Writers of Pro Football Prospectus 2008

Most Recent FO Features

GurleyTod16.jpg

» OFI: SEC Surprises

In an opening week where even the elite teams in college football looked mortal, the SEC had two big surprises in Texas A&M and Georgia defeating their South Carolinian opponents by big scores.

09 Oct 2007

Week 5 DVOA Ratings

by Aaron Schatz

Here are the Football Outsiders team efficiency ratings through five weeks of 2007, 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 for opponent and consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season.

Opponent adjustments are currently set at 50% and will increase each week until they are full strength after Week 10.

DAVE is an early-season formula that combines early-season performance with our preseason projection to get a more accurate picture of how well teams will play over the course of the entire season. (DAVE stands for "DVOA Adjusted for Variation Early.") In this week's DAVE ratings, for teams with five games, the preseason projection counts for 27 percent, and the current DVOA counts for 73 percent. For teams with four games, the split is 40/60. In addition, the weight of Weeks 1 and 2 has been lowered slightly.

To save people some time, we request that you 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>

Commentary follows the numbers.


TEAM
TOTAL
DVOA
LAST
WEEK
DAVE RANK NON-ADJ
TOT VOA
W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 NE 67.7% 1 58.0% 1 75.4% 5-0 42.5% 2 -22.3% 3 2.9% 14
2 IND 51.7% 3 40.2% 2 54.6% 5-0 50.1% 1 -7.4% 10 -5.8% 24
3 PIT 39.4% 5 29.5% 3 49.8% 4-1 12.4% 7 -26.7% 2 0.2% 16
4 DAL 36.4% 2 23.5% 4 46.0% 5-0 26.0% 3 -12.8% 6 -2.4% 21
5 TB 24.4% 4 22.0% 5 23.7% 3-2 13.0% 6 -8.1% 9 3.3% 12
6 GB 21.2% 8 16.7% 8 20.3% 4-1 14.8% 5 -1.4% 14 5.1% 8
7 WAS 17.6% 15 13.8% 10 23.0% 3-1 -0.1% 16 -17.8% 4 -0.1% 18
8 BAL 16.5% 9 15.2% 9 20.3% 3-2 -3.4% 19 -13.5% 5 6.5% 5
9 TEN 16.4% 10 7.4% 12 11.3% 3-1 -4.5% 21 -27.6% 1 -6.6% 26
10 JAC 14.8% 14 19.4% 6 15.8% 3-1 11.5% 8 0.1% 16 3.4% 10
11 PHI 13.2% 11 18.6% 7 12.2% 1-3 5.4% 11 -10.0% 7 -2.2% 20
12 SEA 12.0% 6 8.4% 11 8.4% 3-2 1.1% 15 -4.7% 11 6.2% 6
13 ARI 9.4% 7 2.1% 15 15.0% 3-2 10.5% 9 -0.2% 15 -1.3% 19
14 MIN 6.3% 12 -0.4% 16 7.3% 1-3 -8.4% 25 -8.9% 8 5.8% 7
15 NYG 4.6% 16 4.2% 13 -3.1% 3-2 8.3% 10 -3.2% 13 -6.9% 27
16 HOU 3.3% 13 -3.7% 18 2.3% 3-2 -4.3% 20 3.3% 18 10.9% 1
TEAM
TOTAL
DVOA
LAST
WEEK
DAVE RANK NON-ADJ
TOT VOA
W-L OFFENSE
DVOA
OFF.
RANK
DEFENSE
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
17 SD -0.8% 23 2.9% 14 -6.9% 2-3 2.8% 14 7.5% 21 3.8% 9
18 CIN -2.5% 17 -3.3% 17 -9.6% 1-3 17.1% 4 6.8% 20 -12.8% 32
19 CLE -12.6% 20 -9.8% 19 -23.2% 2-3 -3.0% 18 17.5% 30 7.9% 3
20 MIA -14.6% 26 -14.9% 22 -16.7% 0-5 3.4% 13 18.1% 31 0.2% 17
21 ATL -17.6% 21 -11.9% 20 -16.0% 1-4 -8.6% 26 4.8% 19 -4.2% 22
22 DET -19.1% 18 -18.7% 26 -13.6% 3-2 -6.6% 23 8.1% 23 -4.4% 23
23 BUF -20.2% 28 -14.9% 21 -25.5% 1-4 -19.9% 30 9.6% 25 9.3% 2
24 DEN -22.2% 19 -18.3% 25 -21.3% 2-3 4.1% 12 14.2% 27 -12.1% 31
25 KC -22.9% 22 -22.3% 28 -14.5% 2-3 -18.0% 28 -3.5% 12 -8.3% 29
26 CAR -24.6% 25 -15.3% 23 -9.7% 3-2 -2.8% 17 15.9% 28 -5.9% 25
27 OAK -24.8% 24 -17.9% 24 -10.7% 2-2 -5.2% 22 9.4% 24 -10.1% 30
28 CHI -31.4% 27 -22.9% 29 -28.6% 2-3 -36.8% 31 1.9% 17 7.3% 4
29 STL -33.2% 31 -31.7% 31 -34.2% 0-5 -17.6% 27 17.4% 29 1.9% 15
30 NYJ -33.9% 29 -21.0% 27 -33.4% 1-4 -7.7% 24 29.1% 32 3.0% 13
31 NO -40.6% 32 -25.2% 30 -51.4% 0-4 -19.6% 29 13.1% 26 -7.9% 28
32 SF -43.0% 30 -33.3% 32 -39.7% 2-3 -38.5% 32 8.0% 22 3.4% 11

  • 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. (Note: This is based on 2007 performance only. In other words, DVOA, not DAVE.)
  • VARIANCE measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from least consistent (#1, highest variance) to most consistent (#32, smallest variance).


TEAM TOTAL
DVOA
W-L ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR. RANK
1 NE 67.7% 5-0 5.0 1 -14.0% 30 8.7% 5 2.0% 31
2 IND 51.7% 5-0 5.0 1 -3.7% 21 3.9% 11 4.7% 28
3 PIT 39.4% 4-1 3.2 7 -10.9% 27 -0.5% 18 39.2% 1
4 DAL 36.4% 5-0 3.8 3 -18.9% 32 7.6% 7 12.1% 15
5 TB 24.4% 3-2 3.3 5 -6.9% 24 -9.2% 29 29.9% 4
6 GB 21.2% 4-1 3.5 4 -1.6% 18 -12.5% 30 2.8% 29
7 WAS 17.6% 3-1 2.9 11 -3.2% 20 12.2% 1 10.5% 18
8 BAL 16.5% 3-2 3.2 8 -16.5% 31 11.5% 2 6.3% 25
9 TEN 16.4% 3-1 3.3 6 1.7% 14 -3.1% 20 1.1% 32
10 JAC 14.8% 3-1 3.2 9 -9.2% 26 7.2% 8 4.9% 27
11 PHI 13.2% 1-3 3.1 10 4.9% 9 3.7% 12 19.1% 9
12 SEA 12.0% 3-2 2.8 13 5.5% 8 -17.9% 32 24.8% 7
13 ARI 9.4% 3-2 2.9 12 -1.7% 19 -12.6% 31 26.7% 6
14 MIN 6.3% 1-3 2.7 14 -7.7% 25 -7.3% 27 7.8% 23
15 NYG 4.6% 3-2 2.6 16 10.9% 5 -0.4% 15 15.1% 11
16 HOU 3.3% 3-2 2.6 17 -5.6% 23 3.4% 13 8.7% 20
TEAM TOTAL
DVOA
W-L ESTIM.
WINS
RANK PAST
SCHED
RANK FUTURE
SCHED
RANK VAR. RANK
17 SD -0.8% 2-3 2.3 21 2.5% 13 -0.4% 16 31.0% 2
18 CIN -2.5% 1-3 2.6 15 16.7% 3 -5.4% 23 2.1% 30
19 CLE -12.6% 2-3 2.4 18 19.3% 1 -6.1% 25 12.8% 13
20 MIA -14.6% 0-5 1.8 24 -0.3% 16 10.9% 3 8.0% 22
21 ATL -17.6% 1-4 1.7 25 3.3% 11 -5.0% 22 6.0% 26
22 DET -19.1% 3-2 2.3 19 -3.8% 22 4.2% 10 22.9% 8
23 BUF -20.2% 1-4 1.9 22 17.5% 2 5.1% 9 13.9% 12
24 DEN -22.2% 2-3 1.6 26 4.2% 10 -3.2% 21 30.1% 3
25 KC -22.9% 2-3 2.3 20 -1.5% 17 -5.5% 24 12.5% 14
26 CAR -24.6% 3-2 1.8 23 -12.7% 28 7.8% 6 7.0% 24
27 OAK -24.8% 2-2 1.4 27 -13.7% 29 1.2% 14 19.1% 10
28 CHI -31.4% 2-3 1.1 29 3.0% 12 -2.3% 19 12.0% 16
29 STL -33.2% 0-5 1.2 28 0.5% 15 -0.5% 17 27.9% 5
30 NYJ -33.9% 1-4 1.0 31 10.8% 6 10.7% 4 8.4% 21
31 NO -40.6% 0-4 1.0 30 13.6% 4 -8.2% 28 9.2% 19
32 SF -43.0% 2-3 0.9 32 8.8% 7 -6.7% 26 11.0% 17

Today's DVOA commentary was going to talk in depth about the historic seasons that we are seeing in New England, Indianapolis, and Dallas. Then Tony Romo went out and threw 463 interceptions in just the first eight seconds of Monday Night Football.

OK, I may be exaggerating, but the Cowboys were outplayed by one of the worst teams in the NFL last night. That onside kick recovery may have given Dallas the chance to stay undefeated, but in the DVOA ratings, they are no longer the same team. The ratings for last night:

OFF DEF ST TOTAL
BUF -29.9% -44.0% 23.2% 37.3%
DAL -35.7% -26.1% -7.0% -16.7%

Take out opponent adjustments and special teams, and the teams were basically even: Buffalo offense/Dallas defense at -37.5%, and Dallas offense/Buffalo defense at -32.5%.

Shockingly, Romo didn't have the worst game of any quarterback this week. Monday Night Football doesn't end up in Quick Reads, but for those curious, Romo was worth -7.9 DPAR passing and -2.2 DPAR rushing. That's not as bad as Jon Kitna was against Washington. Five interceptions with 309 net passing yards is a better game than two interceptions with just 74 net passing yards.

One more note: Remember my post from last week about Dallas on third down? Romo threw two interceptions on third down last night, but the Cowboys still converted 10 of 15 third-down opportunities.

So now our discussion of historic greatness comes down to New England and Indianapolis. Last week, I talked about where these teams stand compared to all teams through four weeks since DVOA starts in 1996. After Indy's dominating victory over Tampa Bay, these teams look even better.

Once again, we've gone back and done DVOA if we were replaying the season a week at a time. We're going to look at teams after five games, rather than five weeks, so opponent adjustments are 50% strength for teams without bye weeks and 60% strength for teams with bye weeks, based only on the season to date. Here are the best teams ever through five games:

1999 Rams: 68.7%
2007 Patriots: 67.7%
2006 Bears: 64.7%
1996 Packers: 56.7%
2007 Colts: 51.7%
2001 Rams: 50.7%
2006 Chargers: 47.3%
2000 Ravens: 46.2%
1998 Broncos: 45.8%
2006 Eagles: 45.5%
1999 Jaguars: 45.2%
2004 Eagles: 44.6%

Taking the 2007 teams out of it, seven of the top ten teams were conference champions and four of the top ten teams eventually won the Super Bowl. Three other Super Bowl champions are also in the top 20: the 2005 Steelers, 2004 Patriots, and 1997 Broncos. Of course, both the Patriots and Colts can't win the Super Bowl, and both the Patriots and Colts can't even win the AFC championship.

The Pats and Colts are building these great seasons with offense, but you already knew that. Only seven teams since 1996 have an offensive DVOA over 35% in their first five games. Amazingly, the 2001 Rams, 2004 Colts, and 1998 Vikings are not on the list, but two teams this year rank second and third:

2000 Rams: 50.9%
2007 Colts: 50.1%
2007 Patriots: 42.5%
1999 Redskins: 41.3%
2004 Vikings: 36.1%
1998 Broncos: 35.7%
1999 Rams: 35.3%

The 1999 Rams are only seventh, by the way, because Kurt Warner didn't really become "Kurt Warner" until the second game of the season.

Are the 2007 Patriots and Colts having such great seasons because the overall offensive environment is stronger this year? Not really. Actually, it is probably working the other way around. As many people know, the baselines for DVOA are based on multiple years, which means the league rating for a specific season can be above or below 0%. For the last five years, the ratings go like this:

2002: 1.3%
2003: -1.8%
2004: 1.1%
2005: -1.0%
2006: 0.0%

Yes, last year was exactly average compared to the whole five-year period. Anyway, this year through five weeks, the league rating is 1.2%, roughly the same as 2002 and 2004. However, remove the Patriots and Colts, and the league rating drops to -2.2%. By comparison, if you remove the top two offenses of 2004 (Colts and Chiefs) the rest of the league drops from 1.1% to -1.0%. Remove the top two offenses of 2002 (Chiefs and Raiders) and the rest of the league drops from 1.3% to -0.5%.

