Writers of Pro Football Prospectus 2008

04 Oct 2005

Any Given Sunday: Chargers over Patriots

by Ned Macey

Every week sees surprising results in the NFL, and often they are more surprsing than Chargers over Patriots. What makes this week's upset so specatcular was not only that it took place at New England, but the way the Chargers thoroughly dominated the second half. Patriots fans could shrug off the first loss of the season at Carolina due to extenuating circumstances. What happened Sunday, however, was much more alarming. For the first time since 2002, the Patriots lost at home, and in doing so, they were thoroughly outclassed. As has been well reported, the Patriots were short-handed with the loss of Rodney Harrison, Matt Light, and several cornerbacks. The loss of Harrison in particular left the Patriots particularly vulnerable to the Chargers, whose entire offense is based around a dynamic running back and the game's best tight end. While the Patriots have to worry about giving up so many points, they may simply have played the best offense in football thanks to a healthy LaDainian Tomlinson.

Before we look at how San Diego exploited the Patriots' weaknesses, let us take a second to consider the Patriots team as a whole to identify those weaknesses. Last season, the Patriots faced some serious injuries leading famously to wide receiver Troy Brown's becoming their nickel back. The truth, however, is that by the end of the season, the only major source of injury was in the secondary. While the loss of Ty Law and Tyrone Poole to injury was nothing to laugh at, the Patriots played the Super Bowl with only one other projected starter out with injury, left tackle Tom Ashworth.

From the starting line-up in the Super Bowl, the Patriots have turned over seven new starters, and only one can be considered a clear upgrade.

Super Bowl Last Week
Matt Light Nick Kaczur
Joe Andruzzi Logan Mankins
Keith Traylor Vince Wilfork
Tedy Bruschi Chad Brown
Ted Johnson Monty Beisel
Randall Gay Duane Starks
Rodney Harrison Gus Scott

Looking at these losses, the Patriots have subbed out three Pro Bowl quality players (Light, Bruschi, Harrison) and added only one player with serious upside in Wilfork, and he actually played more snaps than Traylor a year ago. In the Patriots' defense, they planned well in the off-season for their known changes. They let an average player in Andruzzi go and replaced him with first round pick Mankins, a potential short-term decline that will be a long-term improvement. Given their lack of depth in the secondary, they added Starks and Chad Scott. When Tedy Bruschi had his stroke, they signed Brown and Beisel to add flexibility. Still, this was a team that entered the season with less talent than it had a year ago. Injuries to key players this year have only exacerbated the situation.

Even with the departure of Ty Law, the secondary was supposed to be a major strength thanks to emerging players Asante Samuel and Eugene Wilson. Injuries are again crippling the depth of the unit. A year ago, the Patriots were excellent against # 1 receivers, but the lack of depth left them vulnerable to # 2 receivers. The injuries made that true again on Sunday where Keenan McCardell was held to one catch (albeit a touchdown), while Eric Parker and Reche Caldwell combined for six catches, 87 yards, and one TD. Those numbers are not outstanding, but they were compiled in only seven attempts, and all but one (a 19-yard reception on first-and-20) went for a first down or touchdown.

While the Patriots struggled against secondary receivers last season, they were dominant against tight ends. Using DVOA, our statistic that breaks down every play to see how much success a team has compared to league average, the Patriots were the fifth best team in football defending the tight end, with a DVOA of -31%.. The person most responsible for this was Rodney Harrison. Nicely illustrating this is that the Patriots have ranked fifth and second against tight ends since Harrison signed in 2003. The year before Harrison joined the Patriots, they ranked a mediocre 16th. Going into a match-up with football's top tight end, the Patriots had no answer. Antonio Gates was Drew Brees' favorite target, and he responded with 6 catches for 108 yards. Three of those catches were converted first downs when the Chargers were facing third down and five yards or more.

The biggest catch of the day for Gates was emblematic of many of the Patriots' problems. With the score tied at 17 early in the third quarter, the Chargers faced third-and-5 on the New England 39-yard line. On the play, Brees rolled right, had plenty of time, and then threw back to his left down the field. The ball was your typical "jump ball," but Gates easily outmaneuvered Harrison's replacement Gus Scott and came down with the ball on the one yard line. On the next play, Tomlinson ran over Beisel and the Chargers never looked back.

Of course, the Patriots' defensive struggles may prove to be nothing more than a fleeting mirage caused by their facing the best offense in the NFL, the San Diego Chargers. The Chargers have the early lead in offensive DVOA with a 46.7% DVOA. Despite no major off-season additions, the Chargers entered this season with a major weapon they did not have a year ago: a healthy LaDainian Tomlinson. A season ago, the Chargers ranked sixth in DVOA offense, mostly attributable to a passing game that ranked fifth in the league. Their rushing offense was a mediocre 14th. Tomlinson, widely considered the best back in football, ranked 34th in DVOA just ahead of Travis Minor.

What kind of statistics could possibly consider Tomlinson such a substandard back? The kind of statistics that actually look at what happened on the field. Thanks to conventional wisdom that arbitrarily declares 1000 yard rushers are valuable (an average of 62.5 yards per game) and fantasy football that smiled on Tomlinson's 17 touchdowns, many people missed how mediocre his season was. Even for DVOA non-believers, the numbers tell the story: yards per carry dropped from 5.3 in 2003 to 3.9 a season ago. None of this means the real Tomlinson is not an exceptional back, but simply that even the amazing Tomlinson is only middling when hampered by injury.

