DVOA Analysis
Football Outsiders' revolutionary metrics that break down every single play of the NFL season

Week 4 DVOA Ratings

by Aaron Schatz

Cincinnati remains on top of the Football Outsiders DVOA ratings after taking Week 4 off. In fact, the three teams on top of our ratings all had last week off, but Seattle and Denver closed the gap on the Bengals thanks to the introduction this week of opponent adjustments. Those adjustments are currently at 40 percent strength and will increase in strength each week until they are full strength after Week 10. If the Bengals can ward off last year's two Super Bowl teams, perhaps they'll do it with their remarkable consistency. Cincinnati's total DVOA for each of its three games so far: 38.6%, 45.3%, and 41.2%.

Obviously, a lot of talk today has been about New England because they were the most recent team to get clobbered on national television. However, the Patriots are actually one of five teams that ranked in the top eight of the 2013 DVOA ratings but currently rank in the bottom half of the 2014 DVOA ratings. Those teams are:

  • Kansas City, 2.3% DVOA, 18th, 2-2 record / No. 6 in 2013
  • San Francisco: 0.9% DVOA, 19th, 2-2 record / No. 7 in 2013
  • New Orleans: -4.9% DVOA, 21st, 1-3 record / No. 4 in 2013
  • New England: -5.8% DVOA, 23rd, 2-2 record / No. 5 in 2013
  • Carolina: -18.3% DVOA, 27th, 2-2 record / No. 3 in 2013

Yes, the Chiefs are here even though they clobbered the Patriots last night. That Week 1 loss to Tennessee sure does look strange now, doesn't it?

How much hope do these teams have of righting the ship and playing at a playoff level for the rest of the year? Obviously, the answer to that question depends on the specifics of each team's situation, but let's take a general look to see what happens to teams stuck in this same situation.

There were, of course, 192 teams ranked in the top eight of DVOA between 1989 and 2012. Yay for simple math! Four games into the next season, 52 of those teams had a DVOA below 0.0%. That's 27 percent of teams, which actually seems like a lot to me, even with the fact that NFL teams are a lot less consistent from year to year than college football teams or teams in the other major professional sports. Another ten teams had a DVOA between 0.0% and 3.0%, making them comparable to this year's 49ers and Chiefs.

Here's a look at the record of those teams:

  • 9 started 0-4
  • 19 started 1-3 (21 if we include the teams between 0.0% and 3.0%)
  • 18 started 2-2 (23 if we include the teams between 0.0% and 3.0%)
  • 6 started 3-1 (9 if we include the teams between 0.0% and 3.0%)

Here were the worst of those teams; I'm listing 11 of them because the two Tennessee teams are close together and over five DVOA percentage points away from the team ranked 12th (the 1994 Broncos).

Worst Four-Game Starts by Teams Ranked Top 8 in DVOA Previous Season, 1990-2014
    Previous Year First Four Games Final Stats
2013 NYG 9-7 13.4% 7 0-4 -48.7% 31 7-9 -15.7% 27
1996 ATL 9-7 13.9% 6 0-4 -47.0% 28 3-13 -28.7% 28
2009 CAR 12-4 18.0% 6 1-3 -42.6% 29 8-8 7.1% 15
1994 HOIL 12-4 16.3% 5 1-3 -41.8% 27 2-14 -32.3% 28
1999 ATL 14-2 18.8% 7 0-4 -40.9% 29 5-11 -19.2% 26
2005 NYJ 10-6 27.0% 4 1-3 -32.1% 28 4-12 -20.9% 26
1999 DEN 14-2 32.6% 1 0-4 -31.4% 26 6-10 8.6% 12
2007 CHI 13-3 23.9% 5 1-3 -31.3% 27 7-9 -4.8% 20
1995 PHI 7-9 10.9% 7 1-3 -29.7% 26 10-6 -5.3% 18
2009 TEN 13-3 23.8% 5 0-4 -27.0% 26 8-8 -6.6% 21
2004 TEN 12-4 22.9% 2 1-3 -26.6% 30 5-11 -17.6% 26

Only one of these teams recovered to make the playoffs: the 1995 Eagles. The good news for fans of the five teams in question is that none of them have been quite this bad so far in 2014.