By the way, you might be wondering how this is a big year for offense when it seems like every big name running back is averaging three yards per carry. Right now, the league DVOA is 5.5% for passing and -3.7% for rushing. So it's only a big year for one part of the offense.

One last thing, since I know somebody is going to ask. Here are the best defenses ever through five games:

1996 Packers: -45.2%
1999 Jaguars: -38.8%
2006 Ravens: -38.0%
1997 49ers: -37.6%
2000 Bucs: -36.8%
2002 Bucs: -34.3%
2006 Bears: -34.2%

Individual stats pages, offensive line, and defensive front seven are all updated. Each of these other pages will be updated through Week 5 later today or tonight, and we'll let you know when the updates are finished:

For comments on every team, look for DVOA on AOL, every Wednesday. (This will be linked on the FO Goes Mainstream page.)

Posted by: Aaron Schatz on 09 Oct 2007

195 comments, Last at 13 Oct 2007, 4:43pm by Alex

Comments

1
by david (not verified) :: Tue, 10/09/2007 - 4:06pm

I'm actually shocked that the Broncos still have a top half of the league offense. A whole lot of not scoring points makes them look worse than they actually are, I suppose.

2
by Cosmos (not verified) :: Tue, 10/09/2007 - 4:08pm

God dammed Romo, yeah he plays like Farve alright...

3
by Catfish (not verified) :: Tue, 10/09/2007 - 4:09pm

"Three other Super Bowl champions are also in the top 20: the 2005 Steelers, 2004 Patriots, and 2007 Broncos."

This may surprise Denver fans. I assume you mean 1997?

4
by Matt (not verified) :: Tue, 10/09/2007 - 4:09pm

"Three other Super Bowl champions are also in the top 20: the 2005 Steelers, 2004 Patriots, and 2007 Broncos."

Meaning the 1997 Broncos?

But the main reason I am here is not to nitpick Aaron but instead Sportsguy, for the genius of this little hindsight-is-20-20 gem in last week's NFL column.

"But you can't tell me Pittsburgh is better than Seattle, or vice-versa. So why would Pittsburgh be favored by six this week? How does that make sense?"

5
by JJ (not verified) :: Tue, 10/09/2007 - 4:11pm

Three other Super Bowl champions are also in the top 20: the 2005 Steelers, 2004 Patriots, and 2007 Broncos.

I think you mean the 1997 Broncos...

6
by admin :: Tue, 10/09/2007 - 4:17pm

Heh. Three people fix the same typo in a matter of seconds. I love the Internet.

7
by Sidewards (not verified) :: Tue, 10/09/2007 - 4:17pm

Tennessee is 1st in Defensive DVOA?
FIRST?

8
by Rich Conley (not verified) :: Tue, 10/09/2007 - 4:19pm

7.

Vince Young just wins games.

9
by hooper (not verified) :: Tue, 10/09/2007 - 4:19pm

Nah, I think he means the 2007 Broncos. Believe! ;)

Seriously, the AFC West looks bad in this. SD has the highest DVOA at #17, and Denver, KC, and Oakland are #24, 25, and 27 respectively. Ugh.

10
by zlionsfan (not verified) :: Tue, 10/09/2007 - 4:22pm

For those who still believe in the Lions:

DVOA: negative.
DAVE: negative.
VOA: negative.
Offense: negative.
Defense: positive.
Special teams: negative.

Past schedule: negative.
Future schedule: positive.

Sadly, there's also this from the Playoff Odds Report:
Chance of getting #1 pick: 0.0%.

Bandwagon: Found On Road Dead.

(This comment sponsored by the American Colon Society. Punctuation: use it as you wish.)

11
by Will Allen (not verified) :: Tue, 10/09/2007 - 4:23pm

Unsurprisingly, DVOA also says the Packers played better than the Bears on Sunday.

12
by Matt (not verified) :: Tue, 10/09/2007 - 4:25pm

9:

Oakland is clearly ranked too low because 1st Place in the AFC West, baby! Standings are way better than this. Daunte Culpepper is back, biatches!

13
by Rich Conley (not verified) :: Tue, 10/09/2007 - 4:25pm

"SD has the highest DVOA at #17,"

Thats not going to stay that way. San Diego will stomp oakland (who can't stop the run) and move up into the top half. They'll go into their bye at 3-3 with two consecutive stomps, and get to come out against Houston.

14
by Fnor (not verified) :: Tue, 10/09/2007 - 4:25pm

“But you can’t tell me Pittsburgh is better than Seattle, or vice-versa. So why would Pittsburgh be favored by six this week? How does that make sense?�

Why would Chewbacca live on Endor? He's a wookie! It doesn't fit!

Ahem. It just makes me sad to see two historically great teams in the AFC. Being third isn't really very helpful when you're convinced your team couldn't beat #1 or #2. Although I suppose that have the best shot.

15
by B (not verified) :: Tue, 10/09/2007 - 4:25pm

Any chance we can move the Colts back to the NFC so we can have a really competitive super-bowl this year? And this will have the added benefit of keeping Dallas out. It's win-win. Well, except for the endless you know who/you know who debate.

16
by throughthelookingglass (not verified) :: Tue, 10/09/2007 - 4:27pm

7: Read the back cover of PFP 07. :)

Also, have the Niners really been that bad?

17
by hooper (not verified) :: Tue, 10/09/2007 - 4:28pm

Re: 12 (but not aimed at you, just a post of opportunity)

Right now, crowing about first place in the AFC West is like virginity among prostitutes; it sounds special, but means nothing.

18
by sam (not verified) :: Tue, 10/09/2007 - 4:34pm

In response to the question from Quick Reads about David Garrard... he's 5th in DVOA. I think he's playing pretty well.

I don't understand how the Jacksonville defense can be ranked 16th overall in DVOA and best in Variance. They were atrocious against Tennessee in Week 1 I thought. They've improved each week since. How can they be the least varying?

19
by shannon (not verified) :: Tue, 10/09/2007 - 4:34pm

*back* to the NFC? Maybe you mean back to the AFC East?

Man, those were two bad teams in the early 90's.

20
by hooper (not verified) :: Tue, 10/09/2007 - 4:37pm

Re: 19

The Colts were an NFL team before the merger. They were one of the teams that shifted to the AFC when the merger went through; I think that's the reference. "Back to the NFL" wouldn't have made much sense.

21
by Eddo (not verified) :: Tue, 10/09/2007 - 4:38pm

19: The Colts (Baltimore version) were members of the NFL until the 1970 merger. Hence how they played the Jets in Super Bowl III.

22
by Thoreau (not verified) :: Tue, 10/09/2007 - 4:40pm

How does Pittsburgh have 3.2 estimated wins when a) they're 3rd in team efficieny, and b) they've beat the living crap out of 4 teams this year? Is that a typo/should be 4.2 wins, perchance?

23
by gobigred (not verified) :: Tue, 10/09/2007 - 4:40pm

Re: 4
Actually, last week the Steelers were #5 in DVOA and the Seahawks were #6 so just based on overall DVOA for the year, Simmons wasn't far off. Not that I don't think Simmons says ridiculous things, but that was one of the least ridiculous things he has said.

24
by Matt (not verified) :: Tue, 10/09/2007 - 4:44pm

14 - Historically great teams in the AFC? At this point, sure. I'll wait till the end of the season to write the history, however. I don't expect the Steelers to beat the Patriots or the Colts either, but nothing will "convince" me of that until they play them and don't beat them. Just looking at these three teams over the course of the past decade, everybody knew that the Steelers couldn't beat the Colts in 2005, that Manning couldn't overcome his big game curse against Brady in 2006, that the 2001 no-name Patriots couldn't run with the big-name Steelers, etc.

17 - No offense taken, esp. as I was not serious in the least with my post at 12. I wonder if I violated the complaint rules, however, as standings don't really count as a "subjective ranking system."

25
by dryheat (not verified) :: Tue, 10/09/2007 - 4:46pm

#7, You're six posts too late.

#19, The AFC (with the exception of the Jaguars and the Browns or Ravens, depending on how you see the franchise move) is essentially the expansion part of the NFL, created with the NFL/AFL merger. The three exceptions were Baltimore, Pittsburgh, and Cleveland, who were in the NFL, but switched conferences in the name of competitive balance.

26
by John Kim (not verified) :: Tue, 10/09/2007 - 4:52pm

7, Tennessee being first in Defensive DVOA is not a surprise at all IMO. After all they did keep the #1 Offense in the league (DVOA-wise) to only 22 points. Peyton Manning had his worse game of the season so far against the Titans.
The Titans D-Line (Actually, lets be honest here, Kyle Vanden Bosch and Albert Haynesworth with a sprinkle of Tony Brown) have been mauling offensive lines and getting into and disrupting the backfield. Albert Haynesworth is playing like a man posessed and would probably be talked about even more if it wasn't for his stomping spree (okay... it was only a single one, so that doesn't justify calling it a spree) and the fact that he plays for Tennessee.

Vs. the Run Titans have been fantastic, with Haynesworth and Tony Brown sealing off all gaps between the Tackles, and Thornton and Bulluck's speediness effectively shutting down the outside runs.

Vs. the Pass Titans have been decent, and it's all thanks to Cortland Finnegan and Nick Harper. Especially with Finnegan looking like Pacman with his physical play on the WRs.

27
by Catfish (not verified) :: Tue, 10/09/2007 - 4:53pm

Manning and Brady are basically tied for first in DPAR with 57.8. That puts them on pace for 185 DPAR this year, which would be the best of the DVOA era (beating Manning in 06 @ 175 and Manning in 04 @ 170).

Also, raise your hands if you thought Sammy Morris would be the 2nd RB in DPAR at any point this season.

28
by Optimistic Packer Fan (not verified) :: Tue, 10/09/2007 - 4:53pm

22: Their variance hurts them a lot. I'm not sure how their splits work out, but they might also be getting hurt in the formula by poor situational performance.

29
by Rich Conley (not verified) :: Tue, 10/09/2007 - 4:58pm

RE:Tenessee

One of the interesting things about them, is there are only two teams in the top 10 (defenses) that are being adjusted upward. Washington and Tenessee. Tenessee is receiving a big adjustment up (+4%) while Pittsburgh is being greatly adjusted downward (-7%) New England is being adjusted down 1.5%.

Pittsburgh's VOA is over 10% higher than Tennesee's, so SoS is playing a huge part here.

30
by Matt (not verified) :: Tue, 10/09/2007 - 4:58pm

23 -

That's why I said my criticism of the estimable Mr. Simmons was hindsight-based. Perhaps I should have been clearer. I too expected a better game and two more evenly matched teams.

That said, Sunday was a beatdown -- so whatever DVOA said before or after, I'm willing to say that the Steelers are a better team at the moment.

Moreover, I guess I'm willing to go out on a statistical limb and go in way over my head (just to mix metaphors as thoroughly as possible) and suggest that the Steelers #2 ranking in Weeks 2 and 3, when compared to the Seahawks #14 ranking in both of those weeks, was an indicator to consider along with the Week 4 rankings that came in after the Steelers worst game of the year and the Seahawks most dominant game. Now weeks 2 and 3 were still plain old VOA, without any opponent adjustments, etc., but thus far it looks like the Week 4 timeframe when these two teams were next to each other in the rankings was a bit of an anomaly.

I am sure others will disabuse me of that notion, however, if my lack of statistics knowledge is on full display here when I try to talk about trends and anomalies and all that good stuff.

31
by kevinNYC (not verified) :: Tue, 10/09/2007 - 4:58pm

I know no one cares except Giants fans, but to be ranked 13th on defense after being 25% WORSE than the 31st ranked defense after week one is quite a feat.

It's also nice to see them top 6 running the ball and against the run. The only team better in both categories is the Pats.

Derrick Ward is the 5th rated DVOA RB? I don't want to say RBs are "fungible", but....

32
by shannon (not verified) :: Tue, 10/09/2007 - 4:59pm

touche'

Still, feels good having lived through the Rod Rust era.

33
by Rich Conley (not verified) :: Tue, 10/09/2007 - 5:00pm

"Vs. the Run Titans have been fantastic, with Haynesworth and Tony Brown sealing off all gaps between the Tackles, and Thornton and Bulluck’s speediness effectively shutting down the outside runs.

Vs. the Pass Titans have been decent, and it’s all thanks to Cortland Finnegan and Nick Harper. Especially with Finnegan looking like Pacman with his physical play on the WRs. "

you've got that backwards. They're -40% DVOA agaisnt the pass, and -5% against the rush. Surprisingly, their Pass Defense is better without Pacman.

34
by thestar5 (not verified) :: Tue, 10/09/2007 - 5:06pm

15,

Can we at least let the games be played before we crown the AFC? How can you say that the superbowl won't be competitive? Anything can happen in just one game. I mean, the Texans almost beat the Colts without Andre Johnson. Its silly to think Indy is some kind of unstoppable force. Indy started out 9-0 last year before the got beat by the 'Boys. Then they finished the season flat but still won the SB. The SB isn't decided by DVOA after week 5. Lets not crown the Colts and Pats just yet.