This year, the Real Tomlinson is being added to a potent passing attack to create a dynamic offense. Tomlinson through four weeks is first in all of our measurements: DVOA which measures on a per play basis, DPAR which measures total value, and success rate which measures the percentage of plays that are "successful." (Full definitions of all of our statistics are available here.) Tomlinson ranked sixth and second in DPAR in 2002 and 2003, so the mainstream acceptance of him as the league's best back was warranted. In the emergence of the Chargers as an excellent offense, nobody seemed to notice that Tomlinson was having a down year. A healthy Tomlinson, a developing Gates, and full seasons from McCardell and Caldwell may keep the Chargers as the best offense in football all season.

The Patriots' brutal early schedule continues with consecutive road trips to 3-1 Atlanta and 3-1 Denver. The Chargers do not exactly have an easy time of it, as they host Pittsburgh this week before heading to Oakland and Philadelphia in Weeks 6 and 7. Both teams sit at 2-2 and in the competitive AFC cannot afford to lose too many more games. Ironically given the results of this game, the Patriots may be in a better position than the Chargers due to the surprising weakness of the AFC East. After four weeks, the AFC East has only upstart Miami in the top half of our DVOA rankings, while the lowest ranked AFC West team is Oakland at 15th.

The Chargers may really cost themselves a division championship thanks to their hardball negotiating tactics with Gates that led to his missing their Week 1 game against Dallas. The offense is something to behold, but their defense has been poor, with a DVOA of 9.0%, good for 25th in the league. Last year's Chargers' offense was helped by a defense that ranked 11th in DVOA. The defense should improve as the year goes on, but in the all offense all the time environment of the AFC West, the likely division champion is going to be the team that learns how to at least occasionally stop the opposition.

The Patriots still have the horses to make the playoffs given the state of their division and the very good talent remaining, but this team is just not as good as the Super Bowl champions of a year ago. Bill Belichick remains the best coach in football, but even a great coach can only do so much with the talent he is given. This article focused on the deficiencies of the defense, but without Light, Brady looked rushed all game, and the running game was struggling even before Light went down. Before the playoffs come, the Patriots should get back Light, Gay, Poole and Kevin Faulk. If they are lucky, only Harrison will be gone for the season. Still, all three Patriots Super Bowl runs have started with a first round bye in the playoffs, and with two losses already in the books and a tough upcoming schedule, it looks like the Patriots will be playing on Wild Card weekend this year. 10-6 or 11-5 and a division championship is good enough for most teams, but after going 34-4 over the past two seasons, it is not good enough for the Patriots.

Posted by: Ned Macey on 04 Oct 2005

61 comments, Last at 07 Oct 2005, 12:40pm by Sue Ellen

Comments

1
by B (not verified) :: Tue, 10/04/2005 - 3:26pm

Nice article, but it misses the biggest problem the Patriots defense had last week. They just weren't getting anywhere near Brees, and this is the same d-line that dominated the Steelers o-line last week. Last year the Patriots had been able to cover up for defincies in ther covereage by hurrying and harrying the QB. I'm afraid this might be because, with a 4-3 front and less players to rotate into thier fronts, they're just tiring out. Of course, since the Patriots offense couldn't stay on the field in the second half, they gave the defense no time to rest, which compounds the problem.

2
by PatsFan (not verified) :: Tue, 10/04/2005 - 3:35pm

with two losses already in the books and a tough upcoming schedule, it looks like the Patriots will be playing on Wild Card weekend this year.

Ok, so they'll need to call this year's DVD Four Games to Glory instead of Three Games to Glory IV :-)

But seriously, good article. And your insights about TE coverage don't make me feel good about this week, when the Pats have to go up against Crumpler.

BTW, the Boston papers have been reporting that Poole and Gay have been seen around the locker without crutches or limping for a couple of weeks now, so hopefully their return is sooner rather than later.

3
by Ned (not verified) :: Tue, 10/04/2005 - 3:42pm

B-
You're exactly right about the pass rush, and it was in the original draft of the article, but the article was just going on forever. Plus, I think the pass rush will be fine, making this week a bit of an aberration. I do not, however, think Gus Scott with all the coaching in the world will be Rodney Harrison.

4
by jim's apple pie (not verified) :: Tue, 10/04/2005 - 3:48pm

Excellent article. One of the things I was wondering about while watching on Sunday was why the Patriots never blitzed. They were getting zero penetration, Seymour was a no-show, and the Cowboys and Broncos both showed that pressure is the only way to stop the Chargers.

One aspect that might have gone unnoticed outside of San Diego was the upsurge in offense the last two games coinciding with the replacement of OG Fonoti with Dielman. Fonoti is a monster in the run game, his issues have always been in pass protection. He is huge (380 pounds or so) and moves well, but not nearly as well as Dielman. Fonoti was largely ineffective in the Denver game, where he was playing with a broken hand. After that game, he had surgery and it looks like he is going to have another, so I don't expect him back anytime soon. With this being the last year in his contract, I expect the Chargers to part ways if Dielman shows that he can handle being the starter.