Looking at the total group of teams that started with DVOA below 0.0%, 18 of those 52 teams (35 percent) recovered to make the playoffs. Of course, not all those teams are comparable to our five mediocre teams from 2014. We can toss out the 0-4 teams and the 3-1 teams. In fact, one win means so much in a 16-game NFL season, even this early, that comparisons for New Orleans should really be different than those for the other four teams. So here's a more accurate portrayal of history:

As noted above, there are 23 teams in our study group that started 2-2. Thirteen of those teams -- over 50 percent! -- recovered to make the playoffs that same season. Twelve of these teams were within five DVOA percentage points of the 2014 Patriots, and eight of those teams made it to the playoffs including the Patriots themselves twice, in 2003 (-6.6% DVOA through Week 4) and 2005 (-2.3% DVOA through Week 4). There are also 17 teams within five DVOA percentage points of the 2014 49ers or Chiefs -- this includes a few teams that are above 3.0% DVOA, of course -- and 11 of those 17 teams (65 percent) eventually made the playoffs including last year's 49ers, who started out 2-2 with -1.2% DVOA.

On the other hand... 22 of these 23 teams had higher DVOA ratings than Carolina has right now. The exception: the 2001 Colts, who were playing poorly even before Edgerrin James tore his ACL in Week 7, and finished the year 6-10. The only teams to follow a top-eight DVOA season with a similar DVOA start to the 2014 Panthers and still make the playoffs are the 1995 Eagles, as noted above; the 2002 Steelers, who started 1-3 with -21.0% DVOA before winning the AFC North at 10-5-1; and the 2007 Chargers, who started 1-3 with -14.5% DVOA before winning the AFC West at 11-5.

New Orleans has it a bit harder than these other teams because it is now 1-3 instead of 2-2. Only five of the 21 teams in our study group which started the season 1-3 eventually recovered to make the playoffs (24 percent). I just mentioned three of them above. The other two were the 1993 Oilers, who started 1-3 with a -7.9% DVOA but finished the year 12-4 and fifth in the league in DVOA, and the 2004 Packers, who started the year 1-3 with -2.2% DVOA. They didn't really get much better, finishing the year 12th with 2.8% DVOA, but managed to end the year 10-6 to win the NFC North.

I also wanted to take a look at the collapse of the Patriots offense, which is 28th in the league through four games. Here are the teams that rank lower on offense: Jets, Buccaneers, Raiders, and Jaguars. That's a scary list to be on, isn't it? Has there ever been another team to see its offense collapse this massively early in the season after such a run of success? It can be hard to find one because it's so hard to find another team with a similar run of success in the first place. The Patriots have ranked in the top five of offensive DVOA for five straight seasons. The only other teams to accomplish that feat: San Francisco (1989-1995), Dallas (1991-1995), Kansas City (2001-2005), and Indianapolis (2003-2008).

So instead, I went looking for teams that ranked in the top five for just three straight years. There are 27 such spans since 1989, including the current Patriots. Only four of these teams began the following season with offensive DVOA that ranked 20th or worse:

[ad placeholder 3]