35
by admin :: Tue, 10/09/2007 - 5:06pm

Individual pages now updated, so are playoff odds. We've made a couple of adjustments on playoff odds this week, which regresses teams to the mean slightly as the year goes along.

36
by Rich Conley (not verified) :: Tue, 10/09/2007 - 5:13pm

30.

What I'd take from the steelers bouncing around is that Arizona's DVOA is probably much closer to the truth than its DAVE.

37
by John Kim (not verified) :: Tue, 10/09/2007 - 5:13pm

"you’ve got that backwards. They’re -40% DVOA agaisnt the pass, and -5% against the rush. Surprisingly, their Pass Defense is better without Pacman. "

Really? I hadn't checked out the actual numbers for the defense yet. That is surprising. From what I've seen (which is 3 games on a tiny 320x320 part of the screen on sopcast and 2 games on TV) the Tennessee D-Line has been doing a great job of stuffing the runs and not giving up long-runs either, especially up the middle. And that our pass defense was less-than-reliable at certain points but overall, above-average to good.

And as for our Pass D, IMO A lot of it has to do with Cortland Finnegan stepping up big time. Nick Harper was thought to be the only above-average player among our CBs, but Finnegan has been playing great, FWIS.
I think we should make Cortland Finnegan the next cult-hit player, a la Pacman
=D

38
by sam (not verified) :: Tue, 10/09/2007 - 5:13pm

Whoops... thought those Garrard stats looked familiar...

He's actually 4th in DPAR and 3rd in DVOA. People keep calling him a game manager in the line of superbowl-winning Trent Dilfer but I don't think that's an accurate reflection of what he's accomplished this year (especially with a very mediocre group of receivers).

For those who laughed at JAX signing Dennis Northcutt... well, he's 16th in DPAR and 8th in DVOA, so I suppose that he's working out so far.

39
by inkakola (not verified) :: Tue, 10/09/2007 - 5:19pm

so if you know what the baseline DVOA 'could' have been if not for two teams, at what point does it skew the whole 'average performance' thing. how much of a difference to the average do two outliers have to make for there to be a problem? or is this never a problem?

40
by Kal (not verified) :: Tue, 10/09/2007 - 5:23pm

SD has the highest DVOA at #17,�

Thats not going to stay that way. San Diego will stomp oakland (who can’t stop the run) and move up into the top half.

I don't think that SD mauling the 32nd-ranked rushing defense is going to make their DVOA improve that much. And that assumes that Turner actually capitalizes on that weakness of Oakland; he might instead try and pass a lot and get SD killed.

I mean, we did think SD was going to handily crush KC, right?

41
by Carlos (not verified) :: Tue, 10/09/2007 - 5:27pm

sorry if this has been discussed to death elsewhere, but Dick Jauron absolutely blew that game last night!

Gave the 'boys 3 points before halftime with that stupid FG attempt/clock mgt. Then foregoes the run followed by FG attempt with 5 minutes to go that would have ended the game, allowing Trent to throw the pick.

That's just terrible, terrible decision making, w/o even getting in to how do you allow a sideline out pattern with 7 seconds and no timeouts left -- why bother putting defenders in the middle of the field!

42
by Rich Conley (not verified) :: Tue, 10/09/2007 - 5:31pm

40.

Kal, when you can run at will on a team, the team brings safties and LB'ers forward, and then you can pass at will.

Opponent adjustments aren't perfect, and offense moves the ratings more than defense (theres more range in offense, its easier to put up a +50% offense than a -50% defense).

43
by Rich Conley (not verified) :: Tue, 10/09/2007 - 5:35pm

"hen foregoes the run followed by FG attempt with 5 minutes to go that would have ended the game, allowing Trent to throw the pick."

Forgoes the run?

Last Drive: Run for 2, Run for 1, Incomplete pass

2nd to Last Drive: P4, R2, P8, R14, Incomplete, R2, Pick

"Abandoning the run" wasn't the problem. They were getting 2 yards a carry. They probably would have won that game if they'd stopped running so much on 1st and 2nd.

44
by Matt (not verified) :: Tue, 10/09/2007 - 5:40pm

43 - I think his point was about the specific playcalling on that Edwards' pick. Whatever they were getting on the run and whatever they could have done differently during the rest of the game, a Buffalo FG there makes it an 11-point spread with about 4 minutes left and (maybe?) puts the game out of reach.

45
by Lyford (not verified) :: Tue, 10/09/2007 - 5:43pm

w/o even getting in to how do you allow a sideline out pattern with 7 seconds and no timeouts left — why bother putting defenders in the middle of the field! That was the thing that got me, and I haven't heard it even get mentioned anywhere. That was inexcusable - let them catch a pass anywhere, but don't let any Cowboy get closer to the sideline than a Bill, and they win that game. That was unbelievable, and the broadcasters (and all of the commentary this morning) didn't seem to notice it...

46
by Kal (not verified) :: Tue, 10/09/2007 - 5:48pm

Kal, when you can run at will on a team, the team brings safties and LB’ers forward, and then you can pass at will.

except apparently for the raiders, who have the 11th best pass d and the 32nd best run D. In that case it appears that you can't pass at will on them but running at will is fine. Or they stubbornly don't do what you say they do, and just keep getting crushed against the run but stay fine against the pass.

Or maybe they man up well and don't play a lot of zones. That's my suspicion given their personnel and the games I've watched.

Really, the next worst team at rush D is NYJ, and they're 9 points better. But the pass D is pretty hot. If Turner goes out there and decided that Rivers should win the game on his own, well, they might lose. SD should win handily, but they'll only do so if they actually go after the Oakland weakness.

47
by Rich Conley (not verified) :: Tue, 10/09/2007 - 5:55pm

46

One thing to note is that the Chargers Rushing DVOA is -10%. I think thats more of an artifact of playing the Patriots, Packers, and Bears (all better than -10% Rush Defense) than it is a statement about their own team.

They've played two teams with Bad run Ds' so far. The Broncos (41-3) and the Chiefs. They dominated that chiefs game until they stopped running the ball.

Now, barring a Chiefs-style Norv-Meltdown, I think the -10% rushing Chargers run all over the -25% rush D Raiders. I think they run the ball well enough that their run offense approaches positive dvoa for the year.

48
by Carlos (not verified) :: Tue, 10/09/2007 - 6:03pm

“Abandoning the run� wasn’t the problem

Fake quotes = strawman.

Or maybe you're not talking to me?

I'm talking very specifically about:

1-10 at Dal 27: 14 yard run by Lynch.

Now there's 7:09 and you're up by 8. FG makes it 11, and basically means game over. (would really mean game over if they'd not given up the 3 points at the half).

Instead of three straight runs followed by a FG (assuming no first down or touchdown), leaving Dallas down 11 with 4:45 to play, the play calling is

1-10 at Dal 13: Pass (incomplete)
2-10: Run (2yd)
3-8 at Dal 11: Pass (INT)

Just idiotic. I hate it when coaches get really conservative with a lead 5 minutes into the 3Qtr, but this was

49
by NewsToTom (not verified) :: Tue, 10/09/2007 - 6:06pm

TEN's visibly stout yet statistically mediocre run defense becomes more explicable if you look at the Defensive Line stats. The Titans rank 3rd in ALY and 2nd in stop rate, but are 30th in 10+ rate. The Titans are also 24th in Power, which surprises me. At least they're not Green Bay, which hasn't stopped anybody this year in a Power situation.

50
by Will Allen (not verified) :: Tue, 10/09/2007 - 6:18pm

Yeah, once the ball was in chip shot field goal range, with only that much time remaining, it makes no sense to have an inexperienced qb throwing the ball in a red zone crowded with defenders.

51
by Rich Conley (not verified) :: Tue, 10/09/2007 - 6:21pm

50.

I agree, but the run wasn't working. The issue is that the pass got picked off... not that the playcall was a pass. If the play wasn't there Edwards should have thrown it away or eaten the ball.

52
by throughthelookingglass (not verified) :: Tue, 10/09/2007 - 6:21pm

Looking at the individual DPAR pages, and my own subjective observations from watching Dolphins games, is it too early to "crown" Ronnie Brown as an elite back?

53
by Rich Conley (not verified) :: Tue, 10/09/2007 - 6:28pm

52.

Looking at the RB page, I for one, hope that Dallas continues to insist on giving Julius Jones the ball.

54
by B (not verified) :: Tue, 10/09/2007 - 6:30pm

52: I'd say it's about a year to late. Probably to early to crown him as an elite fantasy back, though.

55
by cdcox (not verified) :: Tue, 10/09/2007 - 6:34pm

I handle the regressing towards the mean problem in my playoff odds forecasts by increasing the variance of single game outcomes. That tends to help the weak teams and hurt the strong teams, which has the same effect as regressing towards the mean. I've been doing that for years, because without it, it makes the odds for the front runners unrealistically high. Now if a teams W-L record would just perfectly reflect their DVOA, I think Mike Harris' playoff odds report and mine would converge.

56
by tom (not verified) :: Tue, 10/09/2007 - 6:34pm

Re 47:

I'm pretty sure that "approaching positive DVOA" is horribly dissapointing. They had a 22% rushing DVOA last year, and their unadjusted was higher so they faced above average defenses.

57
by Carlos (not verified) :: Tue, 10/09/2007 - 6:35pm

but the run wasn’t working.

wasn't working at what? All the run had to "work at" was keeping the clock running and keeping possession with the Bills for 3 more plays until the FG.

Of course, Lynch had just ripped off a 14 yard run, so 3 straight runs might have even resulted in more than 3 points.

58
by Jerry (not verified) :: Tue, 10/09/2007 - 6:39pm

Re #25:

The AFC (with the exception of the Jaguars and the Browns or Ravens, depending on how you see the franchise move) is essentially the expansion part of the NFL, created with the NFL/AFL merger. The three exceptions were Baltimore, Pittsburgh, and Cleveland, who were in the NFL, but switched conferences in the name of competitive balance.

Actually, there were 16 AFL teams and 10 NFL teams when the leagues merged. After a lot of discussion, the Steelers, Browns, and Colts agreed to move, accepting money from the league for moving to the "inferior" conference. There was a lot of comment at that point about how putting the Jets and Colts in the same division would create a tremendous rivalry in the wake of Super Bowl III; of course, that rivalry never developed.

59
by throughthelookingglass (not verified) :: Tue, 10/09/2007 - 6:40pm

54: as a top 3 back, MVP candidate in a normal season type player? (that's what I mean by elite, I should have defined that better)

60
by Jerry (not verified) :: Tue, 10/09/2007 - 6:43pm

Re #58:

there were 16 AFL teams and 10 NFL teams

Make that 16 NFL and 10 AFL. Sorry.

61
by Hector (not verified) :: Tue, 10/09/2007 - 6:47pm

51 - No, the play call was the problem. In theory, yes, Edwards should have thrown the ball away or eaten it. But theory doesn't always work - here you have an inexperienced QB playing under the bright lights of MNF against one of the best teams in the league. Stuff happens. Just run the ball, grind off another 40 seconds, and kick the FG to go up 11.

62
by Rich Conley (not verified) :: Tue, 10/09/2007 - 6:52pm

56.

"I’m pretty sure that “approaching positive DVOA� is horribly dissapointing. They had a 22% rushing DVOA last year, and their unadjusted was higher so they faced above average defenses."

I'm not saying its not dissapointing so far, but they've played 3 VERY GOOD run defenses so far, and not done much. They've shredded two bad run defenses. If they shred a third one this weak, and it puts them back around zero, then theres reason to believe they end up well into the positive at the end of the season.

When I said "for the season" I mean "for the season up to and including the game next week"

63
by Will Allen (not verified) :: Tue, 10/09/2007 - 6:53pm

Rich, the Bills didn't need anything more to "work" than a field goal at that point to win the game, and they were close enough to make the field goal highly probable. Hell, they could've done a kneel-down, run another thirty five seconds off the clock, and it would have been more intelligent than allowing the inexpereinced qb to make another decision. When a 28 yard field goal wins the game, then you run time off the clock, and kick the field goal. Period. You don't throw the ball.

64
by Rich Conley (not verified) :: Tue, 10/09/2007 - 6:57pm

"51 - No, the play call was the problem. In theory, yes, Edwards should have thrown the ball away or eaten it. But theory doesn’t always work - here you have an inexperienced QB playing under the bright lights of MNF against one of the best teams in the league."

And lynch could have fumbled the ball. You have an inexperienced running back playing under the bright lights of MNF against one of the best teams in the league.
Edwards had thrown what, 35 passes at that point with no picks?

Bad things DO happen. Defenses FORCE turnovers.

We're harping over this stuff because they lost. If edwards completes that pass for a first down, the bills win that game 31-16, and everyone is saying Jauron pulled a masterful upset.

65
by Admore (not verified) :: Tue, 10/09/2007 - 7:00pm

Houston is #1! #1, baby!

#1 in SPECIAL TEAMS. Read and WEEP Patriots and Colts.

228 yards of FGs! Yeah!

Seriously though, just happy to be in the top half with our injury situation. Also, Kris Brown is a kicking god. And his name? Very easy to spell. That's important.

66
by Gerry (not verified) :: Tue, 10/09/2007 - 7:04pm

"If the play wasn’t there Edwards should have thrown it away or eaten the ball."