Dielman has looked great, and seems to be positively nasty. There was a play in the third quarter in the NY/SD game where Pierce jumped on Dielman's back while he was on the ground and took a swing at his back. There was also a little scuffle in the NE game that Dielman seemed to be involved in. My take is, if the opposing DL and LB's don't like the guy, he is fine by me.

(As long as he isn't chop blocking or anything else Denver-like.)

5
by B (not verified) :: Tue, 10/04/2005 - 4:00pm

I think it was Jarvis Green (best known as the guy who threatened to beat up anybody who said he couldn't replace Seymour in last years playoffs) he was fighting with.

6
by Kibbles (not verified) :: Tue, 10/04/2005 - 4:04pm

I think all of this "best offense in football" talk is still a trifle bit early. To date, they've faced two bad defenses, one hamstrung defense, and one good defense. And that good defense held Tomlinson under 3 yards per carry, and held the Chargers to 41 yards after halftime.

I understand that everyone has bad games, but I don't think "the best offense in football" is allowed to get shut down that badly and still be considered "the best offense in football". I actually don't even think they're the best offense in their division. That distinction belongs to Kansas City, at least when they're healthy, which they clearly haven't been to date.

Personally, I give the "best offense in football" moniker to the Eagles for as long as McNabb, Westbrook, and Owens are playing.

7
by B (not verified) :: Tue, 10/04/2005 - 4:14pm

Maybe it should be "the best offense at or around sea level"

8
by SLB1 (not verified) :: Tue, 10/04/2005 - 4:36pm

The Chargers have a good to great offense, but this article is correct is pointing out that the defense is poor. I watched the SD vs. Denver game, and the Chargers have a DB whose last name is "Florence" who outright sucked in that game. I just chalked it up as a bad game, but he was just as bad in the NE game. In the AFC West, that guy is going to get eaten alive.

9
by SLB1 (not verified) :: Tue, 10/04/2005 - 4:37pm

The Chargers have a good to great offense, but this article is correct is pointing out that the defense is poor. I watched the SD vs. Denver game, and the Chargers have a DB whose last name is "Florence" who outright sucked in that game. I just chalked it up as a bad game, but he was just as bad in the NE game. In the AFC West, that guy is going to get eaten alive.

10
by SLB1 (not verified) :: Tue, 10/04/2005 - 4:37pm

The Chargers have a good to great offense, but this article is correct is pointing out that the defense is poor. I watched the SD vs. Denver game, and the Chargers have a DB whose last name is "Florence" who outright sucked in that game. I just chalked it up as a bad game, but he was just as bad in the NE game. In the AFC West, that guy is going to get eaten alive.

11
by SLB1 (not verified) :: Tue, 10/04/2005 - 4:39pm

Sorry for the repeat post. I think I set a record for putting it in three times. I had a computer lockup and apologize to all.

12
by princeton73 (not verified) :: Tue, 10/04/2005 - 4:42pm

it was worth reading 3 times; Florence actually sucks that bad

I think that injuries to starters are not additive--they're exponential

2 starters out is worse than twice as bad as one starter out, etc

13
by elhondo (not verified) :: Tue, 10/04/2005 - 4:51pm

Re: #6

you do realize that when they say "best offense" around here, that they mean "according to DVOA", right? It's not that they're just throwing out a subjective opinion to offend you.

14
by elhondo (not verified) :: Tue, 10/04/2005 - 4:51pm

Re: #6

you do realize that when they say "best offense" around here, that they mean "according to DVOA", right? It's not that they're just throwing out a subjective opinion to offend you.

15
by Fnor (not verified) :: Tue, 10/04/2005 - 5:14pm

re: #14

I think his counter-point is more that DVOA isn't very D-ish this early in the season.

16
by Pedro (not verified) :: Tue, 10/04/2005 - 5:21pm

I know it sounds cliche, but I haven't heard any pundit mention the loss of Charlie Weis as OC affecting the performance of the Pats. Looking at the job he's done in turning the Irish around this season, I find it hard to believe that losing such a valuable mind on the sidelines would make a negligible impact on NE.

17
by Jerry F. (not verified) :: Tue, 10/04/2005 - 5:53pm

I don't follow college football at all, so maybe this is a dumb question, but is it premature to say Weis has turned the Irish around? It seems like people were pretty excited about the job Willingham did early on as well.

18
by Kibbles (not verified) :: Tue, 10/04/2005 - 5:54pm

Re #12: I think that injuries to starters are not additive–they’re exponential

2 starters out is worse than twice as bad as one starter out, etc

I don't think it's necessarily the number of starters, or the quality of the starters, that determines the impact. I think that there are "good" injuries and there are "bad" injuries. "Good" injuries obviously aren't good, but they're injuries that happen to a position of depth. For instance, in 2003, Denver suffered a massive rash of injuries at the LB position. They lost their #2, #3, and #4 LBs for the season. Luckily, however, LB was the deepest position on the entire defense- those were "good" injuries, because there were backups capable of stepping in and performing at a high enough level to not be a liability. They lost 3 LBs without missing a beat, but the second they lost a single CB, Kelly Herndon, their defense fell apart. Carson Palmer might be the best offensive player on the Bengals (for my money, it's Chad Johnson), but if he goes down to injury, that has to fall in the "good" injury category, because he has a backup in John Kitna who has proven himself capable of performing in that offense at a high level. Likewise, when Frerotte was in Minny, losing Culpepper was a "good" injury, and as long as Volek is in town, McNair is a "good" injury. I think a team can sustain any number of "good" injuries without losing too much (again, refer back to 2003, when Denver lost 3 LBs without skipping a beat).