  • The 1999 Broncos had -23.0% offensive DVOA after four games, which ranked 28th. Of course, John Elway had retired, and Terrell Davis tore his ACL. The Broncos eventually improved their offense as the young players, especially Brian Griese, got more experience. They had 9.5% offensive DVOA the rest of the way, which ranked 11th for the period between Week 5 and Week 17 of 1999. For the full season, the Broncos finished with 3.4% offensive DVOA, ranked 14th, but they missed the playoffs at 6-10.
  • The 2002 Rams had -16.8% offensive DVOA after four games, which ranked 28th. Kurt Warner got hurt in that fourth game and Marc Bulger started for half the season. Torry Holt missed five games. Marshall Faulk and Orlando Pace each missed six games. The Rams finished the year with -10.2% offensive DVOA, which ranked 26th, and a 7-9 record.
  • The 2003 49ers had -12.7% offensive DVOA after four games, which ranked 27th. Like this year's Patriots, they were embarassed in Week 4, falling 35-7 in Minnesota. The 49ers offense had all the same players as the year before, and I remember their offensive collapse being a bit of a mystery. (This was the first season of Football Outsiders.) And the 49ers were the one team on this list to give the Patriots some hope, because their offense recovered (in circumstances much more similar to the Patriots than the 1999 Broncos were). In Weeks 5-17, the 49ers had 14.2% offensive DVOA, which was the seventh-highest rating in the NFL during that 13-week stretch and much closer to the 19.3% offensive DVOA the 49ers had in 2002. The 49ers finished the year with 7.8% DVOA for all 16 games, which ranked 11th. However, they missed the playoffs at 7-9 because of a mediocre defense and poor special teams.
  • The 2012 Chargers had the infamous "what the hell went wrong with Philip Rivers" year. Marcus McNeill was hurt and Vincent Jackson was gone to Tampa Bay. They had -11.0% DVOA after four games, which ranked 20th. The Chargers finished with -10.0% offensive DVOA, which ranked 24th, and a 7-9 record.

The stories of those four teams don't exactly give great hope to Patriots fans. Perhaps they could put things back together the way the 2003 49ers did. Historically, Tom Brady has been better than Jeff Garcia, and I hear Terrell Owens is still around somewhere if the Patriots want to sign him. The Patriots also won't have the field position problems that 49ers team had. While the offense has collapsed, the Patriots special teams are once again in the NFL top ten, so at least some things stay consistent in Foxborough. The Patriots have had above-average special teams every season since 1996. That's the silver lining for Patriots fans. The cloud? Well, scroll down and notice the first schedule strength numbers of the season. The Patriots' schedule ranks 30th through four games. It ranks third the rest of the way. Uh-oh.

* * * * *

Once again in 2014, we have teamed up with EA Sports to bring Football Outsiders-branded player content to Madden 15 Ultimate Team. Each week, we'll be picking out a handful of players who starred in that week's games. Some of them will be well-known players who stood out in DVOA and DYAR. Others will be under-the-radar players who only stood out with advanced stats. We'll announce the players each Tuesday in the DVOA commentary article, and the players will be available in Madden Ultimate Team packs the following weekend. We will also tweet out images of these players from the @fboutsiders Twitter account on most Fridays. One player each week will only be available for 24 hours from the point these players enter packs on Friday.

The Football Outsiders stars for Week 4 are:

  • WR Reggie Wayne, IND: Led all Week 4 WR with 73 DYAR. Caught 7 of 8 passes for 119 yards and a TD.
  • FS Husain Abdullah, KC: 8 tackles, 3 passes defensed, and a pick-six.
  • RT Doug Free, DAL: Dallas RB had 8.3 yards per carry with a 100 percent Success Rate running right.
  • ILB C.J. Mosley, BAL: Led NFL with 11 Stops in Week 4.
  • CB Xavier Rhodes, MIN: Allowed only 2 completions on 8 targets (not counting WR screens) with 4 passes defensed and a run TFL.

* * * * *

All stats pages are now updated with Week 4 information -- or will be in the next few minutes -- including FO Premium, snap counts and playoff odds. I know a lot of people will ask about the KUBIAK fantasy projections midseason update and when it is coming out. It takes a ton of time to put together. I hope to have it out Thursday night but no promises.

* * * * *

[ad placeholder 4]

These are the Football Outsiders team efficiency ratings through four weeks of 2014, 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 strength of schedule and to 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. Because it is early in the season, opponent adjustments are only at 40 percent strength; they will increase 10 percent every week through Week 10. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE. 

DAVE is a formula which combines our preseason projection with current DVOA to get a more accurate forecast of how a team will play the rest of the season. Right now, the preseason projection makes up 45 percent of DAVE for teams with four games played, and 55 percent of DAVE for teams with three games played.