True. But NFL head coaches are supposed to be aware of the fact that NFL rookie quarterbacks are not known for making good decisions in this manner, not having fully adjusted to the fact that they can't get away with what they used to be able to get away with in college and not fully adjusted to the speed of the pro game.

Then again, it is not like the Bills were going anywhere this year. Might as well let the kid learn from experience.

67
by Will Allen (not verified) :: Tue, 10/09/2007 - 7:09pm

Well, Rich, if Lynch is too inexperienced in terms of ball security, then you do a kneel down and kick the field goal. Really, though, we see this situation in overtime with frequency. Once the ball gets to about the 10 yard line, teams stop throwing, even with experienced qbs, because it doesn't make any sense.

68
by Temporarily#24inNBC100KChallenge (not verified) :: Tue, 10/09/2007 - 7:11pm

The team I'm having trouble understanding in both DVOA and the playoff odds report is TB.
The Bucs were stomped in both of their losses, and while it's not embarrassing to be stomped by the Colts, it's strange that they are ahead of a Seattle team who has only been stomped once and has a stomp victory over the Bucs.
The Bucs on the other hand have three stomps, which is vital to DVOA success, but all came over terrible opponents. The Bucs have defeated the 26th, 29th, and 31st ranked teams.

The playoff odds report claims to assign wins "based on a formula that considers the current DAVE ratings". Does this formula take into consideration the propensity of certain teams to play close games? FO has pointed out at length the fallacy in the idea that a team's quality is shown in the ability to win close games (oh, to be a Detroit fan like myself). Close games are pretty close to being random.
When the playoff formula assigns wins and losses, it would need to deviate from DAVE in a certain way in order to accurately predict that good teams like Baltimore and Jacksonville are likely to lose more games than DVOA projects and lousy teams like K.C. and S.F. are likely to win more games than DVOA projects simply because of the way they manage games.
(Good teams should want to maximize the pace of games during the first half while they build their leads, then minimize the pace during the second to reduce player wear-and-tear. Bad teams like K.C. and S.F. want to slow the pace as much as possible all game in order to secure a few fluke victories.)
When you consider the ability of T.B. to build up early leads would seem permanently compromised - if it actually existed to begin with - by the rash of offensive injuries, the close game phenomenon will rob T.B. of a number of wins going forward as they become a defensively-oriented, ball control team.
Even playing in a historically bad division, it's hard to see T.B. as a 90% playoff team. Their most likely record over the next 6 games is no better than 2-4. They have 4 home games against top 13 DVOA teams, and considering their injuries and previous performance against top 10 DVOA teams, it's hard to like them in any of those games. Even if they win one of them, there's a strong likelihood that they lose either at Detroit (which has a very good DVOA at home on the carpet) or at Atlanta.
Following that stretch, they'll be 5-6 and possibly still in first place in the AFC South, but the 10 win estimation would seem out of reach.
Great info by the outsiders as always.

69
by Carlos (not verified) :: Tue, 10/09/2007 - 7:12pm

Rich - your analysis is hopeless.

Could lynch have fumbled? Sure. But what is the ratio of fumbles lost:carries vs. INT:Attempts. It guess the difference is just barely within an order of magnitude.

Plus, with the run, you guarantee to keep the clock running.

Plus, you overlook Jauron's terrible play calling at the end of the first half. You can call it hindsight if you want, but I was there at the time telling my TV what stupid decisions those were.

As a fellow Eli, I wanted to see Jauron succeed, but you're probably a an all-time terrible coach if you get 6 turnovers from the opposing QB plus three return TDs and still lose, and, you know what, Jauron, at least yesterday, was an all time bad coach.

70
by Fergasun (not verified) :: Tue, 10/09/2007 - 7:14pm

I think you guys are talking about kneel to win... the Bills should've played kneel to win.

Rich, thanks for turning the FOMB into your own personal version of Crossfire, every assertion must be challenged by Rich, the brave contrarian, speaker of truth... maybe you can somehow turn discussion of the MNF game into a Colts-Patriots argument...

Denny Green mode
Dick Jauron is who we thought he was, he might've had one good season, but Dick Jauron is who we thought he was. He was sub-500 in Chicago, even with a 13-3 season, got fired from there, Dick Jaruon is who we thought he was. He coached the Lions defense, who thinks coaching the Lions defense isn't an indication of skill? It's BS... their defense didn't even improve when he was their coordinator... DICK JAURON IS WHO WE THOUGHT HE WAS!... what fan gets excited because Dick Jauron is their head coach?! Dick Jauron is who we thought he was.

71
by Dave (not verified) :: Tue, 10/09/2007 - 7:15pm

a lot of people like to question objective rankings, but i just wanted to say that your top 10 ranked teams (NE-JAX) are the exact same teams I'd have in my top 10, althouhg maybe in a slightly different order.

Says a fair bit for the validity of your statistics, IMO, as all your fancy formulas have somehwo come up with a graet rankings.

72
by countertorque (not verified) :: Tue, 10/09/2007 - 7:21pm

14 and 24: Let's not forget the 2004 regular season when the Steelers and their rookie QB couldn't possibly beat either the undeated Pats or the undefeated Eagles and certainly not in back to back weeks.

Just because the Pats and Colts haven't yet laid an egg does not mean that they won't lay an egg. Of course they will.

73
by mm (not verified) :: Tue, 10/09/2007 - 7:38pm

Dick Jauron. Yalie. Say no more.

And I point this out having grown up watching him dominate as the high school phenom he was.

74
by Fnor (not verified) :: Tue, 10/09/2007 - 7:43pm

#72: And, if you recall, the Steelers with their rookie QB got plastered in the postseason, while the other two looked like who (sorry) we thought they were in the regular season.

75
by chip (not verified) :: Tue, 10/09/2007 - 7:43pm

Who would have thought after 5 weeks that the six of the bottom 7 RBs out of 46 (min. 30 carries) would include:

R.Bush
M.Jones-Drew
S.Jackson
R.Johnson
F.Gore
C.Benson

That's five elite fantasy RBs and a #2 (C.Benson.)

76
by Felden (not verified) :: Tue, 10/09/2007 - 7:51pm

Re: 16
Yes, the 49ers offense really has been that bad.

I think the Defense is better than #22, but they're just on the field 40 minutes per game, which will knock them down a bit--they just get tired out eventually.

77
by NewsToTom (not verified) :: Tue, 10/09/2007 - 8:04pm

Re #68
I think the Bucs' 86% division probability has a lot to do with their high playoff probability. The NFC South this year looks like it could challenge the NFC West of recent vintage as being among the worst divisions in NFL history. They're also routing their pathetic foes and only have been really blown out against the Colts-the Seahawks won by 14, but it was a 1-score game until less than 8 minutes left. Meanwhile, the Saints game was even less close than a 31-14 score indicates, as the Bucs were up 28-0, the Rams were just destroyed, and Carolina scored a meaningless TD in the last minute to make it only 20-7. Yes, they'll almost certainly be worse with the offensive injuries. But, right now, we can't be sure just how much worse. Subjectively, I think 9-7 is more likely than 10 wins, and an 85% chance to win the division is about right. And it wouldn't surprise me if 9-7 is enough to get a wildcard berth, even if someone beats them out for the division.

78
by Rich Conley (not verified) :: Tue, 10/09/2007 - 8:07pm

"Could lynch have fumbled? Sure. But what is the ratio of fumbles lost:carries vs. INT:Attempts. It guess the difference is just barely within an order of magnitude."

IIRC, Modern running backs fumble about once in every 40-50 carries. QBs throw a pick about 30-40 attempts. Its pretty close.

The patriots were in the same situation sunday. Up 10, about 8 minutes left, and they threw. They scored a touchdown and put the game away. The only reason it was the wrong call for Buffalo was that it didn't work.

79
by countertorque (not verified) :: Tue, 10/09/2007 - 8:13pm

#74 I think comment #24 already addressed the post season. The Steelers can decisively beat either of those 2 teams. I'm not saying they definitely will. But, it's silly to be down on the #3 team because they aren't ranked #1.

80
by coldbikemessenger (not verified) :: Tue, 10/09/2007 - 8:17pm

In 2006
per team there were on average
398.7 rb carries
6.2 fumbles
3.8 fumbles lost

81
by Will Allen (not verified) :: Tue, 10/09/2007 - 8:19pm

First, Rich, given a 50/50 fumble recovery rate, the odds of LOSING a fumble drop well below the odds of an int. Second, eight minutes isn't five and half minutes, which is what would have been left after a run and a kick, so one doesn't adopt the same strategy at eight minutes as one does at five and a half minutes. Third, Tom Brady doesn't play for the Bills, and one would not expect to adopt the the same strategy with Brady behind center as one would with a very inexperienced qb. Yes, it was a wrong call because it didin't work, but also because it had a much better chance of turning sour due to who was playing qb, and because a field goal makes it highly unlikely the Cowboys can come back, and the chance of a successful field goal is extremely high.

82
by Kal (not verified) :: Tue, 10/09/2007 - 8:19pm

IIRC, Modern running backs fumble about once in every 40-50 carries. QBs throw a pick about 30-40 attempts. Its pretty close.

No, that's about the worst running back that you can have. That's Tiki Barber fumblitis. Most RBs fumble about once in 70 attempts. Here's some example stats of touches per fumble for RBs:

Priest Holmes-- 155.7
Curtis Martin--- 133.8
Corey Dillon---- 102.9
Marshall Faulk-- 101.5
LT Tomlinson--- 97.1
Fred Taylor----- 92.1
S. Alexander---- 88.0
Stephen Davis-- 84.6
Duce Staley---- 81.9
Clinton Portis--- 79.25
D. McAllister---- 74.8
Ahman Green--- 60.6
Ricky Williams*-- 53.5
Jamal Lewis----- 50.2
Travis Henry---- 41.8
Tiki Barber------ 37.1

RBs tend to be a far better proposition than QBs on turning the ball over per attempt - and they're especially better than a rookie QB.

83
by NewsToTom (not verified) :: Tue, 10/09/2007 - 8:27pm

It was clearly the wrong call because I predicted it in the FO IRC chatroom before it happened. As to mistakenly calling for a pick-6, well, I did that just to show I was human. Or I was wrong, take your pick.

84
by sam (not verified) :: Tue, 10/09/2007 - 8:30pm

So here's something interesting... the AFC South and NFC East both have their entire divisions inside the top-16.

85
by B (not verified) :: Tue, 10/09/2007 - 8:30pm

72: Just because they haven't laid an egg doesn't mean they will, either.

86
by jimm (not verified) :: Tue, 10/09/2007 - 8:44pm

Don't like piling on, but the call for Edwards to throw the ball twice inside the 15 with 6 minutes left up 8 was horrible.

Whenever I see a coach like Jauron with a lifetime record well below .500 I always suspect that he has no handle on the statistical strategies of the game.

87
by Will Allen (not verified) :: Tue, 10/09/2007 - 8:58pm

Yeah, I forgot he passes on first down as well. If they run on first and third down, there's actually only about 4:45 left, which seals the game to an even greater degree.

88
by jebmak (not verified) :: Tue, 10/09/2007 - 9:06pm

#30

That ain't mixing metaphors! This is mixing metaphors... (courtasy of Zapp Brannigan)

When we hit this bulls-eye the rest of the dominos will fall like a house of cards. Checkmate.

89
by jimm (not verified) :: Tue, 10/09/2007 - 9:23pm

I can't wait to see one of those last second time out calls by the coaches backfire when some kicker misses the first attempt that doesn't count, and makes the second one.

90
by Matt (not verified) :: Tue, 10/09/2007 - 9:24pm

78 - Will beat me to it, but I can't let it go. Next, Rich will tell us that (as the MNF guys said) Trent Edwards is just like Brady. Same diff. If you can trust Brady to throw to seal the game, then Jauron should trust his rookie. 79 - Right, I specifically didn't mention regular season "upsets" in my list, but those just add to the theory in my opinion. The Steelers beat down the Pats in 2004 reg season but got it handed back to them in the playoffs. Same deal in 2005, but this time it was the Colts in the regular season triumph and the Steelers in the playoffs. (Just to keep piling on with Steelers misery, one can go back to 1997 as well, where the Steelers beat the Broncos pretty handily in Week 15 only to lose the AFCC to them six weeks later.) In the end, I think it just proves that (no matter how good a predictor DVOA may be) you just don't know, and even mid- to late-season regular season results are not necessarily indicators of playoff results between the same teams.

91
by jimm (not verified) :: Tue, 10/09/2007 - 9:33pm

86. Will - of course we do realize there are teams capable of blowing an 11 pt lead with just 6 minutes or so left in the game.

As soon as Dallas missed the 2 pointer I thought of the Ariz/Minn in 2003.

92
by Dutch (not verified) :: Tue, 10/09/2007 - 9:54pm

Today’s DVOA commentary was going to talk in depth about the historic seasons that we are seeing in New England, Indianapolis, and Dallas.

Ok, in all due respect to this statement above. New England has absolutely not been tested. Their opponents are giving up an average of 6.1 yards per play. I know New England has put up 30+ points in most of their games. But all of the teams they have played have given up a large amount of points to other opponents as well. Its not like New England has put up 38 against Pittsburgh , when all other teams are averaging less than ten points against them. Lets not start making these kinds of statements just yet footballoutsiders.

