 for 1 yard (58-T.Cole).
3-1-NYG 39 (3:55) 34-D.Ward right end to NYG 37 for -2 yards (20-B.Dawkins). FUMBLES (20-B.Dawkins), recovered by NYG-67-K.McKenzie at NYG 35. 67-K.McKenzie to NYG 35 for no gain (58-T.Cole).

If the Eagles had scored on their drive, would everyone be talking about the terrible decision by Coughlin to run twice and not be able to pick up a first down? Or the terrible decision to not challenge the fumble on that second play that clearly wasn't a fumble and put them in 4th and 6, so they didn't get a first down on the offsides penalty on the punt?

105
by Kurt :: Wed, 11/12/2008 - 1:22pm

No, nobody would have been talking about the decision by Coughlin to run the ball because (a) the Giants can actually run the ball, and (b) there's a difference between running on 3rd & 4th down with a lead, and running with 2 minutes left on your own side of the field with one timeout in a game you're losing.

As for the non-fumble, you're right that it wasn't a fumble, but I have to believe that the Eagles wouldv'e been a whole lot more careful about jumping offsides on 4th and 1 than they were on 4th and 6. The failure to challenge basically had no effect on the game.

115
by JSap :: Wed, 11/12/2008 - 3:13pm

Ok, so if it wasn't a bad decision to run the ball there, would people have been talking about the players inability to pick up the 2 yards? And I actually strayed from my original point, which was to refute this part of what I was responding to:

"Without looking at advanced or mainstream statistics, I believed that Philly could put 11 guys "In the box" and the Giants still could average 6 or 7 yards per running play."

My point was that the Giants were not able to pick up 2 yards on two carries when the Eagles essentially had 11 men in the box.

As for the non-fumble, I agree they might have been more careful and not committed that penalty, but if a coach is as brilliant as challenges as we were told all night that Coughlin is, shouldn't he have challenged that call to put the Giants in a position where the Eagles would have to be very careful when rushing the punter to NOT be offsides, and to not take a chance on a running into the kicker penalty? My point is that there is a big difference in the receiving team's options when it is 4th and 2 then when it is 4th and 6.

104
by Andrew B :: Wed, 11/12/2008 - 1:19pm

Oh please.

The Eagles played much of the game with 7 men in the box and the safety back deep doubling Burress (who managed one catch for 17 yards on a corner blitz).

When the Eagles did play with more men in the box in obvious running situations, the Giants didn't do very much. 25% of Giants rushing plays went for 2 yards or less.

The Original Andrew

110
by Kurt :: Wed, 11/12/2008 - 2:26pm

"25% of Giants rushing plays went for 2 yards or less."

Jeez, without looking into it, I'd guess that's a pretty terrific stat for the Giants. 75% of runs go for 3 or more? How often does that happen, especially when the team runs 30+ times? Let alone forty-freaking-five times?

Also, there's no evidence at all that the Giants did poorly in obvious running situations. Before the Ward runs with 4 minutes left, which go beyond "obvious running situations" into "everybody in the building knows they're running" territory, their runs of 2 or less came on

1st & 10
2nd & 6
2nd & 7
1st & 7
2nd & 5

That's 5 of their first 39 runs which went for less than 2 yards. Which of those runs, exactly, came in obvious running situations?

124
by Mark S. (not verified) :: Wed, 11/12/2008 - 4:48pm

Are you really arguing that the Eagles were mostly successful stopping the run? The Giants rushed for 219 yards. That's not really spin-able.

130
by Andrew B :: Wed, 11/12/2008 - 6:44pm

The Eagles really didn't try to stop the run. They tried to stop Burress, and they were very successful at that.

As to the 219 yards, they did it at 4.87 yards per run. That's good, but its not world-beating, and its well below the Giants average for the year, which had been 5.21.

That being said, I think the whole defensive game plan was a mistake. They should have concentrated on stopping Jacobs and let Eli try to beat them. While barely being challenged to be a part of the game, Eli completed under 55% of his passes, tossed an interception, was sacked once, and should have been flagged for intentional grounding. He obviously could have done much worse.

The Original Andrew

131
by Kurt :: Wed, 11/12/2008 - 10:04pm

"The Eagles really didn't try to stop the run."

Maybe they should have tried harder.

"They tried to stop Burress, and they were very successful at that."

Sure. And the Rams tried to stop Coles and Cotchery on Sunday too, and by that measure did a terrific job. Also, when they really felt like defending Thomas Jones, he didn't do much. Did you know that 34% of his carries went for 2 yards or less? Way to go Rams!

"As to the 219 yards, they did it at 4.87 yards per run. That's good, but its not world-beating, and its well below the Giants average for the year, which had been 5.21."

Yeah, but the Giants aren't actually that good at rushing. It's just that everyone gameplans to shut down Plaxico Burress.

2
by givetomethespam... :: Tue, 11/11/2008 - 5:42pm

I'm still waiting on a "Worst Special Teams DVOA Ever Watch" for the Vikings. The kick coverage looked almost competent in week 9, but then they allowed a punt return TD this week to one of the worst return men I've ever seen.

13
by Anonymous please? (not verified) :: Tue, 11/11/2008 - 6:17pm

Will Blackmon, worst return man you've ever seen? What in the hell gave you that impression? Sure, he ran back a punt from the endzone, which is retarded, but does that really earn him that label? He's been returning punts for not even a year now and he has 3 touchdowns. I don't know what you're looking for.

135
by Yakul (not verified) :: Thu, 11/13/2008 - 3:43am

Ignoring questions about the talents of Will Blackmon, I also would be interested to see the Worst Special Teams DVOA ever to know how close the Vikings are coming. There is something truly amazing about dreading every time your team has to punt. What made the experience more comical was that right before the punt that Blackmon took to the house, Troy Aikman, presumably looking at Kluwe's gross average for the season, commented "Chris Kluwe has had a good year punting the ball for the Vikings..." You could practically hear the noise of Minnesotans across the state shaking their heads in disbelief, given the bizarre "either he goes or I go" drama that seemed to play out earlier this season between Kluwe and Childress.

4
by sam :: Tue, 11/11/2008 - 5:56pm

Jags are 4-5, not 3-6. Not that it matters.

--
sam! or the original sam from the old FO

7
by joon :: Tue, 11/11/2008 - 6:01pm

and green bay didn't play tennessee. they played the vikings.

not to harp on anything, but does it seem like there have been a lot of errors in the DVOA tables and articles this year?

62
by Josh :: Tue, 11/11/2008 - 11:11pm

I think he was referring to GB's game last week against the Titans.

6
by vinyltoupee (not verified) :: Tue, 11/11/2008 - 6:00pm

Seriously? FO's love affair is getting really stale here. How many games do the Eagles have to lose before your computer rankings stop having fantasies? Did your computer even watch the game? The Giants dominated the Eagles. The Eagles ARE NOT, repeat, ARE NOT, by any reasonable measure, the second best team in football.

32
by Bowl Game Anomaly :: Tue, 11/11/2008 - 7:59pm

No, Aaron's computer didn't watch the game, and it doesn't understand the eagles' poor clock management, bad challenges, and other coaching gaffes. Basically, what DVOA is saying is that the Eagles would be a really good team if they had a good coach.