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

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

1 CIN 41.7% 1 21.2% 3 3-0 19.2% 5 -21.5% 1 1.0% 13
2 SEA 32.7% 4 23.8% 2 2-1 25.4% 2 -8.9% 8 -1.6% 21
3 DEN 30.3% 3 26.8% 1 2-1 19.7% 4 -11.2% 4 -0.7% 18
4 ATL 26.2% 2 12.6% 5 2-2 26.6% 1 10.9% 27 10.6% 2
5 BAL 21.1% 14 9.0% 7 3-1 16.1% 6 -4.7% 13 0.3% 15
6 GB 20.8% 17 18.3% 4 2-2 14.8% 8 -1.1% 18 4.9% 5
7 DET 16.2% 11 5.8% 10 3-1 4.0% 15 -20.6% 2 -8.4% 28
8 ARI 16.1% 12 5.0% 12 3-0 -4.0% 21 -14.4% 3 5.7% 4
9 PHI 10.5% 8 7.9% 8 3-1 -8.3% 22 -7.9% 11 10.9% 1
10 BUF 8.4% 6 3.7% 14 2-2 -10.8% 23 -8.6% 9 10.5% 3
11 NYG 8.3% 28 2.1% 17 2-2 1.0% 18 -11.1% 5 -3.8% 26
12 SD 7.8% 20 9.1% 6 3-1 4.5% 13 -2.5% 15 0.8% 14
13 DAL 7.1% 19 0.5% 19 3-1 12.1% 10 8.3% 23 3.3% 8
14 CHI 6.4% 5 7.6% 9 2-2 4.2% 14 -3.3% 14 -1.2% 20
15 PIT 5.1% 10 4.3% 13 2-2 12.5% 9 10.1% 26 2.7% 10
16 CLE 4.4% 16 -5.4% 22 1-2 20.8% 3 13.4% 29 -3.0% 25
17 IND 2.5% 21 0.7% 18 2-2 9.0% 11 9.9% 25 3.5% 7
18 KC 2.3% 27 -2.0% 20 2-2 5.6% 12 3.5% 19 0.2% 16
19 SF 0.9% 18 5.3% 11 2-2 1.1% 17 -9.3% 6 -9.5% 30
20 MIA -4.7% 24 -5.1% 21 2-2 -2.6% 20 -9.2% 7 -11.3% 32
21 NO -4.9% 13 2.9% 16 1-3 15.7% 7 20.2% 32 -0.5% 17
22 WAS -5.5% 9 -7.1% 23 1-3 3.3% 16 -2.3% 16 -11.1% 31
23 NE -5.8% 7 3.2% 15 2-2 -17.0% 28 -8.0% 10 3.2% 9
24 HOU -7.2% 22 -8.1% 24 3-1 -11.1% 25 -6.5% 12 -2.7% 23
25 NYJ -16.3% 23 -10.3% 27 1-3 -19.7% 29 -1.4% 17 1.9% 11
26 MIN -17.6% 30 -10.2% 26 2-2 -10.8% 24 7.9% 22 1.1% 12
27 CAR -18.3% 15 -9.4% 25 2-2 -2.5% 19 13.1% 28 -2.7% 24
28 TEN -27.9% 25 -17.9% 29 1-3 -16.0% 26 5.8% 20 -6.2% 27
29 STL -31.6% 29 -15.6% 28 1-2 -16.7% 27 14.2% 30 -0.7% 19
30 OAK -33.9% 26 -26.8% 30 0-4 -31.1% 31 6.9% 21 4.2% 6
31 TB -48.3% 32 -26.9% 31 1-3 -30.8% 30 8.8% 24 -8.6% 29
32 JAC -50.1% 31 -33.7% 32 0-4 -33.9% 32 14.3% 31 -1.8% 22
  • NON-ADJUSTED TOTAL DVOA does not include the adjustments for opponent strength or the adjustments for weather and altitude in special teams, and only penalizes offenses for lost fumbles rather than all fumbles.
  • ESTIMATED WINS uses a statistic known as "Forest Index" that emphasizes consistency as well as DVOA in the most important specific situations: red zone defense, first quarter offense, and performance in the second half when the score is close. It then projects a number of wins adjusted to a league-average schedule and a league-average rate of recovering fumbles. Teams that have had their bye week are projected as if they had played one game per week.
  • PAST SCHEDULE lists average DVOA of opponents played this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • FUTURE SCHEDULE lists average DVOA of opponents still left to play this season, ranked from hardest schedule (#1, most positive) to easiest schedule (#32, most negative). It is not adjusted for which games are home or road.
  • VARIANCE measures the statistical variance of the team's weekly DVOA performance. Teams are ranked from most consistent (#1, lowest variance) to least consistent (#32, highest variance).