93
by PanthersnBraves (not verified) :: Tue, 10/09/2007 - 9:54pm

MNF - Somehow, I had the feeling that the Bills had managed to dodge the bullet when they got the goalline interception, but no...

How on the series before, Jauron didn't have his offense run the ball twice and then on third down have the sneak to the center of the field for the field goal is beyond me.

The offense only managed a field goal all game, why have a rookie QB making that throw?

Instead, I have to listen to the 'boys fans at work crow about 5-0, while I have to read up on Matt Moore.

94
by Scott de B. (not verified) :: Tue, 10/09/2007 - 10:07pm

Ok, in all due respect to this statement above. New England has absolutely not been tested. Their opponents are giving up an average of 6.1 yards per play.

Thus the "D" in "DVOA". That's why they go from 75.7% to 67.7%. Even if you double the opponent adjustment (so they end up around 60%), they just drop from 2nd to 3rd on the all-time-after-five-games list.

95
by tom (not verified) :: Tue, 10/09/2007 - 10:17pm

Re 68:

I'm pretty sure stomps don't have anything to do with making the playoffs, it's a indicator for success when in the playoffs.

96
by Dutch (not verified) :: Tue, 10/09/2007 - 10:56pm

93
"Thus the “D� in “DVOA�. That’s why they go from 75.7% to 67.7%. Even if you double the opponent adjustment (so they end up around 60%), they just drop from 2nd to 3rd on the all-time-after-five-games list."

Patriots opponents defenses give up 27 points a game on average thru 5 wks. The Steelers defense gives up like 8. YOu may want to 3.5 times the opponent adjustment.

All you need to do to score against the Patriots is throw the football with dencet TE's at those aging Linebackers and Rodney Harrison. Put those guys in coverage all day.

97
by MarkB (not verified) :: Tue, 10/09/2007 - 11:00pm

San Francisco dead last - Mmmm... the Belichick mind-f*ck is working its magic, and that draft choice is looking better and better. :-)

98
by Carlos (not verified) :: Tue, 10/09/2007 - 11:01pm

80, 81 and 82 are great examples of why this site has the best football discussion anywhere.

78: Made up data and misapplied to boot

80, 81 & 82 stomp out the nonsense

Love it.

99
by Chris (not verified) :: Tue, 10/09/2007 - 11:05pm

68 good post.

100
by thestar5 (not verified) :: Tue, 10/09/2007 - 11:07pm

91,

Exactly. All it takes is one bad game (like the 'Boys had yesterday, although luckily they won) and we won't have to hear this talk about how unbelievable Indy and NE are.

101
by DGL (not verified) :: Wed, 10/10/2007 - 1:56am

Fnor:

Lines quoted for which no apology is ever necessary:

"They are who we thought they were."

"Crown their asses."

And, of course, the immortal: "You PLAY to WIN the GAME."

102
by Pat (not verified) :: Wed, 10/10/2007 - 2:19am

Bad things DO happen. Defenses FORCE turnovers.

Teams can go too conservative, but a defense can't force a turnover if the offense simply kneels. With a rookie QB in the game, there's no way a short pass like that should've been called. Inside the 20, it should've been run, run, run, kick the field goal. Period. They needed the time off the clock, and a FG would make it two scores (and likely two touchdowns!).

Jauron made a similar mistake again later on the next drive, when Edwards threw on third down, with less than 4 minutes to go! You're not going to convert 3rd and 7 with a good percentage. Your chance of winning improves more by running the ball and bleeding time off the clock.

Then, of course, the defense on the 8 yard pass on the second-to-last play of the game was just inexcusably bad. The linebackers weren't just in the middle of the field, they were tight in the middle of the field. Just absolutely no point.

103
by Scott C. (not verified) :: Wed, 10/10/2007 - 2:59am

96:

Exhibit A: The D in DVOA isn't about adjusting to the best team in the league in points allowed. Its adjusting to the Average, hence the A.

Exhibit B: Points Allowed is NOT A DEFENSIVE METRIC. It is a team metric. See http://www.footballoutsiders.com/pregame.php

Turns out, when a QB throws a pick-6 (or a pick-return to 1 yard line), its not the defenses fault! Who'd a thunk it?

Opponent adjustments are at half strength. Things are wacky in the early part of the year. Yes the Pats have faced weak defenses. So what? All those other teams on that list also faced weak opponents to start.

104
by Alex (not verified) :: Wed, 10/10/2007 - 3:08am

IIRC, Modern running backs fumble about once in every 40-50 carries. QBs throw a pick about 30-40 attempts. Its pretty close.

As mentioned above, it's not even remotely close. Just to put it in perspective, Donovan McNabb has one of the lowest interception rates in league history, at one interception every 46.5 attempts. Tiki Barber (who had chronic fumblitis), in his worst season in terms of losing fumbles, 2003, ran the ball 278 times (in addition to catching 69 passes) and had 6 fumbles lost. Even assuming all of his lost fumbles were on runs, his fumble lost rate would still only be one in every 46.3 carries.

105
by steelberger1 (not verified) :: Wed, 10/10/2007 - 7:45am

Pittsburgh is clearly ranked too low because Steely McBeam kicks ass. My opinion is way better than this. Steely McBeam could kick Pat the Patriots ass up and down the field.

106
by steelberger1 (not verified) :: Wed, 10/10/2007 - 7:53am

And who the hell is this DAVE guy?

107
by slo-mo-joe (not verified) :: Wed, 10/10/2007 - 8:24am

96:
All you need to do to score against the Patriots is throw the football with dencet TE’s at those aging Linebackers and Rodney Harrison. Put those guys in coverage all day.
As SD with Gates and the Browns with Winslow have shown, that is clearly not all one needs to do. But of course, never let facts get in the way of visionary coaching.

108
by Sergio (not verified) :: Wed, 10/10/2007 - 9:25am

Woe is me for driving this discussion near Conley's hands again... but... I can't resist.

The problem I saw with Buffalo on Monday was that they simply didn't adjust to the crowded, 8-in-the-box defense the Cowboys put out there at the end. I agree that in the final play, the best possible call would've been a run up the middle or even a kneel-down, but the team shouldn't have been in a 3rd and long situation there (and they were in several points of the game because of the same reason) to begin with.

In other news, the sun is bright and water is wet.

109
by mawbrew (not verified) :: Wed, 10/10/2007 - 9:31am

Nice to see that Braylon Edwards is getting some DPAR respect this year. I haven't seen all of Cleveland's games this year, but Edwards seems to have eliminated (or at least greatly reduced) his drops and is consistently making stunning catches. The sideline grab that Bellichick challenged (and lost) against the Pats was just one of several this year.

Edwards catch rate (56%) hurts his DVOA. But when you consider that his rate is actually above Anderson's overall completion rate (54%) along with his 20+ YPC number, it's pretty impressive.

110
by Ben (not verified) :: Wed, 10/10/2007 - 10:10am

Looking at the offensive DVOA page, I'm surprised to see that the Colts rushing DVOA is almost twice as high as the second place Pats. Is the Colts running game really that good this year? Is that sustainable play? I'm assuming it has something to do with the fact that they are number one in power success rate and have the lowest stuffs rate under the O-Line stats. Both of those are surprising to me. The Colts O-Line has been very good at keeping Peyton upright over the years, but they've never been known for blowing people off the ball...

111
by Starshatterer (not verified) :: Wed, 10/10/2007 - 10:41am

The Colts O-Line has been very good at keeping Peyton upright over the years, but they’ve never been known for blowing people off the ball…
No, but defenses keeping 7+ guys back in coverage looks pretty much the same, from the runner's point of view.

112
by Ben (not verified) :: Wed, 10/10/2007 - 11:02am

Even on Third and Fourth and 2 or less?

113
by Rich Conley (not verified) :: Wed, 10/10/2007 - 11:12am

"Yeah, I forgot he passes on first down as well. If they run on first and third down, there’s actually only about 4:45 left, which seals the game to an even greater degree. "

It does? IIRC, the cowboys got the ball back with roughly 2:00 left and still won. I don't think a field goal would have sealed the game at all.

114
by Rich Conley (not verified) :: Wed, 10/10/2007 - 11:15am

112.

yes, even in 3rd and 4th and less than two. I've never seen teams play nickel in that situation before, but teams are doing it against Peyton.

115
by Ben (not verified) :: Wed, 10/10/2007 - 11:16am

re:112 To follow up to my own post, I was more curious as to what had changed this year. The previous years Power Success rate was in the bottom quarter of the league, including dead last in 2004 when Manning was chucking TDs all over the place.

116
by Rich Conley (not verified) :: Wed, 10/10/2007 - 11:18am

82

So Marshawn Lynch is now among the ranks of players like Curtis Martin, Priest Holmes, etc? These guys were the best of the best. Hes a nondescript rookie. W

117
by B (not verified) :: Wed, 10/10/2007 - 11:35am

115: What changed is they got an athletic Left Tackle who excels at run blocking, but struggles in pass protection.

118
by B (not verified) :: Wed, 10/10/2007 - 11:37am

116: Lynch has 100 rush attempts and zero fumbles, so assuming he will fumble the ball once every 30-40 times is unfounded.

119
by Will Allen (not verified) :: Wed, 10/10/2007 - 11:49am

Yes, Rich, so if one doesn't have confidence in Lynch's ball security skills, you put another running back on the field, or simply do a kneel down. Also, please reflect on the difference between the phrase "seals the game to a greater degree" and the phrase, "seals the game completely". A team down by eleven with 4:45 left needs to score a touchdown, get the two point conversion, AND get the onside kick, then get a field goal, or needs to score a touchdown, likely needs to get the onsides kick, and then another touchdown. Given Romo was still throwing multiple passes that were or should have been intercepted in the last five minutes of the game, that was more unlikely than the Cowboys getting a td and an onsides kick followed by a field goal in the last two minutes. Factor in that if the Bills kick the field goal, and the Cowboys get the td, two pointer, and field goal, the Bills still haven't lost, and it becomes even more clear the the smart play on % of success basis is for the Bills to bleed the clock and kick the field goal.

120
by QBeam (not verified) :: Wed, 10/10/2007 - 11:52am

Regarding the great Jauron debate, I think Rich (113) has made the stake-through-the-heart point. I'll donate a cookie to the charity of their choice to the person who can identify the last time a team with less than a 14 point lead started kneeling on the ball with 5 minutes left to play. Running up the middle without pretense, sure, but kneeling?

The only argument in support of this strategy I've heard so far is that the QB was a rookie, and so couldn't be trusted to throw the ball. It's obviously true that it would be silly to confuse a rookie with, say, Tom Brady. But I think you must really be emotially tangled up with a heart-breaking loss to say that it makes sense for a coach to effectively bench his rookie QB like that. Rookies become veterans precisely by being given the chance to play. Sending the message, loud and clear, that you don't think he can be trusted with more than a TD lead with five minutes to play is not a good recipe for future improvement.

121
by Rich Conley (not verified) :: Wed, 10/10/2007 - 11:53am

Again Will, its only being second guessed because they lost. He completes that pass, and hes a hero. This entire argument is based on hindsight.

122
by Chris (not verified) :: Wed, 10/10/2007 - 11:54am

Peyton doesn't call run plays when he sees 8 in the box. This puts his RBs in favorable situations so they won't get as many -1,0,1 yard runs.

123
by Rich Conley (not verified) :: Wed, 10/10/2007 - 11:55am

118.

Saying that 100 carries is a representative sample is completely unfounded.

124
by Don Booza (not verified) :: Wed, 10/10/2007 - 11:55am

I'd like to rename this EP the "Rich Conley Beat-Down". Sorry if this name has already been used for a dozen other topics at FO.

125
by Chris (not verified) :: Wed, 10/10/2007 - 11:59am

I hate playing MMQB. If the offense runs the ball and loses ( they should have passed), if they passed and lost ( they should have ran).

126
by AmbiantDonkey (not verified) :: Wed, 10/10/2007 - 11:59am

How do the Packers squib kickoffs effect the Bears special teams DVOA? Is it positive because of the field position or negative because of the short returns?

127
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:04pm

124.

How bout "Don Booza adds nothing to the discussion"

We already have some of those, right?

128
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:04pm

At the risk or being pointlessly redundant, Rich, one also cannot ignore the fact that the Bills offense hadn't done much of anything all night against the Dallas defense. The odds of this Bills offense having red zone success, in terms of scoring a touchdown, against this Dallas defense is not good at all. Adopting a strategy that ignores the fact that the other guy's personnel is just flat out better in a particular matchup is never a good idea. What would you say about Wade Phillips if, next Sunday, he decided to play Randy Moss man to man with the ball at midfield with the Patriots behind 35-31, and two minutes left in the game, after the Cowboys hadn't had any success blitzing Brady for the previous 58 minutes, and Moss already had 180 yards and two touchdowns?

129
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:05pm

"ou put another running back on the field, or simply do a kneel down"

Have you ever seen a team kneel with 5 minutes left in the game Will? I haven't.