Now, on the other side of the coin, you have the "bad" injuries- the injuries to players with little to no quality depth behind them. And for all the discussion about "The Patriot Way", where they shun superstars and quietly compile great depth, no team is going to have quality depth at every position. To date, New England has been lucky with injuries. Yes, I understand that they suffered a huge rash of injuries, but all of those were "good" injuries. They never lost Tom Brady, they never lost Rodney Harrison, they never lost Tedy Bruschi. They never lost anyone they couldn't afford to lose.

Now, however, the Patriots are suffering a rash of "bad" injuries (or retirements), and people are wondering at their inability to just plug someone else in.

To use another Broncos example, their secondary is the perfect illustration. Two seasons ago, Kelly Herndon broke his hand at the end of the season, and the entire secondary fell apart. That single injury to the secondary was a "bad" injury, and Manning "exposed" them in the playoffs. Then, last season, they add more depth. Lenny Walls goes out with injury, but Denver's still okay. Then Willie Middlebrooks went out, too, and suddenly Denver was "exposed" again. Now, fast forward another year. Last week, Denver lost both of its starting CBs to injury, but its two primary backups came in and performed at a high level, and Denver shut down Jacksonville's passing attack.

Re #13: I understand they're going by DVOA. I was simply pointing out that the defense adjustments are still very weak this early in the season. The only quality defense SD has played shut them down cold.

19
by elhondo (not verified) :: Tue, 10/04/2005 - 5:54pm

#14

Agreed, I was picking a nit. But my point was that you can't say SD doesn't deserve the title because of the Denver game, any more than you can say that Philly deserves the title because the walloped SF.

I wanted to make the point that the article, at least, wasn't being subjective, and when disagreeing it's generally better to provide some reasoning behind your comments. To make the point that both Philly and KC deserve higher rankings when even their NON-ADJUSTED numbers are lower, kind of stretches the argument thin.

Don't make me post this twice.

20
by Drew (not verified) :: Tue, 10/04/2005 - 6:00pm

Re: #16

I thought the same thing about Weis, and Crennel too for that matter, even before the season started. Good coordinators can be worth their weight in gold. Insert Weis/Crennel fat-joke here.

21
by Pat (not verified) :: Tue, 10/04/2005 - 6:19pm

“Good� injuries obviously aren’t good, but they’re injuries that happen to a position of depth.

Even moreso, though, you have to consider what Carl's been posting regarding the "importance" of certain positions. In other words, you have to consider the "depth" of the waiver wire at certain positions, and at what speed the performance falls off, and how much that position contributes to the team as a whole.

QB, for instance, is a tremendously important position, and the number of NFL-quality quarterbacks out there is very small. Probably less than 100. Now rate all 100 QBs in terms of performance, and if you would say that the top QB is 100, and the bottom QB is 0, then in a perfect NFL, the dropoff from your first level QB -> second QB is ~ 30-50%. Dropoff from first -> third QB is 75-100%, etc.

But take RB, for instance, of which there are an absolute ton of NFL-quality running backs out there. Say there are 10 times more out there (1000), and teams still carry 3, and suddenly the dropoff from first RB->second RB isn't so large (3-5%), first RB->third RB still isn't large (7-10%), and even first RB replaced with a waiver wire RB isn't that large (10%).

Note that essentially what I'm saying is that not only may individual teams have certain positions where they can manage fine with injuries, but there are certain positions which from year to year the team is able to replace, depending on the depth of the talent pool available.

Simple example this year is that the Jets might not have been so strapped at QB if Chicago, Kansas City, Arizona, and Detroit hadn't suffered a slew of QB injuries/replacements. As it is, they, uh, don't exactly have too many choices available to pick from.

22
by RCH (not verified) :: Tue, 10/04/2005 - 6:25pm

The funny thing about Weiss is that I don't remember anyone calling him a genius when he was the OC for the Jets (under Little Bill). He was a great fit for this team and worked well w/Brady but the way that the running game is not producing I don't think that there any any magic plays that would turn things around.

23
by princeton73 (not verified) :: Tue, 10/04/2005 - 7:36pm

The funny thing about Weiss is that I don’t remember anyone calling him a genius when he was the OC for the Jets

and sure as hell no one called Crennell a genius when he was defcon for the Browns in 2000

(quite the contrary)

24
by Oswlek (not verified) :: Tue, 10/04/2005 - 7:37pm

Re: Weis

I do think that his loss is felt, but it isn't as big as people make it out to be. Let's review

Week 1 - Put up 30 on Raiders
Week 2 - Just a terrible performance. Certainly acceptable to question the coaching
Week 3 - Scored 23 on a good Streeler team, and it could have been much more had they not turned it over in the red zone twice.
Week 4 - The second half was questionable, but they put up 17 in the first half and missed a FG. Not only that, but they were on the 6 yard line with first and goal prior to the last FG.