1 CIN 41.7% 3-0 41.1% 4.0 1 6.5% 8 -4.7% 30 0.1% 1
2 SEA 32.7% 2-1 28.6% 2.9 3 19.6% 1 -4.5% 28 9.4% 10
3 DEN 30.3% 2-1 27.0% 2.9 4 12.5% 3 -3.2% 24 9.8% 11
4 ATL 26.2% 2-2 27.1% 3.1 2 -7.3% 24 0.7% 12 29.7% 25
5 BAL 21.1% 3-1 21.3% 2.4 12 8.2% 6 -4.6% 29 12.5% 14
6 GB 20.8% 2-2 14.3% 2.8 5 9.7% 5 -4.1% 27 21.6% 20
7 DET 16.2% 3-1 24.2% 2.6 7 -1.4% 17 -1.2% 18 12.4% 13
8 ARI 16.1% 3-0 9.9% 2.6 9 5.7% 9 4.3% 6 5.2% 6
9 PHI 10.5% 3-1 10.7% 2.5 10 -13.0% 29 0.8% 11 5.9% 8
10 BUF 8.4% 2-2 13.6% 2.3 15 0.6% 16 -2.2% 21 6.6% 9
11 NYG 8.3% 2-2 2.3% 2.3 14 4.9% 11 -1.5% 20 44.4% 30
12 SD 7.8% 3-1 16.7% 2.5 11 1.8% 15 -3.2% 25 5.8% 7
13 DAL 7.1% 3-1 9.3% 2.6 6 -15.9% 31 2.3% 8 25.9% 22
14 CHI 6.4% 2-2 5.3% 2.6 8 3.4% 13 -2.6% 22 15.5% 17
15 PIT 5.1% 2-2 7.8% 1.9 17 -10.3% 28 2.8% 7 17.2% 18
16 CLE 4.4% 1-2 4.2% 2.3 13 7.1% 7 -3.0% 23 0.4% 2
17 IND 2.5% 2-2 12.6% 1.5 25 -9.3% 26 -1.3% 19 15.3% 16
18 KC 2.3% 2-2 6.8% 2.1 16 -2.0% 19 -0.5% 17 45.5% 31
19 SF 0.9% 2-2 4.2% 1.6 23 10.0% 4 1.9% 9 15.0% 15
20 MIA -4.7% 2-2 7.1% 1.9 18 -7.2% 23 0.4% 14 11.8% 12
21 NO -4.9% 1-3 -5.1% 1.7 20 5.0% 10 0.4% 13 22.4% 21
22 WAS -5.5% 1-3 -4.9% 1.7 19 -9.6% 27 -3.3% 26 46.9% 32
23 NE -5.8% 2-2 -2.8% 1.5 26 -13.5% 30 8.8% 3 41.5% 28
24 HOU -7.2% 3-1 -2.4% 1.6 24 -5.7% 22 -5.1% 31 2.0% 4
25 NYJ -16.3% 1-3 -9.9% 1.2 28 2.4% 14 -0.3% 16 4.8% 5
26 MIN -17.6% 2-2 -15.7% 1.4 27 -4.0% 20 0.2% 15 42.1% 29
27 CAR -18.3% 2-2 -17.3% 1.6 22 -1.5% 18 7.8% 4 27.5% 24
28 TEN -27.9% 1-3 -27.2% 1.6 21 13.4% 2 -7.0% 32 32.8% 26
29 STL -31.6% 1-2 -24.4% 1.0 29 -19.6% 32 9.2% 2 27.1% 23
30 OAK -33.9% 0-4 -35.2% 0.3 31 -8.5% 25 9.3% 1 2.0% 3
31 TB -48.3% 1-3 -44.8% 0.9 30 -4.7% 21 7.1% 5 39.4% 27
32 JAC -50.1% 0-4 -50.7% 0.2 32 3.8% 12 1.3% 10 21.4% 19