Actually, that may be a possible in-game coaching measure: Do some coaches consistently over- or underperform their team's DVOA? Is DVOA biased in favor of poor game managers? Does it make sense to compare DVOA to actual wins?:

Higher DVOA = better pregame preparation than in-game management
Higher actual wins = better in-game management than pregame prep

Of course the effect would have to be consistent over several years to be significant. And I'm not sure how to account for strength of schedule. But I bet guys like Reid and Holmgren would be exposed if you could get it to work right. And what coach's record outperforms his DVOA?

(Formerly "The McNabb Bowl Game Anomaly")

51
by Ashley Tate (not verified) :: Tue, 11/11/2008 - 8:48pm

Actually the Week 6 DVOA write-up included an extended discussion on whether DVOA favors certain teams and which teams over/under performed DVOA:

http://www.footballoutsiders.com/dvoa-ratings/week-6-dvoa-ratings-3

I believe the conclusion RE the Eagles was that they have underperformed DVOA in the past 3 years and overperformed in the years before that--all under Reid.

70
by Bowl Game Anomaly :: Wed, 11/12/2008 - 12:08am

Actually, that was a study of whether DVOA favors West Coast offense teams, looking at all WCO teams together, with only the Eagles getting close examination. I still think it would be worth exploring on a coach-by-coach basis. Maybe there would be no significant results, but you can't know unless you've tried.

(Formerly "The McNabb Bowl Game Anomaly")

125
by The Ninjalectual :: Wed, 11/12/2008 - 5:20pm

So Philly used to overperform their DVOA. The formula for calculating DVOA changes every year... did Aaron notice this overperformance and make adjustments to the formula and effectively "overreact"?

146
by tuluse :: Thu, 11/13/2008 - 5:25pm

I believe he recalculates all past DVOA's with the newest formula. So they overachieved based on the newest DVOA.

36
by phillyangst :: Tue, 11/11/2008 - 8:18pm

Hey, don't hate the [Iggles] player... hate the [sabermetrics] game!!!
DVOA loves Philadelphia!!!

82
by Alex51 :: Wed, 11/12/2008 - 6:39am

The Eagles ARE NOT, repeat, ARE NOT, by any reasonable measure, the second best team in football.

Maybe not. But I think I can find a reasonable measure that puts the Eagles at 3rd in the NFL: Margin of Victory. The Eagles have outscored their opponents by an average of 7.9 points/game, behind only the Giants, who rank second with 11.3 points/game, and the Titans, who rank first with 11.4 points/game.

So, the Eagles lost by 5 points to a team that is (according to MOV) about 3 points better than they are. Nothing strange about that. I think DVOA isn't being all that crazy here. Unless...

That's it! MOV loves Philadelphia!! Scoring more points that your opponent just makes your team look like it's good in the eyes of these crazy statistical rankings.

Did your computer even watch the game?

I find the thought of Aaron Schatz setting his computer on his couch next to him to watch the game (with a webcam, maybe?) hilarious.

148
by Dales :: Thu, 11/13/2008 - 7:22pm

I find it very easy to believe that the Eagles are among the very best in the league this year.

I also find it very easy to believe that the Eagles are almost as good as the Giants this year, and maybe even (despite the records) better.

What I don't find easy to believe is the idea that they played like it on Sunday night or were almost as good as the Giants were that night. They didn't and they weren't.

You should save the 1 ratings for people who are just trolling or being a-holes, and not for people who are making good-faith arguments or engaging in discussions that lead to good information on both sides of the arguments coming out, by the way.

Peace.

150
by Alex51 :: Fri, 11/14/2008 - 12:29am

You should save the 1 ratings for people who are just trolling or being a-holes, and not for people who are making good-faith arguments or engaging in discussions that lead to good information on both sides of the arguments coming out, by the way.

Usually, I save the 0 ratings for people who are just trolling. And while I don't doubt that your arguments were made in good faith, I thought they weren't very well supported, at least by FO standards (if I were rating you by Yahoo! message board standards, it would've been a 5, but come on, this is FO).

Specifically:

-You used a very small sample size to reach a conclusion in here:

And I believe the response to that is that the Eagles (ETA- recent, three and a half year) ability to underperform DVOA is becoming difficult to dismiss as a fluke.

And if you'd looked at a larger sample (like the last six and a half years, or the entire DVOA era), you probably would've noticed that the Eagles have shown no general pattern of over or under performing their DVOA during the DVOA era or Andy Reid's tenure as coach, which would undercut your argument.

-Here, you seemed a little overly critical towards DVOA, although that might not have been your intention. Also, I'm not sure what evidence you based your argument on:

DVOA apparently does not understand that while teams similar to the Eagles will score touchdowns x percent of the times in the two minute drill, the Eagles won't, and that this is repeatable.

If you want to argue that DVOA is missing something about the Eagles' two minute drill, your argument should include some evidence about their two minute drill. For all I know, the Eagles might not even be below average in the two minute drill, or being bad in the two minute drill might not be repeatable in the way you claim. Now, having watched the Eagles' two minute drill, I wouldn't be surprised if they were consistently bad in that area, but I'd need some specific evidence before I'd believe that DVOA was wrong, and my subjective impressions were right. If you have that sort of information, great, share it. If not, your argument is being based on your own subjective impressions, and I'm not sure why I should trust those over DVOA.

-And here, you used a question begging argument:

Anyhow, that game Sunday was a perfect example of a game that wasn't as close as the score and wasn't as close as DVOA seems to think it was. If those two teams played 100 games with each team performing at the same basic level of performance as they did on Sunday, the Giants would win a substantial majority of them.

The game Sunday was only an example of a game that wasn't as close as the score/DVOA thinks it was if we assume that the Giants would win a substantial majority of the time if both teams played 100 games just like they did on Sunday, which is the whole point in question. We don't think the Giants would win a substantial majority of those games, so the argument comes down to your subjective beliefs about 100 hypothetical games vs. our subjective beliefs about 100 hypothetical games.

That said, I will agree that I was too harsh initially, and while I still don't think that post was your best, it wasn't bad enough to warrant a 1.

151
by Boston Dan :: Fri, 11/14/2008 - 4:35pm

"Did your computer even watch the game?"

Priceless!!!!!!!!!!!!

8
by Pete (not verified) :: Tue, 11/11/2008 - 6:09pm

Will access to the new stats defense vs. receivers table throw into question PFP07's maxim, "Two fundamental matchup concepts that will help/guide your sit/start decisions: (1) QBs and RBs lose about 1/3 of their value against the best defenses in the league as opposed to the worst, while the effect is about half as strong for WRs (they lose 1/6 [15%] of their value). (2) The effect described in (1) is roughly the same for all players at a given position, regardless of how good they are." ???? Or has that already been debunked implicitly now that Barnwell writes a weekly matchup column with positive and negative percentages. I had not read how he gets those percentages.

9
by Aaron Schatz :: Tue, 11/11/2008 - 6:10pm

No idea why I seem to be making more typos this year than past years. The Packers thing is fixed. The Jaguars thing will be fixed. As for a "worst special teams" watch, the Vikings would need to have another really awful game before they got near the 2000 Bills.

16
by Unverified Telamon (not verified) :: Tue, 11/11/2008 - 6:24pm

We tried to warn you about hiring Emmitt Smith for your copy editing.

71
by Staubach12 :: Wed, 11/12/2008 - 1:14am

Hmm... Another Wade Phillips team with poor special teams? Amazing!