130
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:09pm

"Adopting a strategy that ignores the fact that the other guy’s personnel is just flat out better in a particular matchup is never a good idea."

Will, if you want to say that, they shouldn't have thrown a pass in the entire second half. They should have kneeled 3 times and punted each drive. Why is it okay that they threw from the 20-30, but not at the 13 yard line?

And there was six and a half minutes left, not 4. At most you take it down to 5:45.

131
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:12pm

Actuallly it isn't Rich, given I was saying at the time that the Bills should bleed the clock and kick the field goal. Look, by your reasoning, one could simply say that playcalling and clock strategy are irrelvant to discussing football, since if something works, it never becomes an issue. The fact is that there are some strategies in regards to time management, likelihood of a successful particular play, and the personnel matchups, which provide a significantly greater chance of winning a game than other strategies. Good coaching doesn't always win the game, but it almost always involves adopting the strategy which has the greatest likelihood of winning the game on a percentage basis.

132
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:17pm

"Look, by your reasoning, one could simply say that playcalling and clock strategy are irrelvant to discussing football, since if something works, it never becomes an issue."

no, thats not what I'm saying Will. What I'm saying is that short passing in that situation is VERY common, and works most of the time. Generally teams play the run, so short passing is MORE successful.

Buffalo got the ball back after that, with the score STILL THE SAME, and pretty much lost the game because they ran on 1st and 2nd and it set up a 3rd and long that wasn't succesful. Was that the right call?

133
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:18pm

". Good coaching doesn’t always win the game, but it almost always involves adopting the strategy which has the greatest likelihood of winning the game on a percentage basis."

Right, I agree. Continuing the drive is generally what gives you the highest winning percentage. They get a first down there, and the game is a whole lot more "sealed" than if they kick the field goal.

134
by Al 45 (not verified) :: Wed, 10/10/2007 - 12:24pm

Rich:

Sorry, but this isn't hindsight. I know I was watching the game and, before the play, turned to my friend and said, "QB sneak to the middle, or RB dive... let time run off the clock, kick a chip shot field goal, and the game is pretty much sealed."

That's not hindsight, it's common sense. If Brady were behind center, and we were talking about the Patriots, then yes, I wouldn't have had a problem with a pass play. Brady is infinitely more experienced and has shown a much greater ability to make quality reads. Trent Edwards hasn't shown any of that.

You're at under 6 and a half minutes to play and up by 8. A field goal makes it 11, which means the other team needs 2 touchdowns to beat you, or a touchdown, two point conversion, and a field goal just to tie the game.

It's a no brainer. It's not hindsight.

[quote]It does? IIRC, the cowboys got the ball back with roughly 2:00 left and still won. I don’t think a field goal would have sealed the game at all.[/quote]

The Cowboys got the ball back with a little over 2 minutes left and only down by 8 points. If the Bills run the ball, it's likely less time on the clock and up by 11 points. That means they would have had to have scored a touchdown with under 20 seconds left (since the two point conversion failed) in order to pull out the win.

So, yes, a field goal there would have all but sealed the game.

The play call to have Trent Edwards drop back and pass when they are in chip shot field goal range (for Lindell), was absolutely inexcusable. It would have been wrong even if he completed the pass. The reward wasn't worth the risk.

135
by Al 45 (not verified) :: Wed, 10/10/2007 - 12:26pm

Rich:

Buffalo got the ball back after that, with the score STILL THE SAME, and pretty much lost the game because they ran on 1st and 2nd and it set up a 3rd and long that wasn’t succesful. Was that the right call?

The score was still the same because they stupidly threw the ball and got it intercepted. When they got the ball back, they were about 20-25 yards worse in field position, which meant they didn't have an easy opportunity to put points on the board.

I'm sorry, but with an inexperienced QB, with an 8 point lead and under 6 and a half minutes left to play, the smart call (and definitely not hindsight) is to keep the ball on the ground (whether a QB sneak or RB dive) and get 3 points to make it a two score game.

Jauron blew it.

136
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:26pm

Rich, kicking a field goal from the 30 is significantly different than doing so at the the 13, and taking time off the clock with runs has a much higher value when up by 8, with six minutes left, with a field goal attempt of 30 yards in the future, than the value of taking time off the clock with runs when there is, say, eight minutes left, and a field goal attempt of 45 yards in the future. The difference between six minutes left in the game, and eight minutes left in a game, and between a 30 yard field goal, and a 45 yard field goal is significant. Even this ignores the fact that as the red zone approiaches, a defense which has superior personnel to the opposing offense has an easier time defending. This is especially true with regards to the Cowboys, given that their primary defensive weakness is saftey play in deep pass coverage.

Finally, Rich, I do believe the Bills ran on first down as well, throwing incomplete. If they run instead, the clock is below five minutes when the field goal is made.

137
by Al 45 (not verified) :: Wed, 10/10/2007 - 12:28pm

[quote]Continuing the drive is generally what gives you the highest winning percentage. They get a first down there, and the game is a whole lot more “sealed� than if they kick the field goal.[/quote]

Even if they got the first down, the chances of them turning that into 7 points was minimal, as they hadn't shown any proficiency, other than one drive, at moving the ball all night.

Basically, that call came down to risking nearly guaranteed points for about another minute and a half to two minutes of clock. Sorry, but most people take the points.

138
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:32pm

Rich, you once again completely ignore the fact that the odds of the Bills being successful in continuing the drive in the red zone against the Dallas defense is not good at all, for the simple reason that the Dallas defense has significantly better personnel than the Bills offense. Making likelihood of success calculations which ignore the personnel who are on the field is the epitome of bad coaching.

139
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:37pm

Again, Rich, your "generally" ignores the players who are actually on the field. Why in the name of Paul Brown do you thing it is wise to strategize in a manner which ignores the personnel who are acutally on the field?

140
by Mr Shush (not verified) :: Wed, 10/10/2007 - 12:38pm

"They’ll go into their bye at 3-3 with two consecutive stomps, and get to come out against Houston."

Wait, they "get" to come out against a team with a superior record and DVOA, who have played over half the season so far without two of their best three offensive players, both of whom are expected to be healthy by then, and with only two healthy receivers? Who have the league's best special teams despite playing two and a half games without either of their two legitimately elite return men? Lucky Chargers.

In other news,

A bunch of NFC teams are clearly ranked too high because the opponent adjustments aren't full strength and they've only played one game against the Patriots and Colts combined. Looking at DVOA and making minor tweaks based on your own informed subjective judgment is way better than this. CowG!RLZ SUKC!!! TExAsn starters hamerd them in weak 3 of pre season with Raging Bulls d-fense and Andre Johnson + Jaccoby JOnes destroying stupid midget Americas' JOKE corners. GO TEXANS!

Ahem. When your team has a winning record after 5 games for the first time ever, it's easy to get carried away.

141
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:48pm

138.

No Will, I'm not ignoring that. If you want to make that argument, they should have kneeled the ball 3 times on the 27 yard line and kicked.

Buffalo was moving the ball well on that drive, and there was no reason to think that 3rd and 8 was any more risky than any other one.

142
by Alex (not verified) :: Wed, 10/10/2007 - 12:52pm

So Marshawn Lynch is now among the ranks of players like Curtis Martin, Priest Holmes, etc? These guys were the best of the best. Hes a nondescript rookie. W

No, he's not Curtis Martin or Priest Holmes, but he'd have to be worse than early-career Tiki Barber to turn the ball over at a rate as high as the even the best QBs in the league. Running is much, much safer than passing.

143
by Rich Conley (not verified) :: Wed, 10/10/2007 - 12:52pm

"Wait, they “get� to come out against a team with a superior record and DVOA, who have played over half the season so far without two of their best three offensive players, both of whom are expected to be healthy by then, and with only two healthy receivers?"

Mr Shush, its 5 weeks into the season. The Chargers have played 2 of the top 6 teams, and played Chicago before their entire defense got hurt. Theres reason to believe their DVOA isnt representative of the actual quality of the team. Houston has played Indy, and 4 awful teams. Theres reason to believe their DVOA isn't representative either.

DAVE is still more accurate than DVOA at this point, and San Diego is higher. The game is also IN San Diego. If San Diego isn't a 7-10 point favorite in that game, I'd be surprised.

144
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:54pm

Apparently, Rich, you are not reading my posts, so there is little reason to continue. No, kneeling three times with eight minutes left, to kick a 45 yard field goal, against a defense whose personnel superiority will become more pronounced as the goal line approaches, is not the same thing as kneeling three times with six minutes left to kick a 30 yard field goal.

Have a nice day, Rich.

145
by throughthelookingglass (not verified) :: Wed, 10/10/2007 - 12:55pm

140: We played them in preseason. Who the hell takes the third game of the preseason game like it's bullshit? We played them the third game. Everybody played three quarters. The Cowboys are who we thought they were!

146
by Al 45 (not verified) :: Wed, 10/10/2007 - 12:57pm

Buffalo was moving the ball well on that drive

Not really. The drive consisted of 8 plays, including the interception. There was a run play for 2 yards, pushed another 5 because of a face mask penalty. A swing pass for 4 yards, another 2 yard run, a dump pass for 8 yards, a good Lynch run for 14 yards, incomplete, 2 yards, interception.

The pass plays that 'worked' were all dump or swing plays. Not pass plays into a tight zone (which it was as they got further down the field) 8+ yards down the field.

Also, they never had a 3rd down until that point and their best play came on a run.

Only one play on that drive had gained enough yards to convert a 3rd and 8 and it was a broken play where the ball was dumped off.

Sorry, but it just wasn't the right call.

147
by Will Allen (not verified) :: Wed, 10/10/2007 - 12:59pm

Rich, I'll actually end our little conversation by noting that you just seemed to imply that third and eight on the 13 has the same odds of being converted as third and eight on the 27, against a defense with generally superior personnel, and whose primary weakness is safety play in deep pass coverage.

148
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:00pm

144.

No Will, its not, but is 15 yards worth the risk of losing the ball, against an obviously superior opponent?

You're arguing its not.

Whats the difference in field goal percentage at 30 yards vs 45 yards?

Why should I be bothered to read your posts Will, when you obviously don't read mine before making a retort.

149
by RickD (not verified) :: Wed, 10/10/2007 - 1:02pm

re: 86
One of those aging LBs had two picks on Sunday. Perhaps you've heard of this Mr. Seau?

You will also see next to him on many plays somebody named Thomas who came over from the Ravens during the offseason. Neither Seau nor Thomas played in last year's AFC Championship game. Nor did Harrison for that matter. Nor did rookie safety Brandon Merriweather, not that he's climbed that far up the depth chart yet.

But OK, the Pats are still soft in the middle. Right...

150
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:03pm

"Rich, I’ll actually end our little conversation by noting that you just seemed to imply that third and eight on the 13 has the same odds of being converted as third and eight on the 27,"

No, Will, I am not. Could you please stop putting words in my mouth and actually read the post.

What I am implying is that the odds of TURNING THE BALL OVER is not significantly higher at the 13 than it is at the 27.

151
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:04pm

"But OK, the Pats are still soft in the middle. Right…"

yeah, they're so soft that their Pass DVOA is what, -25%?

152
by Will Allen (not verified) :: Wed, 10/10/2007 - 1:06pm

Uh, Rich, you just ignored the difference between eight minutes left and six minutes left. Look, if you don't think a 30 yard field goal is the same as a 45 yard field goal, or, presumably, that eight minutes left is the same as six minutes left then don't write....

"No Will, I’m not ignoring that. If you want to make that argument, they should have kneeled the ball 3 times on the 27 yard line and kicked."

....because that is a statement which ignores the difference between a 30 yard field goal and a 45 yard field goal, and the difference between six minutes left and eight minutes left.

153
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:08pm

Had they kneeled the ball 3 times on the 27 Will, the clock would have been about 5:00 when they kicked.

154
by Will Allen (not verified) :: Wed, 10/10/2007 - 1:13pm

Well, then Rich, you seem to have simply ignored the difference in the likelihood of success between the Bills converting a third and eight from the 1l, against the Cowboys, as opposed to making the same sort of conversion from the 27, and the likelihood of conversion plays a large role in deciding whether to risk a pass at all. That would be another severe mistake.

Also, the odds of an int do probably climb in that situation, given the inexperience of the qb, and having more defenders crammed into a smaller space.

155
by mawbrew (not verified) :: Wed, 10/10/2007 - 1:13pm

Re: 120 - "Sending the message, loud and clear, that you don’t think he can be trusted with more than a TD lead with five minutes to play is not a good recipe for future improvement."

I think this is a fair point, but I'm not sure that allowing him to throw an interception that is charactterized (fairly or unfairly) as costing his team a game that their defense had won is a good recipe for further development either. There are risks either way.

Re: 126 - "How do the Packers squib kickoffs effect the Bears special teams DVOA? Is it positive because of the field position or negative because of the short returns?"

That's a terrific point. I think the assumption in the FO analysis is that length of opponents kickoffs is something that teams have no influence on. At least, I think I've read that here before. But obviously in this case Hester clearly did influence those kicks. So I would think that (if handled normally) the Packers special teams would be docked for the short kicks/bad field position but that the Bears special teams wouldn't benefit from it. Though they may make an exception in this case.