Where the offensive coaching comes into play is with the O-line. They just look totally out of synch. Due to that, the Pats are reluctant to run the ball. Plus, the longer developing pass routes are not being called as much. Therefore, the play calling is becoming predictable. Much like 2002.

I think once the O-line settles in, nobody will be talking about Weis.

Crennell, surprisingly is the bigger loss. I didn't think he would be, but with all the injuries a coach of his calliber would be very helpful about now.

25
by Oswlek (not verified) :: Tue, 10/04/2005 - 7:38pm

Re: Weis

I do think that his loss is felt, but it isn't as big as people make it out to be. Let's review

Week 1 - Put up 30 on Raiders
Week 2 - Just a terrible performance. Certainly acceptable to question the coaching
Week 3 - Scored 23 on a good Streeler team, and it could have been much more had they not turned it over in the red zone twice.
Week 4 - The second half was questionable, but they put up 17 in the first half and missed a FG. Not only that, but they were on the 6 yard line with first and goal prior to the last FG.

Where the offensive coaching comes into play is with the O-line. They just look totally out of synch. Due to that, the Pats are reluctant to run the ball. Plus, the longer developing pass routes are not being called as much. Therefore, the play calling is becoming predictable. Much like 2002.

I think once the O-line settles in, nobody will be talking about Weis.

Crennell, surprisingly is the bigger loss. I didn't think he would be, but with all the injuries a coach of his calliber would be very helpful about now.

26
by Pedro (not verified) :: Tue, 10/04/2005 - 7:42pm

re #18 and #21: Isn't this exactly what we're compensating for with PAR and DPAR? Obviously, having a good backup benefits your team if your starter goes down. That's why we judge our backups by the same value.

Look at the DPAR values for elite QBs vs. elite RBs. The difference in points earned by a good quarterback vs. HIS replacement is HUGE when compared to a good RB vs. his replacement. Check it...really.

However, I'd never wish an injury on anyone.

27
by Pedro (not verified) :: Tue, 10/04/2005 - 7:42pm

re #18 and #21: Isn't this exactly what we're compensating for with PAR and DPAR? Obviously, having a good backup benefits your team if your starter goes down. That's why we judge our backups by the same value.

Look at the DPAR values for elite QBs vs. elite RBs. The difference in points earned by a good quarterback vs. HIS replacement is HUGE when compared to a good RB vs. his replacement. Check it...really.

However, I'd never wish an injury on anyone.

28
by Carl (not verified) :: Tue, 10/04/2005 - 7:46pm

"and this is the same d-line that dominated the Steelers o-line last week."

Actually, the dominated part of the Steelers' o-line was the right side, a preseason point of concern for many of us who must observe the Stillers.

I have a feeling Indianapolis will discover the right, too, thanks to the Pats' Schlieffen Plan of a game at Heinz.

29
by Scott C. (not verified) :: Tue, 10/04/2005 - 7:56pm

About the strength of schedule. I'd say that the Chargers' schedule is significantly more difficult than any in the league. If you only look at the teams played and the home/road, NE looks like a tough schedule too, but the AFC east just got a lot weaker and the AFC west plays the NFC East.
But besides that there is one thing that makes the schedule of the Chargers rediculous:

FOUR games against opponents coming off a bye. Lets see:
@ New Egland last week then:
Pittsburgh (after their bye)
@ Raiders (short rest, afer their bye, away)
@ Eagles (after their bye)
KC
@ Jets (after their bye)

Rest of the schedule? bye, then Bills, @Redskins, Oakland, Miami, @Indy, @KC, Denver

There are three teams on the bottom half of DVOA (current) on the whole schedule. Redskins, Jets, Patriots. All away, only the Jets look truly weak right now.

Hardest schedule in the league by FAR if you factor in home/road, playing teams after their bye, and road games on short rest. Oakland after their bye on short rest on the road? What?!! then follow with Philly on the road after their bye?

On average every team should get one other team after their bye, the Chargers getting 4, and one on the road on short rest in a division matchup is surely a first in all of NFL scheduling history.

30
by Carl (not verified) :: Tue, 10/04/2005 - 8:01pm

"I don’t think it’s necessarily the number of starters, or the quality of the starters, that determines the impact."

Regression has shown us that the injuries to watch are the ones to starters in key positions. There is no statistical correlation between the NUMBER of injuries and losses. In fact, the Colts and Rams typically record the most injuries nearly every year, and yet have a higher number of wins than most teams, usually.

The most important starting positions to protect, all other things being equal: QB, WR, OL, DL, LB in that order.

There really is not much of a change in a team's fortunes when they lose starting DBs, TEs, Ks, Ps or RBs. In fact, teams tend to do slightly better when their RB goes out.

This analysis possibly might change, of course, with teams that have schemes that rely heavily on a particular player (such as, perhaps, New England's Safety business or the Ravens' rusher). But more often than not, certain positions are pretty fungible (RBs, TEs, Ks, etc.), and because of the high injury rates teams tend to stockpile these guys pretty cheaply.