10
by Unverified Telamon (not verified) :: Tue, 11/11/2008 - 6:12pm

As I was looking at the playoff odds, I found myself wondering how the Steelers took such a big hit when they won their last game. Then I realized that, as a Redskins' fan, I'm just really wrapped up in my own suffering -_-,

11
by Mig (not verified) :: Tue, 11/11/2008 - 6:15pm

This love affair between DVOA and the Iggles is starting to annoy me.

38
by phillyangst :: Tue, 11/11/2008 - 8:22pm

Jealousy will get you no where. DVOA loves Philadelphia!!!

12
by Todd M. (not verified) :: Tue, 11/11/2008 - 6:15pm

"They looked almost as good as the Giants."

Really? No, really?

66
by the silent speaker (not verified) :: Tue, 11/11/2008 - 11:43pm

Yes, really. They looked almost as good as the Giants, the same way the Browns looked almost as good as the Cowboys earlier this year.

14
by jimm (not verified) :: Tue, 11/11/2008 - 6:19pm

"(They lost to the Vikings, and moved up three spots anyway because the system thinks they outplayed Minnesota by a small amount.)"

Hey I respect this site but if the system thinks GB outplayed Minnesota in that game then the system is completely useless.

15
by Anonymous please? (not verified) :: Tue, 11/11/2008 - 6:20pm

Yeah, man! Adrian Peterson is infallible!

17
by kevinNYC (not verified) :: Tue, 11/11/2008 - 6:31pm

Add me to the group who doesn't think "the Eagles looked almost as good as the Giants". For all the talk about challenges, the biggest play of the game IMO was Jacobs' 2nd quarter helicopter fumble. The Eagles were about to get run off the field in the first half.

In other news, the Giants' running DVOA is now slightly less than double the 2nd best team's.

Matt Ryan and the Falcons have the 3rd best passing DVOA? I guess it's time to admit maybe Mike Vick was the problem.

20
by Dales :: Tue, 11/11/2008 - 6:46pm

"For all the talk about challenges, the biggest play of the game IMO was Jacobs' 2nd quarter helicopter fumble"

At least on that part you can say that the Eagles forced that fumble with some nice hard hitting.

93
by Mark S. (not verified) :: Wed, 11/12/2008 - 11:43am

I’m going to somewhat sit the fence here and say that I basically agree with both positions: I do think the Eagles are a good team, but I also think the Giants really dominated that game. It was a blowout that was turned into a close contest basically because of one anomalous play: the Jacobs fumble. I say anomalous because I don’t ever remember Jacobs leaping into the air like that before…he doesn’t even dive when he is going for the goal line or third-and-oneAnd I’m going to go out on a limb and say he won’t be trying it again anytime soon. And, as noted, the Eagles really could have been run right off the field by halftime if not for that play.

Also: it’s a moot point except in the eyes of DVOA and fumble luck, but that Ward “fumble” before the Eagles’ final possession would have been overturned on replay. He was down before the ball came out. All game, the Eagles’ players were stripping the ball out of guys’ hands while they were already on the ground (and did anyone see the flying elbow drop Dawkins laid on Jacobs near the end zone?).

That said, the Eagles did some things well – they completely protected McNabb against a great pass rush. No sacks and few if any hits/hurries – so their pass blocking was great, even if the G-men totally stuffed the run.

144
by Dales :: Thu, 11/13/2008 - 12:41pm

"I do think the Eagles are a good team, but I also think the Giants really dominated that game."

You and I are in accord. That is what I took away from that game.

"Also: it’s a moot point except in the eyes of DVOA and fumble luck, but that Ward “fumble” before the Eagles’ final possession would have been overturned on replay. He was down before the ball came out."

Great point.

29
by JasonK :: Tue, 11/11/2008 - 7:32pm

"In other news, the Giants' running DVOA is now slightly less than double the 2nd best team's."

The individual stats are also getting gaudy:

Jacobs: 2nd in DYAR, 3rd in DVOA, 3rd in SuccRate
Ward: 4th in DYAR, 1st in DVOA, 4th in SuccRate

(Bradshaw's numbers aren't nearly as good, mostly because he's had 2 fumbles in only 36 carries.)

18
by vesini (not verified) :: Tue, 11/11/2008 - 6:32pm

While the Eagles high DVOA rating rankles me as well, here's a thought ... If you beat a team by 5 points and both teams score at least 30, how is the losing team significantly worse than the winning team?

In the meantime, can someone from FO please, please, pretty please do an in-depth analysis of the Eagles and DVOA ... Should I really hate Andy Reid this much?

52
by Ashley Tate (not verified) :: Tue, 11/11/2008 - 8:52pm

As I posted above, this was done in week 6:

http://www.footballoutsiders.com/dvoa-ratings/week-6-dvoa-ratings-3

19
by Dan J (not verified) :: Tue, 11/11/2008 - 6:35pm

Without wanting to state the obvious, there is absolutly no way the Packers outplayed the Vikings, at any stage of that game, apart from the punt return.
Something not quite right about that, and that is the first time I have really had cause to think that about this site.
Other than that, top work fellas

22
by Independent George :: Tue, 11/11/2008 - 6:53pm

I've no objections to the Giants being number one, but I'm nervous about how high they're scoring. They're arguably the best team in the league, but they're not dominating the way their DVOA would seem to indicate. Their remaining schedule is just brutal, but three wins could seal the division (especially if they include Washington and Baltimore).

49
by lawrence (not verified) :: Tue, 11/11/2008 - 8:44pm

Saying that "they're not dominating the way their DVOA would seem to indicate" makes no sense. They have scored 102 more points than they have allowed, the only team better is Tennessee with +103.

24
by Dales :: Tue, 11/11/2008 - 7:03pm

Actually, now that I look more, did the Eagles do that much better on the kickoffs? They had a lot of return yards, sure. But the average starting field position for them after kickoffs (not even counting the one they fumbled away) was the 32 yard line. The average starting field position for the Giants after Eagles' kickoffs was the 27 yard line. That is not nearly the advantage you would have thought given difference in return yards.

25
by DGL :: Tue, 11/11/2008 - 7:06pm

"What do these topsy-turvy rankings mean in the short run?" More griping about how DVOA sux!

43
by phillyangst :: Tue, 11/11/2008 - 8:33pm

DVOA doesn't "sux"... it [in the long run] loves Philadelphia!!!

27
by Dales :: Tue, 11/11/2008 - 7:30pm

"Beginning this week, we're offering a "Half Season Special" subscription to Football Outsiders Premium. For $25, or just over half the normal subscription fee"

Aaron, I was hoping you would do something like this, and on Thursday you will get another customer. Good deal.

42
by The Ninjalectual :: Tue, 11/11/2008 - 8:32pm

Meanwhile, DirecTV refuses to relent and last I checked, the NFL Sunday Ticket costs $250, the same as it did before week 1. Fuck you, DirecTV! I will spend my money at bars instead of buying your product.