137 comments, Last at 03 Oct 2014, 11:54pm

1 Re: Week 4 DVOA Ratings

Miami's ST are worse than Washington's? Have the coverage units been that bad, or do the KO return in the BIlls game and the blocked punt hurt that much?

Phil Simms is a Cretin.

2 Re: Week 4 DVOA Ratings

What are the biggest discrepancies between Run and Pass offense ever? San Diego is currently at more than 92% DVOA difference between run and pass offense.

7 Re: Week 4 DVOA Ratings

Going by numerical difference between the two DVOA values can be misleading, because a historical good pass offense can distort the numbers even with a decent supporting run offense. Even with that, 2009 San Diego is tops in both numerical and ordinal difference. They were 1st in pass offense at 59.6% and 31st in rush offense at -13.3% for a 30-place difference. Second is 2003 Tennessee, who were 1st in pass offense at 54.7% and 30th in rush offense at -11.4% for a 29-place difference.

On the flip side, 1992 Seattle had the biggest difference going the other direction, but that was solely because they had the worst pass offense in FO's database. 2004 Atlanta is far and away the winner in biggest ordinal difference at -28, with a 31st-ranked pass offense at -35.5% but 3rd-ranked rush offense at 16.9%. They were also at -18 in 2005 and -22 in 2006. Nothing says pre-prison Vick more than those numbers.

43 Re: Week 4 DVOA Ratings

Yeah, I get that the numerical difference can be misleading for a lot of reasons. I'm just interested in how extreme this one is so far.

From watching the games, this looks quite a bit worse than 2009, and the numbers back that up. However, it has only been a few games, and the offensive line chaos will not likely last all season -- and Mathews should be back at some point.

48 Re: Week 4 DVOA Ratings

I'm just happy they realized they needed a new kicker before he cost them any games. Other than that, I think the special teams aren't bad. Sam Martin has been amazing on both kickoffs and punts, and the coverage teams are decent. Jeremy Ross just needs to learn to take a touchback when it's appropriate.

4 Re: Week 4 DVOA Ratings

If the Vikings win the VOA contest against the #6 Packers, or even play it close, after getting the best (I presume) of the #4 Falcons, but losing convincingly to the #21 Saints and #23 Patriots, I think Minnie will need a "Matt Cassell sucks and Adrian Peterson was indicted 48 hours before kickoff" adjustment.

5 Re: Week 4 DVOA Ratings

Xavier Rhodes is really coming on, as is another Vikings cb, Josh Robinson, It's amazing what improved coaching can accomplish. On the other hand, the guy they added, Captain Munnerlyn, has not done well.

9 Re: Week 4 DVOA Ratings

San Diego doesn't seem like the 12th best team after whipping Seattle and barely losing on the road against a good Arizona team. Maybe their Buffalo performance really hurt them, but it seems most teams struggle the week after playing Seattle.

Atlanta doesn't seem like the 4th best team, as it seems their dominating Thurs night game has over-valued them.

BTW, not a fan of San Diego, but their ranking doesn't seem to make sense...

10 Re: Week 4 DVOA Ratings

Whipping? Seattle was within a score to the very end. Seattle turned it over on downs and there was just enough time that SD was forced to take a field goal. If any one of a number of plays had gone differently, for example SD recovered all 4 fumbles in the game, Seattle may well have won.

21 Re: Week 4 DVOA Ratings

While the score may have been close, San Diego did dominate Seattle for quite a bit of that game. Good on Seattle and poor job by the Chargers for letting the Seahawks hang around long enough to get a chance late in the game.