The Pats somewhat confound this because they are very good at finding replacement value throughout the organization (heck, the QB was a third stringer who became a HOFer overnight).

See also, Steelers.

I would not be surprised if the Pats' O-linemen on the bench don't turn out to be slightly better (and cheaper) than the guys in the trenches now, making a somewhat uncomfortable Visa commercial reunion (cut!) at next year's camp.

The injury rate in the NFL for QBs and OLs (high priced talent) is relatively low, of course, but it's fairly high for WRs and DLs, which is why, I think, some fans are surprised to see their team fade at the end with injuries to their wideouts or defensive line.

They probably didn't realize they were such important components to a winning team.

Because of the importance of line play on both sides of the ball, this is one of the reasons my preseason projections didn't wash out well for Green Bay or Minnesota.

But these projections also helped me predict the rise of the Bengals.

Regression + Markov = I'm beating Vegas right now.

31
by Vern (not verified) :: Tue, 10/04/2005 - 8:05pm

I think 11-5 may well be good enough for a bye this year. The AFC schedule rotated this year so that all the best teams are playing each other - Cincy, Pittsburgh, SD, Colts, Denver, KC, Jax, Ravens, etc.

Yeah, the Colts probably have a lock on the #1 seed, but after that? SD has two losses already, Pittsburgh has 1, and each have yet to play the Colts, Cincy etc.

It's not a talent drop, just that they can't ALL win if they're playing each other.

32
by Led (not verified) :: Tue, 10/04/2005 - 8:16pm

"The funny thing about Weiss is that I don’t remember anyone calling him a genius when he was the OC for the Jets (under Little Bill)."

The man made Ray Lucas look like an NFL QB in 1999. With the combo of Rick Mirer and Ray Lucas taking most of the snaps, the team was 12th in offensive DVOA that year (Chrebet was hurt for 4 games, too). And the offense was pretty darn good in 1998 (see link). I remember thinking Weiss was a very good coach at the time, comparable to Belichick as a strategist.

The thing is, although Belichick had established his reputation back in the days with the Giants, you didn't hear much about his coaching during the 3 years in NY either. Tuna was a rock star with the NY media and dominated the coverage.

33
by TomC (not verified) :: Tue, 10/04/2005 - 10:48pm

Thanks to Carl, I will henceforth refer to the left side of the Steelers line collectively as "Maginot".

34
by thad (not verified) :: Tue, 10/04/2005 - 11:29pm

re kibbles
the eagles?
are you serious?
the eagles have a horrible run pass ratio. the eagles may have a very good regular season offense but it is very difficult to win in the playoffs without
A a higher than average number of running attempts
b a higher than average number of rushing first downs.
Incompletions are brutal. I am not an expert in the fine tuning of DVOA but i read the palmer book. Teams are punished for a high number of incompletions. You can beat a lot of teams during the regular season that way. Much harder in January

35
by Josh, FL (not verified) :: Tue, 10/04/2005 - 11:56pm

Re: Weis

One thing I've noticed is that run plays haven't been called that much this year. One of the things about Weis was that he always stuck with the run (most of the time, at least with Dillon) even when it wasn't working, and Belichick hasn't been as patient with it this year. Before, even if the run wasn't working, opponents still had to respect it cause Weis stuck with it. But it seems like since they aren't sticking with it this year, the defense can just pin their ears back and relentlessly rush and blitz Brady. If you noticed, they were fairly successful running against Pittsburgh, and the line was able to better protect Brady and they won.

Is it possible that perhaps Bill Belichick is just a mere mortal like the rest of us? LOL

36
by Josh, FL (not verified) :: Wed, 10/05/2005 - 12:00am

Re: Weis

One thing I've noticed is that run plays haven't been called that much this year. One of the things about Weis was that he always stuck with the run (most of the time, at least with Dillon) even when it wasn't working, and Belichick hasn't been as patient with it this year. Before, even if the run wasn't working, opponents still had to respect it cause Weis stuck with it. But it seems like since they aren't sticking with it this year, the defense can just pin their ears back and relentlessly rush and blitz Brady. If you noticed, they were fairly successful running against Pittsburgh, and the line was able to better protect Brady and they won.

Is it possible that perhaps Bill Belichick is just a mere mortal like the rest of us? LOL

37
by julian (not verified) :: Wed, 10/05/2005 - 12:31am

I think its far too early for Pats fans to worry about securing just a wild card berth.

With the Bungals resurging, Pitt is no longer guaranteed the division title, and the AFC West could still be a 3-way race up until the final game of the season.

Only Indy, playing in the weakest AFC Division, seems a sure lock to secure a bye in the 1st round.

38
by Sean (not verified) :: Wed, 10/05/2005 - 2:14am

Actually, Parcells took away Weiss' playcalling duties early in '99 and assumed them himself, so it was Parcells calling the plays during the Ray Lucas run. That's not to say that Weiss was twiddling his thumbs the entire season, but he wasn't calling the plays.

39
by KillerB (not verified) :: Wed, 10/05/2005 - 2:50am

I've got to give props for any reference to the Pottsville Maroons.

Just think, if this team still existed not only might they be playing in a stadium called Yuengling Field, but more importantly, the Washington Redskins wouldn't exist!