108
by Andrew B :: Wed, 11/12/2008 - 1:39pm

So, since you refuse to buy Sunday Ticket yourself because it is $250, you are going to go and spend more money at bars (say $20 per weekend) that do buy Sunday Ticket than you would have spent at home with an $11.99 30 pack and the Sunday Ticket all for yourself? Makes a ton of sense. (/sarcasm)

The Original Andrew

126
by The Ninjalectual :: Wed, 11/12/2008 - 5:36pm

$250 / 7 weeks remaining= $35/week. If I had lived in a house with DirecTV to begin the season, it would have been $14/week over all 17 weeks. But I didn't have that option until week 5 or 6. My bar tabs are closer to $15/week incl. tips than to $10. And my team has either been off or been nationally televised for 3 weeks in a row now, so no need to go to a bar. NFL Ticket would be much more expensive for me than doing what I am doing, unfortunately. If they dropped it to say, $150 for the last 7 weeks, I would be all in.

Also, apropos of nothing, I live in Colorado (the microbrew capital of the Rockies) where an excellent 6-pack runs $9. Sierra Nevada Pale Ale is only considered "very good" around my parts =)

28
by Temo :: Tue, 11/11/2008 - 7:31pm

Dallas in 6 games with Romo as QB: 24.7% Offensive DVOA (3rd in the league), 4-2 record with a 2 point loss to the Redskins and an OT loss to the Cardinals.

Dallas after Romo misses 3 games: offensive dvoa decreases to 3.8% (17th), 1-2 record which includes blow out losses to the Rams and Giants.

Screw you, Brad Johnson. Screw you, soft defense. Screw you, Barnwell injury research.

72
by Key19 :: Wed, 11/12/2008 - 1:14am

Honestly, its gotten to the point where I want the Cowboys to not only make the Playoffs, but beat the Giants in the process. Just to show how screwed up DVOA is in terms of compensating for short-term and quantifiable horrible streaks and as a result how screwed up the Playoff Projections are.

75
by Love is like a bottle of gin (not verified) :: Wed, 11/12/2008 - 1:42am

If those streaks are so quantifiable maybe you should tell them how to quantify them? I am sure someone of your deep insight into the streakness of football teams has all the equations worked out...

DVOA isn't screwed up when it fails to consider these things, it simply considers what it considers. The reason these things are not considered now is because no one has come up with a statistically rigorous and useful way of handling them.

DVOA is a tool and acting like it should be the end all be all final word is what gets you so unnecessarily hot under the collar. You could also use a amalgamation of betting lines to produce a ranked listing of teams, or take a coaches poll or do 50 other things, none of them will be perfect, but all might be informative. DVOA is just another piece of information.

30
by jimm (not verified) :: Tue, 11/11/2008 - 7:43pm

Mewelde Moore 14th in DYAR, 7th in DVOA. The guy finally gets to play and no surprise to me he plays as well as most of the top backs and far better than backs on the same team.

Every year on Minnesota he had a higher DVOA than the other backs he was competing for time with and every year Tice and Childress found reasons not to play him. Is it any wonder neither of these coaches had any success.

31
by Housh P. :: Tue, 11/11/2008 - 7:58pm

To those who are upset about the Eagles being second, which teams are such no-brainers to be above them? Tennessee, maybe, but is it really obvious that Baltimore, Arizona, Pittsburgh or Green Bay are better than Philadelphia?

44
by Love is like a bottle of gin (not verified) :: Tue, 11/11/2008 - 8:34pm

Exactly, I have no problem with the Eagles being second right now. Do I think they are actually the second best team, no. Do I think they will still be second at the end of the year, no. Do I feel like I have enough evidence to back up these subjective feelings, no. There is a huge problem with underdetermination in the NFL and really the Eagles could be anywhere in the top half of the league but they won't play enough games for us to know for sure.

I think I would take BAL, WAS or TEN against them on an neutral field, and of course the NYG, but that is about it. Those of you who follow spreads know that Vegas (which seems very knowledgeable to me always) agrees more with FO than conventional wisdom (which 9 times in 10 is rank teams according to wins) about the Eagles.

48
by MTR (not verified) :: Tue, 11/11/2008 - 8:43pm

I think a big difference between DVOA and human opinion shows here. Humans use the thinking shown in the college polls where a loss always drops you. The computer says "yup, thought they would lose that one" and keeps them right where they were.

94
by Mark S. (not verified) :: Wed, 11/12/2008 - 12:05pm

I agree - and I would even take WAS off that list. Aside from their 2 divisional upset victories, the 'Skins haven't been all that impressive really. They got trounced by the Giants and Steelers, lost to the Rams, survived big scares against the Browns and Lions, and have yet to blow anyone out all season (they haven't won by more than one score yet). Don't get me wrong, I will be rooting for them hard to basically end Dallas' season on Sunday, but I'm not overly optimistic.

AS for PHI, as I mentioned elsewhere in this thread, I do believe they are one of the better teams in the league. The teams I would rate as better are: TEN, NYG, and BAL, and that's about it. I think they're about even with PIT.

58
by Anonymous? (not verified) :: Tue, 11/11/2008 - 9:58pm

Agreed. The point of objective stats is that they're objective. Now, some other objective compilation of stats might shake down to a different ranking, but overall you can't look at one team to evaluate the fitness of a statistical model, you have to look at the whole model. If the rankings had defensive powerhouses with barely over .500 records as five of the six top teams, I would start to wonder if DEFENSE was ranked too heavily in DVOA. but it makes no sense to say ONE TEAM is ranked too high, unless you can point to the actual factor that you think makes the team over-rated.

there are objective and subjective rankings. subjective rankings have the ability to include other factors not accounted for in the statistics, and that is about all i think they're good for. when a star player gets injured, the football pundits are quick to drop that team in the rankings, whereas an objective model will often lag behind in devaluing the affected team. poor coaching is better represented in subjective rankings than in objective rankings, unless those objective rankings somehow include coaching as a factor, so maybe that's why the eagles are much lower in ESPN's power rankings.

that being said, my team, the ravens, sit at #12 this week in ESPN's power rankings. you're team might also be underrated (or overrated!) in those standings. in terms of in-game DVOA, the ravens have only failed to post a higher DVOA than their opponent in one game this season. despite an absolutely dominant performance on sunday, and despite being tied with the giants for the second longest current winning streak this season, the ravens are ranked as the lowest 6-3 team in ESPN's power rankings. this is because (in my opinion) the football pundits who do those rankings generally follow each other around and base their opinions on overall game statistics and records instead of watching games closely and leaving pre-conceived biases at the door. they also (in my opinion) overvalue offensive powerhouses over stingy defenses. this has to do with the nature of defensive teams. a defensive team can abolutely DOMINATE a game and still come away with a final score of 17-10 (see for example the ravens first game this season, which looked like a blow-out from any perspective other than final score). the football experts, on the other hand, would rather see a team win a game 49-35. that to them is a far more dominant team.

anyway, the ravens play the giants this sunday. do you really think the match-up will be a #1 team blowing out a #12 team as ESPN indicates, or do you think the match-up will be a little closer? if you've watched both of these teams, it should be apparent that the ravens run D will likely halt the giants rushing (potent as it is), and the game will come down to a seasoned QB forced to pass more than normal on a decent pass defense versus an up-and-coming offense trying their new "tricks" and solid run-game against a stingy defense.

i'll take the peculiar outlier (in terms of record) of an objective statistical model over the less-than-useful predictions of some so-called football experts any day.

33
by SonOfDad (not verified) :: Tue, 11/11/2008 - 8:00pm

Giants ranked 4th on special teams?