Also, 2 of those fumbles were behind the line of scrimmage with only Chargers around them so it isn't exactly surprising that they fell on them and Seattle recovered a fumble of their own as well.

56 Re: Week 4 DVOA Ratings

I'd be interested to see the DVOA breakdown for the game. From viewing, it looked like Seattle was clearly outplayed throughout the entire game, and that the final score was closer than the balance of play would indicate. Traditional statistics bear that out, as the Seahawks were at large TOP and yardage disadvantages.

And I've got to play devil's advocate here: if you're going to take positive plays away from the Chargers, you have to account for Percy Harvin's touchdown that clearly wasn't a touchdown. He stepped out of bounds 30 yards away from the end zone. Seattle, I believe, also had one or two fumbles that they managed to recover.

95 Re: Week 4 DVOA Ratings

They were clearly outplayed on defense, but only in the sense that San Diego Kept converting third downs. You might as well say that Seattle was clearly outplayed by Denver on the final two drives before halftime in the Super Bowl before the final play of each drive. In turn, Seattle outplayed San Diego on offense.

Also, Seattle lost a fumble on a kickoff.

73 Re: Week 4 DVOA Ratings

I'm not sure DVOA handles TOP advantages properly. Well, more specifically if it handles play count advantage.

DVOA is the per play quality of a team. If a team manages to take plays away from an opponent, and that opponent is a good one, they press their advantage (SD vs Seattle) if they take plays away from a bad one, it might actually go the other way.

In this case, I think it is potentially a real flaw in DVOA. TOP doesn't matter, but total plays do, and the ability to keep an opponent's offense off the field is not only a function of the defense.

Seattle's offense did very well against San Diego, which is reflected in DVOA, but they had so few plays that the total value (DYAR) didn't add up to a win.

Therefore, I'd be interested not in a DVOA breakdown of this game, but a DYAR breakdown.

94 Re: Week 4 DVOA Ratings

I don't think a DYAR breakdown would be very interesting, because it would simply say that San Diego outperformed Seattle; you might as well look at the final score for that. There wasn't anything like extremely long plays or an imbalance of fumble recoveries (there were zero fumbles for both teams, and no interceptions either) that DYAR would pick up on, after all.

Edit: Oops, must've looked at the wrong page. Seattle did lose a fumble on a kickoff.

105 Re: Week 4 DVOA Ratings

vs YAR?

I wonder if games with high YAR / VOA differences (due to one team with many more plays, keeping the ball away from the other), correlate with when VOA doesn't predict the winner well.

106 Re: Week 4 DVOA Ratings

I do.

I suspect in this game, Seattle outperformed or nearly matched San Diego in VOA, but underperformed in YAR.

Both offenses played better than their respective defenses (positive YAR), but San Diego's offense had almost 2x as many plays. On a per play basis, the difference would be much smaller than as a total.

12 Re: Week 4 DVOA Ratings

Going by the stats, they neither whipped Seattle nor played Arizona close. Seattle outgained them by 2.2 yards on a yard per play average, and Arizona outgained them by 1.4. DVOA doesn't like teams that meander their way through drives, as San Diego did against Seattle (17 third down attempts on only 9 drives, and 5.0 yards per play), because it's historically not sustainable.

25 Re: Week 4 DVOA Ratings

Converting 3rd downs is nice, but not having to convert 3rd downs is even better. The third downs they had to convert were of 13, 9, 9, 9, 8, 8, 8, 8, 8, 7, 7, 5, 2, 2, 2, 1 and 1 yard in length, which means there were quite a lot of failed downs before them.

61 Re: Week 4 DVOA Ratings

True, but San Diego continues to perform admirably on third downs, against some of the best defenses in the league. It may not have historical precedent, but that doesn't mean it's not sustainable. If a team continues to do it game in and game out, it may suggest that they're just really good at converting long third downs. And with what we know about Rivers' strengths, is it really so hard to believe?


If you watched the game, they beat Seattle convincingly. They also played Arizona close. San Diego led by eleven points going into the final quarter. It took an anomalous number of dropped passes to kill most of their drives. Even then, they still probably win if Rivers doesn't fumble a routine shotgun snap, pushing them out of field goal range.