40
by Bruce Dickinson (not verified) :: Wed, 10/05/2005 - 2:53am

hey thad,
the last time i checked, the eagles have gone to 4 straight NFC championships and only lost the superbowl by a field goal. please do not tell me that this constitutes as 'losing in the playoffs'.

suppose the Pats took TO before 2004 and Philly took Corey Dillon (regardless of TO being TO). does that mean the eagles would have done more 'winning in the playoffs'.

it might. but this team will probably go to the superbowl this year (barring injuries), and could probably beat the colts without an ideal pass/run ratio. their record since 2003 is 33-8 (3-2 in playoffs; 2 garbage week end of season loses excluded). they've pretty much been running the same pass oriented offense the entire time.

41
by Doug (not verified) :: Wed, 10/05/2005 - 9:29am

This win wasn't really surprising to me -- looking at the teams' respective historical trends, San Diego is nearing their peak while the Patriots just passed theirs. What you end up with is a team fighting to stop the bleeding going up against a team that feels unbeatable.

Taking a look at some of the other teams' trends, we can expect Indianapolis to peak alongside San Diego while Pittsburgh begins a slow descent after peaking last year.

And speaking about Philadelphia in comment 40 -- no Super Bowl rings anytime soon...the NFC teams won't regain any sort of dominance until they've been beat up by the AFC for a couple more years.

42
by dryheat (not verified) :: Wed, 10/05/2005 - 9:40am

#39 re: Maroons

You're going to have to spell that one out for me, because in my version of history, the team in Washington was formed as the Boston Red Skins.

43
by Pat (not verified) :: Wed, 10/05/2005 - 12:28pm

And speaking about Philadelphia in comment 40 – no Super Bowl rings anytime soon…the NFC teams won’t regain any sort of dominance until they’ve been beat up by the AFC for a couple more years.

This is silly. Last year should've showed that just because the rest of one conference is weak doesn't mean the top team is weak.

Heck, Philly's 2-0 against the AFC right now, and 1-1 against the NFC. :)

44
by Doug (not verified) :: Wed, 10/05/2005 - 12:36pm

re#43 -- how does last year show anything like that? Philly, the strongest team in the weaker conference, still couldn't beat the AFC's best.

45
by Daniel Warehall (not verified) :: Wed, 10/05/2005 - 1:13pm

#44,

That's the point. Philly lost because NE was slightly better than them.

Whether the AFC is much better than the NFC had nothing to do with whether Philly was going to win the Super Bowl. (Unless, NE was stealing the players from the teams they beat...)

46
by Doug (not verified) :: Wed, 10/05/2005 - 1:35pm

on the contrary, it has everything to do with who was going to win the Super Bowl. I would be glad to provide the evidence if you so desire.

47
by B (not verified) :: Wed, 10/05/2005 - 1:38pm

Since we're writing off New England, let's see how the other top AFC teams stack up to Philly.
Indy: Philly's defense should be able to harry and confuse Peyton, and Indy's defense doesn't stack up well against Philly's offense.
San Deigo: Martyball, nuff said.
Pittsburgh: Philly struggles against power-rushing teams, so this will be a good matchup.
Denver: See above.
Cincy: Come on, they're still the bengals.

48
by james (not verified) :: Wed, 10/05/2005 - 3:19pm

Regarding Philly,

What league you play in only comes into account when you consider the amount of tough games played.

Every team with a -10 or lower DVOA can be expected to turn even the best offense into an average offense thus allowing no more than 20 points. (This of course disregards the effect of turnovers and special teams.)

Therefore to me Philly with their very above average offense a very above average defense would have won the superbowl 4-6 times if that game were played out 10 times. The same would result if they played Pitt

The colts would have had no chance.

Philly was the third best team in the leauge last year at worst.

What has that got to do with what league they were in.

49
by Fnor (not verified) :: Wed, 10/05/2005 - 4:29pm

I couldn't disagree more with the "peaking" argument. If you read this past summer's series on caponomics, you find that sort of behavior among "boom and bust" teams. The idea is that they amass so much dead weight and cap burden that they have to purge themselves and rebuild. Not all teams work that way. Did Denver peak? Perhaps, but they were consistently good with Elway and then consistently not-quite-so-good without. What about Phillidelphia? Did they peak last year? Or are they peaking this year? And how does that explain the 2001 Patriots?

Just saying "this team peaked" and assuming that all teams rise and fall on a simple quadratic curve where they have one single "best" year is an ill-founded and simplistic analysis.

50
by Doug (not verified) :: Wed, 10/05/2005 - 5:17pm

Well, all I have to say is that 29 of 32 Super Bowl champs since 1972 have come from the dominant conference..that is an 91% accuracy rate, which is more than enough for me to believe that the dominant conference wins Super Bowls.

What about the other three years? Another good indicator of which conference is dominant is the number of 10+ win teams that a conference has...using that trend in addition to the winning percentage trend, you only end up with ONE Super Bowl that doesn't pan out (1977, Dallas over Denver).

51
by Daniel Warehall (not verified) :: Wed, 10/05/2005 - 6:24pm

Doug,

How are you determining which is the "dominant" conference?