Because Carney kicks tons of field goals? The fumbled kickoff from Philly? The poor long snapping ability of James Harrison?

I'm a Giants fan, can someone explain that one?

98
by Mark S. (not verified) :: Wed, 11/12/2008 - 12:23pm

Don't know for sure, but off the top of my head I'd say, in order of importance:

1. Carney's made all but one FG all year (which was blocked and returned for a TD in the Niners game)
2. Feagles is excellent at pinning opponents inside the 20
3. They don't give up big returns (Sunday vs. PHI was I think the most they've allowed all year, and really it was solid rather than big returns)
4. They return game is a pretty solid, really. Hixon especially has had some nice ones

34
by Aaron M. (not verified) :: Tue, 11/11/2008 - 8:02pm

Doesn't it seem like teams that do a lot more short passing (Brian Westbrook and the Eagles, Faulk and Welker for the Pats last year) do much better in DVOA maybe because a higher percentage of their passing plays work? I don't know, just throwing it out there.

If fumbles are randomly recovered, what about tipped passes? Do those count as "clean" interceptions? Because that kind of seems like luck to me too.

59
by Rich Conley (not verified) :: Tue, 11/11/2008 - 10:03pm

"Doesn't it seem like teams that do a lot more short passing (Brian Westbrook and the Eagles, Faulk and Welker for the Pats last year) do much better in DVOA maybe because a higher percentage of their passing plays work? I don't know, just throwing it out there."

Yeah, they do rank higher. Thats because they usually score more points. The patriots broke the scoring record with all those short passes.

109
by Andrew B :: Wed, 11/12/2008 - 2:02pm

"Doesn't it seem like teams that do a lot more short passing (Brian Westbrook and the Eagles, Faulk and Welker for the Pats last year) do much better in DVOA maybe because a higher percentage of their passing plays work?"

"Yeah, they do rank higher. Thats because they usually score more points. The patriots broke the scoring record with all those short passes."

The 2008 Eagles are on track to score 446 points, which would be the franchise all-time record, and are the 4th highest scoring team in the league right now. They are also 10th in lowest points allowed against on Defense.

(sarcasm)So yeah, they must really suck big donkey cocks this year like everyone is saying upthread from you, Rich. Totally overrated. No way are they a successful team.(/sarcasm)

The Original Andrew

116
by Tundrapat (not verified) :: Wed, 11/12/2008 - 4:00pm

So yeah, they must really suck big donkey cocks this year...

Really? Is this part of their training regimen? Or another of Andy Reid's questionable or asinine coaching decisions?

Get it? Donkeys...asinine...? Anyone? (sigh) Never mind.

Ahem - "DVOA totally sucks because it appears at first blush to be overly favoring a good team that is universally unloved (even in Philly). The ESPN Experts Picks are way better than this."

In all honesty, I do have to inquire about why the DVOA system thinks that Green Bay 'outperformed' the Vikings on Sunday (caution - 'homer' now posting about his team). That's really, really curious. They scored 14 points directly off of returns (interception and punt, no downs from scrimmage involved). The had two additional interceptions, one of which gave them a pretty short field.

Was the 'outplaying' due to special teams? Because it looked like the Vikings outperformed the Pack, even with Frerotte really trying hard to keep Green Bay in it.

79
by ammek :: Wed, 11/12/2008 - 4:56am

Wes Welker has negative DVOA.

Can we put the 'DVOA luvs WCO' myth to sleep, please?

90
by Rich Conley (not verified) :: Wed, 11/12/2008 - 11:21am

The problem with Welker is the way the patriots use him isn't typical of a WR. He's used almost like a running back. As far as the patriots are concerned, about half the time he gets the ball its an extended hand-off.

5 yards on a catch on first down is great to the patriots, despite the fact that its a "below average" reception.

37
by Yazata :: Tue, 11/11/2008 - 8:18pm

What information does premium database provide besides historical and current DVOA data? What I'm looking for is fantasy football lineup predictions for a given week. Does premium provide projections on what each player will perform for a given week? Or any other quantifiable data that helps me choose my lineup for the week, and maybe help me with trades?

Also, the premium 2006 DVOA free sample (www.footballoutsiders.com/premium/beta) is not accessible right now so I couldn't really see what Premium access would give me.

Could someone with premium access please let me know how the access helps them for fantasy football?

PS: I did purchase the 2008 Kubiak projections and used them for my draft, and so far they have been pretty nice, though some have been an outright miss (e.g. Santonio Holmes).

46
by Love is like a bottle of gin (not verified) :: Tue, 11/11/2008 - 8:38pm

To be fair everywhere missed on Holmes. People still don't take regression to the mean seriously enough, even for second and third year players it is a stronger trend than improvement.

65
by Aaron Schatz :: Tue, 11/11/2008 - 11:41pm

Sorry about the Premium BETA being down. It apparently never made it over from the old site and something has come up each time we planned on moving it. It should be set by Thursday.

69
by Yazata :: Wed, 11/12/2008 - 12:08am

Aaron,

I tried to purchase the Premium Access Half Season Special and everything goes well till I hit the purchase button and Paypal gives me a blank page. Is it possible to get this fixed before the first game this week i.e. Thursday, November 11, 2008?

39
by The Ninjalectual :: Tue, 11/11/2008 - 8:22pm

Washington is clearly ranked too low because even RaiderJoe thinks they are a lock for the playoffs. Our Superbowl chances are way better than this. JaMarcus Russell for defensive rookie of the year!!!!!!

41
by Will Allen (not verified) :: Tue, 11/11/2008 - 8:32pm

Jimm, Mewelde Moore didn't play much in Minnesota because he got hurt a lot, and Tice and Childress don't play running backs not named Adrian Peterson if they can't block, a talent which is not reflected DYAR and DVOA. I havn't watched the Steelers much this year, so maybe Moore has improved in this area, but I've also heard that the QB with the long last name is pretty beat up. If this is so, Moore's blocking skills might be a factor.

I'm no fan of either Childress or Tice (except for arguing that Tice was somewhat hamstrung by hideous ownership), but I HATE running backs who can't/won't block. It seems pointless to have guys on the field who get your qb killed, unless those guys have HOF talent in other areas.

123
by DFJinPgh (not verified) :: Wed, 11/12/2008 - 4:44pm

Lawrence Phillips. Steve Young. Will, I agree with that oh-so-rational hatred ...

45
by Will Allen (not verified) :: Tue, 11/11/2008 - 8:37pm

Yeah, it seems odd that Aaron's numbers say that the Packers outplayed the Vikings, but the punt return DOES count, as do Frerotte's interceptions. I'd be curious to know what the numbers look like absent those four plays.

50
by MTR (not verified) :: Tue, 11/11/2008 - 8:47pm

Aaron, it seems the teams with the easy past schedules are always clustered in the top of DVOA. I noticed the same thing last year. Since I don't know how you do your calculations I can't suggest what might be wrong, but it seems to be consistently odd.

92
by Rich Conley (not verified) :: Wed, 11/12/2008 - 11:29am

That has to be looked at under the idea that DVOA isn't finished until all the samples have come in. The opponent adjustments aren't perfect by any means. You almost always see a team's DVOA go down when it loses to a better team and go up when it beats a worse team. If DVOA were perfect, it wouldn't move in any sort of predictable way in these events.