156
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:17pm

"Well, then Rich, you seem to have simply ignored the difference in the likelihood of success between the Bills converting a third and eight from the 1l, against the Cowboys, as opposed to making the same sort of conversion from the 27, and the likelihood of conversion plays a large role in deciding whether to risk a pass at all."

No, Will, I am not ignoring it. I am saying that the difference isn't significant enough that you need to go into marty-ball.

If you want to insist that it was the bad decision, show it. Prove it. Give me the expected points of passing and running from that situation.

157
by Will Allen (not verified) :: Wed, 10/10/2007 - 1:20pm

No, looking at the synopsis of the drive above, there would have been one less running play in the drive, then, so there would have been more than five minutes. The difference would not be as large as I posited, however. About 45 seconds instead of two minutes. Even so, the Bills attempting to get another first down from the 27 is significantly different than trying to do so from the 13.

158
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:25pm

"Even so, the Bills attempting to get another first down from the 27 is significantly different than trying to do so from the 13"

Yes, but the expected points is significantly higher, IMO. Seven points puts the game completely out of reach. Three doesn't.

What's the chances Edwards throws a pick there? 1 in 20? So a 5% chance of turning it over throwing vs a 2% chance of turning it over running. Is a 3% chance of turning the ball over a big enough risk that you should give up a significant chance at gaining 4 additional points?

159
by mawbrew (not verified) :: Wed, 10/10/2007 - 1:26pm

Re: 156 - "Give me the expected points of passing and running from that situation."

Also please adjust for the presence of a rookie QB, an above average defense, and the presence of Jerry Jones in the owners box.

Sheesh.

By the way, I was screaming for Cleveland to run the ball (rather than risk the pick) on third down and goal during their first drive vs. the Pats. Sometimes it just seems inevitable that the interception will occur.

160
by Carlos (not verified) :: Wed, 10/10/2007 - 1:29pm

I count 31 or so posts by Rich Conley out of 156 posts overall. Including gems such as:

Again Will, its only being second guessed because they lost.

When you just keep repeating the same point, but not persuading anyone, why would you think the solution is more repetition?

What makes you think what you have to say is worth making 30+ posts over, especially when you're not saying anything new in the latter 28+ posts?

161
by Mr Shush (not verified) :: Wed, 10/10/2007 - 1:33pm

Rich, I know SD are almost certainly better than their DVOA. Thing is, so are Houston. As I said, Green and Johnson have missed over half the season to date, but are expected to return well before the San Diego game. As for DAVE, it may be more accurate for most teams, but the projection system simply had no way of knowing that Schaub would be as good as he is, or that Okoye would have such an immediate impact. The Chargers may well be favoured, and if Norv gameplans to attack the various glaring weaknesses in the Houston secondary and Rivers executes, they could win big. But it's not as if all SD's problems are attributable to SOS. They lost at home to the Chiefs, who Houston beat handily. DVOA suggests the Falcons and Dolphins are significantly better than their records. The Chargers' variance is gargantuan and their coaching is attrocious, so I wouldn't rely on them beating anyone, and certainly not a genuinely above average team.

162
by Will Allen (not verified) :: Wed, 10/10/2007 - 1:39pm

No, Rich, I'm don't have the data at my fingerprints to prove that the #30 DVOA offense, with a rookie qb who hasn't played the entire year, has a significantly better chance of converting a third and eight from the 27, as opposed to a third and eight from the 11, against the #6 DVOA defense. I have noodled around a little, and it appears that a thirty yard field goal has a chance of being converted somewhere in the mid to high 90s, whereas as 45 yard field goal is somewhere in the 65-70 percent chance of being converted.

163
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:44pm

Will, I'm not arguing they should have kneeled the ball from the 27, although you seem to think I am. I'm arguing that going for it on third down wasn't the wrong call. It just didn't work.

You're arguing that it is extremely risky to throw the ball there. I'm arguing that it may not be so risky that its not worth the chance at 4 more points.

164
by Will Allen (not verified) :: Wed, 10/10/2007 - 1:48pm

Actually, Rich, in a situation where the best running back on the roster in terms of ball security is on the field, and he has been told to maintain ball security above all else, I suspect the odds of a fumble are around .5% at most. Again, one can always do three kneel downs, and take 98% chance of making the filed goal, leaving less than five minutes on the clock. I also suspect that the odds of a rookie qb with a lousy offense, against a good defense whose pronounced weakness is deep pass coverage by the safeties, has a significantly greater than 5% chance of throwing an int on a third and eight from the 11. This is not mean situation, and it is a mistake to calculate the odds as if it were.

165
by John (not verified) :: Wed, 10/10/2007 - 1:48pm

Can one use the DVOA rankings to help
place "friendly" wagers against the line
for the upcoming week of games.

An example would be appreciated.

Thank you

166
by Kal (not verified) :: Wed, 10/10/2007 - 1:51pm

I know you guys are going through a rough spat, but I figure you crazy kids love each other and will figure out a way to make it work. I wish you both luck, Will and Rich.

What I am implying is that the odds of TURNING THE BALL OVER is not significantly higher at the 13 than it is at the 27.

The odds of turning the ball over are significantly worse if you throw than if you run at every yard marker on the field. Therefore, if it is absolutely crucial that you do not turn the ball over, you run the ball.

167
by Will Allen (not verified) :: Wed, 10/10/2007 - 1:51pm

Rich, stop writing this.....

“No Will, I’m not ignoring that. If you want to make that argument, they should have kneeled the ball 3 times on the 27 yard line and kicked.�

...and I'll stop pointing out that there is a difference between kneeling at the 27, and kneeling at the 11.

168
by Kal (not verified) :: Wed, 10/10/2007 - 1:57pm

Rich, another way to put it is this:

In the situation Buffalo faced, the only way that they could realistically lose is if they do not score points and take more time off the clock there. The best way to not score points or take time off the clock there?

Throw.The.Ball.

Whereas if they run at that point, they are far less likely to turn the ball over and more likely to burn time, both of which heavily increase their chances of winning.

I don't even see how this is arguable at this point. Throwing the ball is more likely to turn it over. It's more likely to take 0 time off the clock. Given the stats on the night, it's about as likely to be a successful play. It is in every way a more risky proposition. You're arguing that scoring a TD there is worth it, but the best case scenario is that you force your opponent to score two TDs - which is harder than scoring a TD, a 2pt and an FG, but not by much.

The worst case scenario means that you've kept it a 1-possession game and taken no time off the clock. Which...is exactly what they did.

The smart money says get the points.

169
by Rich Conley (not verified) :: Wed, 10/10/2007 - 1:58pm

"and I’ll stop pointing out that there is a difference between kneeling at the 27, and kneeling at the 11."

Did I ever say there wasnt? NO. I was saying that if its worth risking it there, it may be worth risking it at the 11.

170
by Rich Conley (not verified) :: Wed, 10/10/2007 - 2:01pm

"It is in every way a more risky proposition. "

Kal, I'm not saying its not. I'm just saying the risk may not be significantly higher.

The worst case scenario happened. Not the average scenario. If you want to deal with worst case scenarios, how about a botched handoff run back for a TD? How about a blocked field goal?

We're arguing about this because the ball WAS picked off, not because the chance of it being picked off was very high. Its typical monday-morning-quarterback.

171
by Will Allen (not verified) :: Wed, 10/10/2007 - 2:03pm

Rich, if it is worth risking at the 27, since it is worth risking at the 11, that certainly implies that the situations are very similar at those yard lines, at those times in the game, with that personnel on the field. They aren't.

Now, I really must go.

172
by Al 45 (not verified) :: Wed, 10/10/2007 - 2:07pm

We’re arguing about this because the ball WAS picked off, not because the chance of it being picked off was very high. Its typical monday-morning-quarterback.

Sorry, Rich, that's BS. Yes, we're arguing about it because it did go drastically wrong. If it didn't, it would be a footnote.

However, the fact we're arguing about it because it went wrong does not make it "typical monday morning quarterback". That assertion states it's in hindsight. It's not. I know I sat there, before the play happened, and said they would/should just run it (whether a QB sneak to get the ball to the middle of the field or a RB dive play, didn't matter) and not risk a pass... and, when they did drop back to pass, I was absolutely shocked. I'm quite positive I'm not the only person who felt that way.

It becomes a bigger discussion because of the outcome, but regardless of the outcome, the decision was still wrong.

It's the whole argument that I could attempt to run across the Mass Pike while hordes of cars are plowing down at 80mph. I might even make it across. The outcome doesn't make the decision right. Furthermore, the fact that I made it would lessen the dissent against the decision than if I did not. Not apples to apples, I know, but you get the point.

People aren't arguing the decision because of the outcome. They're arguing the decision because, based upon the available information, it was the wrong decision.

173
by Kal (not verified) :: Wed, 10/10/2007 - 2:17pm

The worst case scenario happened. Not the average scenario. If you want to deal with worst case scenarios, how about a botched handoff run back for a TD? How about a blocked field goal?

I'm talking about reasonable worst-case scenarios. An interception is not the absolute worst thing that can happen, but it's up there - but that's not really what I'm saying.

I'm saying that the normal thing that happens when you throw the ball with a rookie QB on a 3rd and long situation is an incomplete. You've then not moved the ball at all, taken no time off the clock, and made it a harder field goal to kick. That also corresponds to about the worst 'basic' scenario for the Bills at that time.

Another way to look at it is this: taking one minute off the clock is far more valuable at that moment of the game than getting a first down, especially when you consider all the other possibilities that can happen.

An interception wasn't likely to happen. It was more likely than a fumble, but it was not likely. The likely outcome, however, was that they do not succeed at getting the first down, and of those non-successes it is likely that they do not make the clock move. This isn't rocket science; the Bills didn't need to get a first down there, but they did need to keep the clock rolling and allow themselves to score at least a field goal.

Choosing to pass meant the chances of both of those things were drastically diminished.

174
by Wanker79 (not verified) :: Wed, 10/10/2007 - 2:46pm

I guess you can add me to the list of people who thought passing in that situation was idiotic (a.k.a. everyone except Rich).

Rich, you keep harping on the point that the risk between a rookie QB making his 3rd career start attempting to pass with such a short field "may not be significantly higher" than just running into the middle of the line. Can you prove that? Do you have anything to back up that assertion? If you don't it's not that big of a deal since I don't have any hard data to prove otherwise. But if we're all arguing from an intuitive point of view, I find it hard to believe that anyone can look at that situation (regardless of the outcome) and not see how much riskier a pass would seem to be.

175
by DJH (not verified) :: Wed, 10/10/2007 - 3:06pm

Rich, take this from a guy who would rather lurk and read than post…and doesn’t have a dog in this hunt.

You are quickly becoming an incorrigible little prick who is slowly and steadily ruining these comment threads.

Your omnisciencent, “lawyering� semantical bullshit charade has run it’s course, and I would ask you to please stop.

176
by crack (not verified) :: Wed, 10/10/2007 - 5:32pm

What were the Bills thinking? They should never have hired Jauron. They should hire whoever wins the hindsight tontine going on in this thread.

177
by QBeam (not verified) :: Wed, 10/10/2007 - 7:11pm

I share the frustration about how this thread has become dominated by the Monday-morning QBing of the Monday night game (heh), but I'm perplexed at the hostility being expressed towards Rich. For one thing, Will has been his partner in crime all day. Assuming arguing the point ad nausium is the crime, that is. The fact that Rich is catching the flack and Will isn't suggests that it's not.

Regardless, Rich has obviously got the better end of the argument. Even if we buy Will's theory that there is some important difference between the 27 yard line and the 13, there was still more than 7 minutes on the clock when they got there. That's way, way too much time to go into a prevent defense, and so it's way, way too much time to abandon your offense. Furthermore, there is one huge difference between being on the 13 yard line and being on the 27--the chance of scoring a TD goes way, way up. Which means the opportunity cost of abandoning your best offense and planning to just kick a FG goes way up. One more completed checkdown pass and the Bills probably go up by 15. Lights out. To that point in the game, the Bills had converted about 2/3 of the time (12 first down conversions vs. 5 punts/FGs). So just keep playing your offense, and you've got something very close to a 65% chance to score a TD and put the game away.

Against that chance it would have been silly to settle for a FG, just to avoid a 3% chance of an INT. The risk of missing out on a FG just doesn't justify that exchange. There's not that much difference between an 8pt lead and an 11pt lead. As it happens, Dallas missed the 2pt conversion, but even if they hadn't, even that doesn't give Dallas the win, absent the recovered onside kick. So now we're multiplying the 3% risk of an INT times the 3% chance of a recovered OK. So yeah, I think it's pretty fair to say that, if this attack on Jauron isn't hindsight, it's just goofy.

178
by coldbikemessenger (not verified) :: Wed, 10/10/2007 - 7:45pm

re 177
With 7:09 left and a first down the Bills can
1. run three plays milking 2:15 of the clock.
2. kick a fg
3. Presumably get the ball back and run another 2:15 off the clock.

Now Dallas has about 160 seconds(2:40) to score a fg and a td. A few special teams plays would reduce this a bit more.

That seems very very hard.
Not impossible, I am sure it happened sometime.
But you are putting a lot of pressure on Dallas to move fast twice.
Its not that you don't want an interception, you don't want any incomplete passes at all.