52
by thad (not verified) :: Thu, 10/06/2005 - 12:11am

hey Bruce
Last time I checked losing in the playoffs meant losing in the playoffs.
In 2001 the Eagles ran on 42%on their plays. The nfl average was 44
In 2002 the Eagles ran on 46% of their plays. The NFL average was 43.
In 2003 the Eagles ran on 44% of their plays. The nfl average was 45.
in 2004 the eagles ran on 39% or their plays. The nfl average was 45.
For the past 4 years combined the Eagles have run on a total of 43% of their plays and the NFL average has been 44%. For comparison the Rams have run on 38% of their plays and big surprise coming here, the Steelers have run on 51% of their plays.
This year the Eagles have run on a league low 29% of their plays, well below their historic average.
In 2005 they have an excellent passer rating so far. But they lead the league with 64 incompletions, that is horrible. If they continue in this manner I will be shocked if they reach the NFC championship game

53
by Doug (not verified) :: Thu, 10/06/2005 - 8:49am

#51 -- I created some graphs using data from 1970 to 2004, then created trend lines to show the fluctuations. The stats I used were overall conference winning percentage, inter-conference play records, and the number of 10+ win teams per conference per year.

The three graphs wind up matching almost exactly, which isn't surprising considering that all three statistical categories are pretty much reflecting the same things (how many times the AFC and NFC beat each other).

54
by Carl (not verified) :: Thu, 10/06/2005 - 6:05pm

"How are you determining which is the 'dominant' conference?"

Uhhh, good question. I'm not even sure I've ever seen the construction of the words.

"I created some graphs using data from 1970 to 2004, then created trend lines to show the fluctuations."

But this doesn't address the changing nature of the game itself, from the 1978 reforms in the rules that opened up the game to passing; the 1993 agreement on the CBA (salary cap and free agency); or the fallout from teams that didn't heed the lessons from the CBA.

These shifts haven't exactly been glacial, and I'm pretty sure they haven't been confined to "conferences," especially in the post-AFL/NFL merger, post-1978, post-1993 salary cap era.

"The stats I used were overall conference winning percentage"

So what? Why won't anyone trust Markov models?

"inter-conference play records"

Interesting but woefully incomplete. Today, teams play only 4 of ONLY 16 games out of conference. In previous years, it was even less than that.

Again, I make my plea for poor, poor Markov.

I've seen statisticians try the same argument in MLB, and it didn't wash. Maybe some of the baseball guys in here like Will can tell us about sabremetric surveys like that.

55
by B (not verified) :: Thu, 10/06/2005 - 6:10pm

I'll bite. Who's Markov and what's his model?

56
by Carl (not verified) :: Thu, 10/06/2005 - 6:10pm

Speaking of Markov, Pat, Sagarin still has Philadelphia at the top in Pure Points, which seems right.

NE is third, but I think that's deceptive because it's not necessarily predictive with their current injury issue.

Also, I think Philadelphia is one bad hit from their QB being gone for the year (well, isn't every team? Not as much as Philly's damaged goods), which would really poop up their season.

He also rates the NFC West as worse than the NFC North, which I don't think any mathematical formula could ever explain, but there it is.

57
by Carl (not verified) :: Thu, 10/06/2005 - 6:35pm

B, in this case I'm using "Markov" as a brand, not as an exact discipline.

In my mind, I'm considering a range of stastical evaluating tools, from Markov decision processes to linear optimization.

First, some vocabulary. Sagarin points to his adaption of ELO Chess on his screen. Elo is a Hungarian mathematician who developed a predictive model for chess that Sagarin has adapted for part of his football rankings.

When I mention a Markov model, I'm suggesting a combined linear regression Markov Chain model that's dynamic in structure to predict outcomes of games in the NFL.

This involves using a dynamic programming algorithm that's been pretty publicly vetted in its computational testing.

Without getting all mathy on ya, the basic question is easy: "If A beat B by X, then what will A do to C?"

So you have a lot of random variables to consider, and you want to posit a future that's conditionally independent of the past (predictively).

Andrei Markov, I should add, was a Russian mathematician who studied sequences of random variables in which future variables are determined by the present variables.

Think probability with a whole lot of series study.

Now, what I've done is taken regression and Markov and added in factors based on our historical research into injuries to fine-tune the model (no, we don't have much data at this time of year; yes, I'm kicking Vegas' ass anyway).

Aaron is doing some of these same things.

So you give me some bunk about "motivation" or "bulletin board material" or "toughness" and I give you stochastic sequences of characters!

I win!

58
by Carl (not verified) :: Thu, 10/06/2005 - 7:28pm

Ok. Now that I've reread that, I can see that it will probably deaden all conversation in here.

Disregard it.

Substitute here: Manning kicks Brady's ass!

59
by princeton73 (not verified) :: Thu, 10/06/2005 - 7:39pm

Ok. Now that I’ve reread that, I can see that it will probably deaden all conversation in here.
Disregard it.

not me! I thought it was bitchin'

60
by Andrei Markov (not verified) :: Thu, 10/06/2005 - 7:49pm

Markov kick Brady ass, da.

61
by Sue Ellen (not verified) :: Fri, 10/07/2005 - 12:40pm

Wait a second here. Smart, "mathy" and knows football? That's kinda sexy.

Are you single, Carl?