But DVOA isn't perfect. Its good. Playing strings of good opponents depresses your DVOA and playing strings of bad opponents inflates it.

53
by jimm (not verified) :: Tue, 11/11/2008 - 9:00pm

Will - I don't know why you think Moore was such a poor blocker. The only times Childress and Tice used him with any regularity was third down passing plays which suggests he was an OK blocker. I haven't watched him in Pittsburgh but Roethlisberger has been taking sacks at a high rate for two years.

I wouldn't take him over Peterson, I'm not nuts, but I'd sure as hell have kept him over Taylor, particularly when you could probably get something in a trade for Taylor.

55
by Will Allen (not verified) :: Tue, 11/11/2008 - 9:24pm

jimm, I think that because I did see him on occasion try to pass block, if only in pre-season. He stunk. Also, Tice said he was, and if there is anything Tice knows, it is blocking.

Taylor gets on the field because he CAN pass block.

56
by John Doe (not verified) :: Tue, 11/11/2008 - 9:34pm

Fred Robbins (Giants mammoth DT) played the entire game with two broken hands, at least that's what the announcers said. He had a gigantic round ball of a cast on each one.

57
by Raiderjoe :: Tue, 11/11/2008 - 9:42pm

not worried about 28th ranking. broncos in first place bu got 21th rnaking. More imporamt is record. brocnos are 5-4 and Raiders are 2-7. Enough time to make up that. Raiders gets one game closer when b eat broncos on Noemnber 23.
Denver Broncos might not win another game except game vs Cifes on decemver 7. Broncos going to lose to Falcons, Raiders, Jets, Panters, Bills. Maybe beat Chargers,.
Chargers have lots of tough games left such as Steelers, Clots, Falcons, Raiders, Buccaneers.
Last week of seeson will determine who wins division. Bet on it.
Lets see Broncos play at San Diegop Chagrers , and Raiders play at Buccaneers. Going to get some revenge for Super bowl 37 when Raiders win to make playoffs and loss knocks buccaners out ofplauogs..

re: ninjalecital
Is poster drunk?
JaMarcus Russell is great up and cominger Qb for Raiders. Doesn't play defense. Also not rookie.
About other comment he made poster is rigt about Redkins
Redskins are lock for playoffs.

63
by Josh :: Tue, 11/11/2008 - 11:15pm

Awesome, dude. Just awesome.

Raiderjoe for Poster of the Year!

68
by the silent speaker (not verified) :: Wed, 11/12/2008 - 12:04am

But isn't the mark of a truly great player that they transcend their positions? Like by winning DROY as a second-year QB.

77
by Jim (not verified) :: Wed, 11/12/2008 - 2:53am

Hey Joe,

Sorry I missed you on the 31st, wanted to stop in and say goodbye. Hope you're doing OK man.

Jim
Go Bucs

80
by Arson55 :: Wed, 11/12/2008 - 5:07am

You're right about the Clots being a tough team. I don't see how anyone could ever beat them. They're a lock for the plauogs.

Man, how can you not love Raiderjoe?

120
by Tundrapat (not verified) :: Wed, 11/12/2008 - 4:19pm

Still trying too hard, joe. It's just not the same.

64
by jonnyblazin :: Tue, 11/11/2008 - 11:24pm

In terms of awesome matchups next week, how about the Giants #1 ranked rushing attack (+24.4%) vs. the Ravens #1 rush defense (-31.9%)? Earth, Wind, and Fire vs. Ray Lewis, Haloti Ngata and friends should be great to watch.

Of course, if the Giants are smart they'll just attack the Ravens crappy CBs through the air. But where is the fun in that?

83
by Anonymous? (not verified) :: Wed, 11/12/2008 - 8:40am

crappy CB, not CBs. Samari Rolle is back from injury as of last week (and played a great game). Especially with a nickel package on likely pass plays (Corey Ivy is a good nickelback, although he makes a terrible substitute for any other position), the ravens secondary is not as weak as past weeks (like the god-awful Clots* game).

*thanks for this Raiderjoe, I am forever indebted to you.

73
by Joshua (not verified) :: Wed, 11/12/2008 - 1:34am

I wonder how Belichick and the Patriots stack up viz the discussion of performance relative to DVOA. My impression is that the Pats have had some pretty good seasons in years where their DVOA was not so hot. Particularly in 2001 and 2005, and even this year. They have consistently done well also when their dvoa was top 10 but not top, like in 2003 and 2006. Is that in-game coaching? I have certainly seen some games where I thought that Belichick was coaching poorly, eg. last year's Super Bowl and this year's game against the Colts, but those games are amazingly rare... do the pats overperform their dvoa? are there other teams that consistently do better than dvoa would project?

76
by Utvikefan (not verified) :: Wed, 11/12/2008 - 2:22am

I think your computer could perhaps be taken in for a check up, since it liked Green Bays 1 of 11 3rd down conversion rate THAT much.

78
by ammek :: Wed, 11/12/2008 - 4:40am

Come on, be fair. The distances to go on Packers third downs were: 6, 13, 15, 2, 9, 14, 13, 11, 3, 5 (which became 10 with a false start) and 7. So it would be more accurate to say Green Bay sucked on first, second and third down. The punter was terrible, too, and they missed a FG, so let's add fourth down to the list.

How on earth does DVOA think the Packers outplayed its opponent? I can only suggest:
- the interceptions (although the Packers had two fumbles and recovered both);
- opponent adjustments on the running game; basically, it's de rigeur to run for 200 yards on the Pack, whereas Ryan Grant managed a respectable 4 yards/carry on an elite Vikes defense;
- number of plays: Minnesota ran 50% more plays than the Pack.

As yet, I don't think QB hurries are included in DVOA. But they should be, because the numbers need to understand that Aaron Rodgers was running for his life Sunday, which is why Green Bay produced nada in the passing game.

84
by ammek :: Wed, 11/12/2008 - 9:05am

OK, can anyone explain this? The Packers' DVOA for rushing offense ranks higher (22nd for the season) than Minnesota's (23rd). That's fairly counter-intuitive, but on closer inspection Minny's VOA is superior -- in other words, there's an opponent adjustment of about +/- 10% which boosts Green Bay.

However, on the individual stats page, we find Adrian Peterson ranked 18th in DVOA but Ryan Grant only 36th. Now, backup Brandon Jackson has been more valuable than his Purple counterpart Chester Taylor (though try telling as much to Mike McCarthy, who employs Jackson only in the Holmgren-inspired doomed-draw-or-screen-on-third-and-long role). But even then, I can't see how Green Bay makes up the gap.

85
by Tim R :: Wed, 11/12/2008 - 9:11am

I think in PFP 2005 there was an article in the Rams chapter which showed that Mike Martz was the coach whos teams had the most wins over DVOA expected wins.

86
by jimm (not verified) :: Wed, 11/12/2008 - 9:14am

Ammek - I've pointed out numerous occasions where the sum or the parts don't add up in DVOA. For instance the Vikings rank 8th in adjusted line yards - which suggests the Vikings have a good offensive line in terms of run blocking - yet they rank 23rd in rushing. Does that mean Chestor Taylor and Peterson are awful backs and the line is great?

91
by Rich Conley (not verified) :: Wed, 11/12/2008 - 11:23am

The adjusted line stats are just short of useless. They don't show anything near what they're meant to show. The assumption they're built on (long runs are the back's ability, less than 10 is the line's) is just plain false.