179
by coldbikemessenger (not verified) :: Wed, 10/10/2007 - 7:47pm

Also,
I have no real data to back me up but
I have noticed over the years that the first play of a two minute offense can take a lot of time off the clock.
Often it seems like one play, boom, 25 seconds gone.
Has anyone else noticed this?

180
by jimm (not verified) :: Wed, 10/10/2007 - 9:03pm

If Buffalo still had Flutie they could have tried a drop kick.

181
by Frankie (not verified) :: Wed, 10/10/2007 - 9:37pm

Has FO ever considered implementing a maximum posts per topic rule?

I think if everyone were limited to no more than 5 to 7 posts per topic, the level of intelligent, insightful, and meaningful discussion could be improved.

182
by young curmudgeon (not verified) :: Wed, 10/10/2007 - 10:10pm

oh, well, another thread hijacked...

183
by Pat (not verified) :: Thu, 10/11/2007 - 12:44am

Buffalo got the ball back after that, with the score STILL THE SAME, and pretty much lost the game because they ran on 1st and 2nd and it set up a 3rd and long that wasn’t succesful. Was that the right call?

Actually, they pretty much lost the game because they ran on 1st, and 2nd, and not on third. If they had run on third down, even if they hadn't picked it up, they probably would've won the game.

That fact is hindsight, but the idea isn't. Running on first and second down was right. Passing on third was stupid. You're winning, up by 8, the other team can only tie at best, and there's less than 4 minutes left. Bleeding the clock is much, much more valuable than trying to keep the ball.

184
by Saints_just_playin_for_draft_picks (not verified) :: Thu, 10/11/2007 - 12:35pm

I am deeply offended that the New Orleans Saints are ranked only 31 on the charts. I mean, really, this is an insult to the San Francisco 49rs. The 49rs don't actively play the game to loose it, they actually seem somewhat intent on winning occassionally. The Saints, on the other hand, actively seek out and find new ways of looking terrible in every facet of the game.

I only hope that the Saints can indeed go 0-16 and draft at least one good receiver and one good cornerback and can come up with a defensive coordinator that can squeeze blood from a turnip. They could probably use a good offensive line coach too.

185
by Matt (not verified) :: Thu, 10/11/2007 - 2:00pm

177 -

Not that anyone is still reading this, I hope, but some of your statistical claims are utter B.S.

You say: "To that point in the game, the Bills had converted about 2/3 of the time (12 first down conversions vs. 5 punts/FGs)."

The Bills did have 14 first downs by the end of the game. They also finished 3 for 13 on third downs, 1 for 2 on fourth downs, punted five times, and went 1 for 2 on FG attempts. So to say that they'd converted 2/3 of the time is wildly misleading, at best. They converted third downs less than 25% of the time, 4th downs at a 50% clip (thanks, I take it, to the fake punt), and mounted one scoring drive on offense.

You can't use "converted" as a substitute for picking up a first down -- at least not as an argument for what they should have done on 3 and 8.

Here is how their offensive drives ended, excluding McGee's KOR TD, with the number of plays in parentheses:
Punt (8)
Downs (4)
Punt (6)
FG (15)
Missed FG (4)
Punt (7)
Punt (5)
Interception (8)
Punt (3)

You conclude: "So just keep playing your offense, and you’ve got something very close to a 65% chance to score a TD and put the game away."

There's absolutely no support for your 65% figure in your erroneous claim that they'd "converted" 2/3 of the time. Now for all I know a team may have a 65% of scoring a TD once it reaches the 13 yard line, as Buffalo had at that point, but it's not because of the number of punts they had compared to the number of first downs they made during the course of the game.

186
by hooper (not verified) :: Thu, 10/11/2007 - 2:32pm

Well, to change things up a little:

The Broncos have an offensive DVOA of 12, with passing and rushing each at 13. According to their offensive stats, their offensive variation is 29th at 1.3%, which is behind the Colts (32), Vikings (31), and Saints (30). That would mean that the Broncos have been roughly average league-wide, and consistently so. Given the loss of Walker in a couple games, the loss of Smith (admittedly aging anyhow) so far this season, a relatively new QB, the loss of Nalen's bicep for most of the San Diego game, and yet another backfield of runners, I would thinks this speaks well of the Broncos offense. They're not the greatest in the league, but their play is reliable for what it's worth. I would guess this either means they've hit a ceiling, or they're in a position to improve on a fairly regular schedule. Thoughts?

Regarding the Broncos D: #27 with the top variability (27.4%), where only the Giants at 25.0% are even close, explains the bulk of their struggles as of late. The variability seems to be a function largely of the annual overhaul of the D-line, with the Bailey and Bly injuries in the last game not helping any. To have that much variation and be ranked that low means that the lowest moments have been extremely bad (not that we didn't already know that). The best moments don't even need to be all that great, which is completely demoralizing for this Broncos fan. Still, there is obviously a lot of room for improvement, and I would think that a lot of improvement should naturally occur just as a result of the new defensive players and schemes gelling together over time. Pass D at 20, and run D at 30 seems to also indicate that the hemorrhaging is primarily a D-line issue.

Lastly, the ST DVOA is 31 - gaah! - though they are comfortably joined by KC and Oakland at 29 and 30 respectively. ST has been a Denver problem for years, and I don't know why they don't improve that any. Certainly, the SD game wouldn't have been as frightful if the special teams play had merely been average.

Opinions?

187
by hooper (not verified) :: Thu, 10/11/2007 - 3:05pm

Re: 179

There seems to be a timing issue in the heads of a lot of coaches where they don't want too much time on the clock. I believe the logic is close to this:

This drive is our last drive of the game. If we fail, they will kill the remainder of the clock and we lose. If we succeed, we have either a tie or a lead, and we have to kick the ball back to the opposition. If we succeed with 10 seconds left on the clock, it'll be really hard for them to mount a comeback. If there's 35 seconds left on the clock, their job is much easier. Let's try to score AND leave them with no time.

Obviously, letting some time fritter away comes at its own risk, but I think a lot of coaches see that as a better risk than the comeback drive.

188
by QBeam (not verified) :: Thu, 10/11/2007 - 3:15pm

Matt, Judging from the fact that you listed the results of *drives*, rather than the results of *series*, I think you may have overlooked one of my assumptions. Namely, I assumed that the odds of getting a TD from 1st-and-10 on the 13 are basically the same as the chance to get another first down. I think that's a reasonable assumption--even if they did happen to gain 10 yards w/out getting in the end zone, they'd still have to get stuffed at the line three more times in a row (barring penalties).

Which is why your list of how the drives ended is beside the point. Up to that point in the game, (the 1st-and-10 on the 13 w/ 7:09 to go), Buffalo had a 1st-and-10 situation 17 times, and 12 of those times they managed, one way or another, to get at least 10 yards and a new 1st down. Five times they failed, and had to kick. That's 12/17, or just over 70%. I discounted to 65% because there was at least some chance they'd convert and get stuffed three times.

Another way to think of it is that a drive that begins on the 13 yard line is a whole lot more likely to result in a TD than a drive starting on the far 20. Which, of course, was my point in the first place. Once Buffulo got to the 13, it would have been madness for them to give up on trying to get a TD. Once they got that close, it made even MORE sense for them to run their offense and try to go up by 15.

189
by Will Allen (not verified) :: Thu, 10/11/2007 - 4:04pm

Qbeam, to equate a first and ten at midfield with first and ten at the opponent's 13, in terms of the chances of gaining at least 10 more yards, is not an accurate way to look at the game.

190
by stubert (not verified) :: Thu, 10/11/2007 - 4:19pm

When Dallas was backed up at its own 11, the BEST thing that could have happened for them was an interception returned for a TD, or at least a long return. If you’re Buffalo’s coach and you have 95% + chance to run time off the clock and put your team ahead by two possessions, and you opt out of that strategy to do the ONLY thing that could give your opponents their best chance to win, you’ve made the wrong decision. Jauron made the wrong decision. It was the wrong decision before Edwards dropped back to pass, regardless of the outcome of the play. It isn’t hindsight. It’s about giving your team the best chance to win the game and Jauron didn’t.

191
by QBeam (not verified) :: Thu, 10/11/2007 - 5:16pm

Will, that's a valid point. Red zone offense is tougher than the rest of the field--the backfield runs out of places to hide, as a friend of mine puts it. But how much harder to you want argue it is? Did their chance to convert go down from 70% to 50%? 30%? If they convert, its game over--15pt lead, and all that time run off the clock. How low does that chance to put the game away have to be before it's outweighed by the danger of an INT, TD, 2pt conv.?

192
by Will Allen (not verified) :: Thu, 10/11/2007 - 6:00pm

With this particular set of personnel, qbeam? I'd say that that the chances of them gaining another 10 yards is about the same as the chance of throwing an int. We're talking about a guy who has had only a couple of starts, against a very good defense whiose only pronounced weakness, deep pass defense by the safeties and cbs, does not come into play on the 13 yard line. My bet is that in that siutation, Dallas kicks Buffalo's ass more than 90% of the time. The way for the Bills to maximize the chance of victory, as measured by what strategy, played out through a couple of thousand trials, ended up most frequently with the Bills winning, was to take as much time off the clock as possible, and then kick a field goal with a 95%-plus chance of success.

193
by Matt (not verified) :: Thu, 10/11/2007 - 7:08pm

188 -

I understand and accept your explanation to some small degree, but Buffalo had also failed to "convert" in your terms on the sets of downs where they kicked a FG, missed a FG, and turned the ball over on downs. I still don't trust all of your numbers, calculations, or arguments, but even on your terms doesn't that mean that they had "converted" 12 out of 20 times?

Moreover, since we are talking about a 3rd and 8 playcall here, aren't the more relevant numbers the ones that show many times the Bills had converted on third down -- rather than how many 1st downs they'd gotten all over the field during a wide variety of game scenarios? Buffalo was 3/13 on third downs on the night. (For the season they are at 34%.) Edwards had a high completion percentage on the night, at 23 for 31, but his yards per attempt came out to 5.67 for the game.

Taking those stats in conjunction with Will's several points, don't all of these numbers factor more heavily into the decision to throw on 3rd and 8 than the 12 times the Bills "managed, one way or another, to get at least 10 yards and a new 1st down"?

194
by Dont worry Im NOT a Giants fan (not verified) :: Sat, 10/13/2007 - 7:30am

It's been interesting reading this thread that's for sure. Sorry I'm late too it but just in case anyone is still checking it, I'd like to add some points.

First, does Jauron even call the offensive plays? I would guess not, since he has the defensive expert rap, so lets not crucify the guy for the calls OK?

Second, I think part of the Bills strategy was to be fairly aggressive. For example, fake punt in the 1st quarter, there was also some trickery later with Roscoe Parrish on 3rd down that got stuffed from memory too. Given that line of thinking, I have no problem with them looking to exploit Dallas with a pass on 3rd & 8 in a situation where Dallas would be quite wary of a run. That shows balls, trying to put the game away rather than limply admitting you are sitting back and waiting for Dallas to come back at you.

Third, back to the call again. The highest percentage call isn't always the right one. In theory it is of course, but think about this. We know that passing the ball is more effective than running in most situations, but is the Mike Martz offense best in the NFL? No. Coaches can't just robotically call the highest percentage plays in every situation - it is far too predictable. Multiple posters have said that they expected the Bills to run, wouldn't Dallas be expecting that too?

I can't believe the flack over this call. The really tragic call was the 2nd last play where they gifted the Cowboys 7 yards on a quick out - that was the inexcusable that we should be discussing because that truly cost them the game.

And for what it's worth - Rich usually I don't care much for your arguments, but I think you said some surprisingly good stuff in patches on this thread - don't agree with all of it of course, but you didn't deserve the crap most people gave you.

195
by Alex (not verified) :: Sat, 10/13/2007 - 4:43pm

The highest percentage call isn’t always the right one. In theory it is of course, but think about this. We know that passing the ball is more effective than running in most situations, but is the Mike Martz offense best in the NFL? No.

Key word, most. It's not the highest percentage call in all situations, so you shouldn't always call a pass. The Mike Martz offense isn't the best in the NFL for several reasons, mostly talent-related, but also because he calls passes in situations where passing is not the highest percentage call. If he ran when running was the highest percentage call, then his offense would be better.

Coaches can’t just robotically call the highest percentage plays in every situation - it is far too predictable. Multiple posters have said that they expected the Bills to run, wouldn’t Dallas be expecting that too?

So? Even if Dallas knew exactly what was coming, they wouldn't have been able to do anything to stop the Bills from running time off the clock and then kicking a field goal. Even if Dallas stuffed the RB for a 3 yard loss, the Bills would go from kicking a 20 yard field goal to kicking a 23 yard field goal. Big deal, they'd still make it 95% of the time at least.

You're assuming that Buffalo needed to conceal their intentions in order to have a successful play. While that is the case in many situations, there are plenty of situations where it doesn't matter at all.

For instance, at the end of a game, when up by 1 point, on first down, with one minute on the clock, with the other team out of timeouts, the percentage play is a QB kneel. The other team will be expecting it, but that's ok. It won't matter, because there's basically nothing they can do to stop the play from succeeding.