102
by ammek :: Wed, 11/12/2008 - 1:04pm

I wouldn't go that far, because I think Peterson is an exceptional case. He rarely loses yardage on a run; ALY believes this is because his line does not allow defenses to penetrate, and in most other cases it would be, but AP avoids losses because he has such great vision and breaks so many tackles. Equally, Peterson does not break many long runs -- and this is because teams sell out against the run when facing Minnesota, dropping a safety to prevent a breakaway rush.

In both cases, Peterson is quite atypical. Maybe he is unmeasurable.

What I would like to see is a breakdown of Minnesota's ALY between Peterson's carries and other backs' carries (mostly Taylor). That might give us a clue as to how the line is really performing. My hunch is that it's not great.

88
by froramder (not verified) :: Wed, 11/12/2008 - 10:08am

Clearly whatever formula you're using somehow ignores Adrian Peterson's value. It's cute that you think he's not even a top 15 back and everything, but you're denying the blatant reality of the situation. Did he really gain those yards per carry with flukes, again? The Minnesota line is good but it's not THAT good, the center is old and the right side is mediocre.

Adjust your formula or people like me will stop taking it seriously. If it isn't correlating to whatever you're trying to measure consistently, and it's not, then try some new numbers. Not just this but your rankings and all sorts of unintuitive and more importantly uncorrelated to the numbers. Also the obvious reality that Adrian Peterson is an elite running back. Seriously.

95
by ammek :: Wed, 11/12/2008 - 12:05pm

Please use the correct formula for this kind of complaint. You know, Adrian Peterson is clearly ranked too low because...

I think it's possible that the numbers are telling us something interesting and valuable about Peterson and the Vikings offense. I'm just unsure as to what it is. The passing game is higher ranked than the rush offense: could this be an example of "Kenton Keith Syndrome", whereby DVOA overvalues the offense's least expected (and its most under-used) weapon -- in Minny's case, the forward pass -- because other teams commit very little time and effort to stopping it?

97
by Love is like a bottle of gin (not verified) :: Wed, 11/12/2008 - 12:22pm

What the numbers are telling us is teams sell out against the run vs MIN and that si why the rushing numbers look bad.

139
by ammek :: Thu, 11/13/2008 - 10:41am

Defenses do the same against Tennessee and Baltimore, yet their DVOA rushing numbers correspond more closely to what we'd intuitively expect.

96
by mrh :: Wed, 11/12/2008 - 12:14pm

Hey, we're (Chiefs) #29. Woo-hoo. Of note, we could easily be 5-4:
Week 1 - Bowe drops a possible tying TD pass in the end zone in the last 2 minutes. It's a stretch but KC could have won.
Week 8 - late game loss to Jets. Winnable game.
Week 9 - OT loss to Bucs (note questionable PI call at end of Week 10 game to set up potential winning 2 pt play offsets questionable OPI call in this game that negates probable clinching 1st down)
Week 10 - failed 2 pt conversion on almost final play

None of this contrafactual history would have moved KC off the #29 position in DVOA. But they would be 5-4, tied with the Broncos (and ahead on H2H tiebreaker). They would still suck and DVOA would recognize it, but we'd be reading all sorts of press about the clutch Chiefs pulling out another late-game win. Anyone can see that a 1-8 team is bad - it's separating the bad 5-4 teams (Denver or my imaginary Chiefs) from good ones (Philly) that marks a good system, subjective or objective.

99
by Love is like a bottle of gin (not verified) :: Wed, 11/12/2008 - 12:26pm

exactly.

106
by Will Allen (not verified) :: Wed, 11/12/2008 - 1:25pm

Football Outsiders methods are much, much, more useful for measuring performance for total team, and offensive and defensive units, than they are for individual or sub-unit performance. That isn't a harsh criticism; until charting data becomes even more detailed and available much sooner, and then used to adjust for individual and subunit performance stats, capturing context well will be impossible. That will take a significant amount of labor.

I'll say it again: objectively, statistically, measuring football performance in way which reflects the interdependence between teammates within a football play is very, very, very hard, whch is why measuring the total team performance is so much easier.

107
by tomgdavis@juno.com :: Wed, 11/12/2008 - 1:28pm

Why all the trashing of Andy Reid's coaching? He's a very good, perhaps great, coach as measured by the simplest yardstick. He wins. He's 93-60. Whether he butchers clock management (he doesn't IMO) or fails too often on third and short (he does) is immaterial. He's still 93-60. And losing all the close ones this year isn't the mark of some coaching character flaw. George Seifert went 10-6 in 1991 and missed the playoffs despite outscoring the opposition by almost 10 points per game (393-239). It was just bad luck. Seifert was 14-2 in the other three out of his first four years.

Head coach is the most important job in pro football. Even more important than quarterback. The good and great coaches are pretty consistent too. There are a few coaches that have won for three or more franchises:

Parcells (77-49-1 NYG, 29-19 NYJ, 32-32, NEW, 34-30 DAL)

Schottenheimer (44-27 CLE, 101-58-1 KC, 8-8 WAS, 47-33 SDG)

Vermeil (54-47 PHI, 22-26 STL, 44-36 KC)
Vermeil had a losing record in Saint Louis, but it doesn't take much of a leap of faith to attribute the Rams post-1999 success to him and not Mike Martz.

Chuck Knox (69-48-1 LARams, 37-36 BUF, 80-63 SEA)
Knox's second stint with the Rams was a disaster, but he still won at his first three jobs.

And lots of guys have won in two cities:
George Allen, Don Coryell, Jon Gruden, Mike Holmgren, Tom Coughlin, Tony Dungy and Don Shula

Not many losing coaches get three jobs but Marion Campbell stank in two stints with the Falcons and one with the Eagles. Norv Turner is well on his way to his third losing coaching stint. To be only 15-10 with the talent he has on the Chargers is unforgivable. So is firing Schottenheimer to hire him, but I digress.

Dan Henning, Ray Perkins and Herman Edwards have stunk in two cities.

I've presented my evidence first and the thesis last, but here it is:
If great coaches consistently win when they move from one franchise to another then coaches who spend their whole coaching career with one team (Noll, Landry, Gibbs, Walsh, Grant, Halas, Cowher, Fisher, Madden and Andy Reid) and win consistently must also be great coaches.

If you've read this far you deserve my two favorite trivia questions:
1. Who are the only two men to be the first head coach for two different franchises?
2. What division were the Tampa Bay Bucs originally in?

119
by JSap :: Wed, 11/12/2008 - 4:14pm

I agree - I am not the biggest Andy Reid fan out there, and I admit he has flaws, but the Philadelphia fans who are in a rush to fire him need to realize it is not that easy to find a good coach just ask the Cowboys after Jimmy Johnson left or Miami after Shula left. I think sometimes you need stability and eventually you catch that lightning in a bottle. Look at these stats for his first 9 seasons:

86 wins, 58 losses, 5-6 in the playoffs and 1 Super Bowl loss.

Oh, wait a minute - those are Bill Cowher's stats after 9 years! Here are Reid's:

88 wins, 56 losses, 8-6 in the playoffs and 1 Super Bowl loss.

Look, I'm not saying that Andy Reid will have the same success that Cowher had in years 10-15, but I bet the Steelers fans who wanted Cowher out after missing the playoffs three years in a row are glad no one listened to them!

129
by Rich Conley (not verified) :: Wed, 11/12/2008 - 6:30pm

[quote]If great coaches consistently win when they move from one franchise to another then coaches who spend their whole coaching career with one team (Noll, Landry, Gibbs, Walsh, Grant, Halas, Cowher, Fisher, Madden and Andy Reid) and win consistently must also be great coaches.[/quote]

I'm not sure, with the current importance of GMs that I completely agree. If Turner keeps his job in SD, and AJ continues to give him a TON of talent to play with, he ends up with a great record despite being an awful coach. The guys who never move are tough to evaluate because you can't take the GM out of the picture.

That, and Gibbs v1.0 was a great coach. Gibbs v2.0, not so much.

132
by Raiderjoe :: Wed, 11/12/2008 - 10:13pm

1. answres is Paul Brown. coach Browns first and bengals first.
other guy is Dom Capers when coach Pantehrs and texans firts.

2 AFc West
Bucs finish 13.5 games behind Raiders and season only had 14 games for each team
Raiders beat Bucs 49-16 that year., really taught them a lesson

152
by Unverified Telamon (not verified) :: Fri, 11/14/2008 - 5:52pm

Andy Reid Just Wins, baby.

They should trade McNabb for Kyle Orton, who also Just Wins.

113
by Will Allen (not verified) :: Wed, 11/12/2008 - 2:52pm

Interesting post, tomgdavis. George Seifert's career is a huge enigma for me. I'll say it again; if Roger Craig doesn't drop the ball in the waning moments of the NFC Championshig Game in the 1990 season, it's quite likely that Seifer retires after the Super Bowl against the Chargers, or at least is never fired by the 49ers, ends up with the highest winning perecentage in NFL coaching history, and is mentioned in the same sentence as Lombardi. Which is why his career in Carolina is so interesting. Maybe Walsh left such a tremendous foundation that it couldn't be fouled up.

114
by Will Allen (not verified) :: Wed, 11/12/2008 - 2:57pm

Oh, and you are also correct; replacing Schottenheimer with Turner was too incompetent for words. A.J. Smith is a legend in his own mind!

133
by hector :: Thu, 11/13/2008 - 2:18am

Boldin and Fitzgerald have to be pretty close to even in real-life value, no? I realize Fitzgerald is the position prototype, size, physicality, unreal hands, but Boldin is more dangerous after the catch and better at some other things. I don't see a big difference, and they strike me as a 1 and 1A, not a 1 and 2.

134
by tomgdavis@juno.com :: Thu, 11/13/2008 - 2:38am

Rich, if GM, and not head coach, is the key job then we should be able to find lots of coaches that were big winners in one city and stunk up the joint in another or vice versa. I tried to find some examples. There aren't many. Hank Stram with the Saints, Dennis Green with the Cardinals and Seifert with Carolina were all pretty bad. These three counter-examples would hardly seem to negate the evidence that good coaches in general win everywhere.
And while I grant that Gibbs v2.0 was a shell of his former self remember that he was much older (see age of coaches article in PFP 2008).

Will, no Walsh-laid foundation could be so strong that Norv couldn't f--- it up.

raiderjoe, Bucs *only* finished 13, not 13.5, games behind Raiders in 1976 season. Raiders got thrashed by the Pats in their only loss and finished 13-1.

137
by ChrisFromNJ :: Thu, 11/13/2008 - 4:21am

Also Dan Reeves with the Giants, Mike Ditka with the Saints, Joe Gibbs v. 2.0, etc. There are quite a few.

138
by ammek :: Thu, 11/13/2008 - 5:00am

Tom Flores in Seattle, Bum Phillips in New Orleans, Chuck Knox in Buffalo, Jim Mora in Indianapolis, Lindy Infante in Indianapolis, George Wilson in Miami, Ray Rhodes in Green Bay, Mooch in Detroit, Lou Saban in Denver, Sid Gillman in LA, Buddy Parker in Pittsburgh, Curly Lambeau in Washington, Art Shell v.2.0, Ted Marchibroda v.2.0 and 3.0.....

140
by Wanker79 :: Thu, 11/13/2008 - 11:08am

Bill Belichick with Cleveland

147
by tuluse :: Thu, 11/13/2008 - 5:41pm

It's not really fair to compare look at Ditka's and Gibbs's second stints like that. The game changed so much while they were away, and the responsibilities of a head coach changed a lot.

145
by Rich Conley (not verified) :: Thu, 11/13/2008 - 2:10pm

"Rich, if GM, and not head coach, is the key job then we should be able to find lots of coaches that were big winners in one city and stunk up the joint in another or vice versa. I"

No, you're not going to find a lot of that, because coaches that stink in the first place rarely get second jobs.

And secondly, The importance of the GM is a relatively new phenomenon, and is a product of the cap age.

136
by Brandon (not verified) :: Thu, 11/13/2008 - 3:45am

Wow who would of guessed that DVOA would still be in love with the 5-4 Eagles. When will Barnwell and Schatz come out and confess their love for the Eagles.

141
by tomgdavis@juno.com :: Thu, 11/13/2008 - 11:20am

Impressive list of coaches that were "big winners" in one city and "stunk up the joint" in another Ammek. Did you research it? While you were sober?

I'll grant you Saban and Flores. I'll have to fight you on all the others.

First the easy ones:

Buddy Parker was a big winner in Detroit. He also won in Pittsburgh (51-47-6).

Chuck Knox was 37-36 in Buffalo. And won everywhere else also.

Ray Rhodes was 29-34-1 in Philly and 8-8 in Green Bay.

Jim Mora was 32-32 in Indy.

Lindy Infante didn't win anywhere. He had the same ghastly .375 winning percentage in Green Bay (24-40) and Indy (12-20).

Now the slightly harder ones:

George Wilson won in Detroit (53-45-6) and he did stink in Miami. However, I wouldn't qualify 53-46-6 as "big winner" and Wilson stunk in four years for an expansion team in an era when ALL expansion teams stunk.

Sid Gillman was 28-31-1 in LA which doesn't come anywhere near "stinking up the joint".

Ditto Curly Lambeau at 10-13-1 in Washington.

Marchibroda did stink in his second stint in Indy and also with the Ravens. However, he was more than sixty years old at these two jobs. (Again see article in PFP 2008).

Art Shell was 60 and took over the reins of one of the most pathetic franchises in recent memory when he went 2-14 in 2006. And when Shell won in his first stint no one pretended he was a great coach. Which probably explains why no one offered him a head coaching job in the twelve years between his two stints with the Raiders.

Mariucci was a big winner in San Francisco and he truly stunk up the joint in Detroit. However, he went from having Carmen Policy as GM to Matt Millen. Give the man a break.

149
by SB (not verified) :: Thu, 11/13/2008 - 9:32pm

The Bears have played +.500 ATL, PHI, CAR, TB, IND, MIN, and TEN - that's 7 of 9 games. And they've played the Lions twice. So of course they have only the 9th most difficult schedule through the first ten weeks.

Ladies and gentlemen, your 2008 Detroit Lions.

153
by Mig (not verified) :: Mon, 11/17/2008 - 12:32pm

How about them (really really really overated) Iggles ! Let's see where DVOA places them tomorrow. I just can't wait.