Against the Bills, the Chargers totally controlled the game.

The current FO rankings underestimate them, but as the season wears on, and opponent adjustments become a bigger factor, I expect the Chargers will continue to leap up the charts. They are very, very good.

63 Re: Week 4 DVOA Ratings

I don't remember any research on third down conversion consistency in season, but it's not well correlated season to season. A big difference between third down DVOA and that of first and second down was one of the markers for regression.

74 Re: Week 4 DVOA Ratings

They are performing pretty good on other downs too, I think.

The _running game_ is not performing well on other downs, but the passing game has been very good on 1st and 2nd down, so I don't think there are any markers for regression due to down discrepancy there.

81 Re: Week 4 DVOA Ratings

Do drops not count as missed third-down conversions or something? The DVOA numbers accurately reflected their performance in the game.

Against the Bills, San Diego converted only 2-5 third downs in the first half, including failing all three times in the second quarter. The game was hardly out of hand at that time.

92 Re: Week 4 DVOA Ratings

Drops certainly count, but so do the third down conversions against Seattle. My argument is only that if one is anomalous, and subject to a regression towards the mean, it's the former, not the latter.

Anyway, as another poster mentioned, the truth is in the numbers. DVOA has San Diego as a top 3 passing offense, and the worst ranked rushing offense. Watch their games and it's clear what's going on: first down is two yards and a cloud of dust. So is second down. But Rivers seems to be pretty good at converting 3d and anything. NY Times has a nice write-up on this, where they call him the league's MVP so far.


His QB rating on third downs this season is over 120. On 3d and 10 or greater, it's 158.3 (or perfect).

96 Re: Week 4 DVOA Ratings

DVOA is suggesting that 2 failures out of every 3 plays is probably not a formula for sustained success - which is what 2 yards on first down and 2 yards on second down is. You are right that Rivers has been superb at converting, but the smart money wouldn't be on him sustaining that over the course of a season.

108 Re: Week 4 DVOA Ratings

THey don't do run-run-pass every time. There are quite a few third and 1's and third and 2's. In the Seattle game, they consistently got third and short.

The smart money is also on their offensive line and running game regressing towards average too and not being _this_ bad the whole season. If they continue to have an injury bug there, it may stay as bad. If they get healthy it will very likely get better.

111 Re: Week 4 DVOA Ratings

Also, DVOA is not just success rate +.

It knows that a two yard run on first down "failure" is a much smaller failure than -2 yards on first down. It doesn't count 2 of 3 plays as failures, it adds all of them up -- if three plays equals first down, at the end of the day, that is good.

116 Re: Week 4 DVOA Ratings

If that's all that DVOA did, why even bother with it? You could just look at the play-by-play to give you that information. The point of DVOA is so that going 0 yards, 0 yards, and 10 yards to get a first down is considered less sustainable than going 5 yards, 4 yards, 1 yard.

98 Re: Week 4 DVOA Ratings

You don't think a 10-17 conversion rate (10-16 really, considering the last one was a kneeldown) against one of the best defenses is also anomalous?

As for your second paragraph, you're only making the point that DVOA is accurate in its assessment that San Diego's offense is not operating completely optimally. If Rivers is so successful on third down he should be just as successful, if not more so, on first and second down, so if they would've thrown it more often they probably would have a better offensive DVOA rating.

101 Re: Week 4 DVOA Ratings

Regarding your first question. Based on what I've seen from Rivers' play this year (and throughout the majority of his career), it's certainly far less out-of-the-ordinary than the relative drop-a-palooza we saw from his receiving corps agains Arizona. The guy is just flat out good.

Regarding your second point, I can't say I disagree. I imagine the entire Chargers fan base moans every time they see a run on first down. That said, maybe the coaches are using the run to set up the pass in a way that DVOA can't capture. That is to say, perhaps futility in the Chargers' run game somehow facilitates the dominance we see in their passing game. Not saying I beleive that, but it's interesting to ponder on the disconnects between strategic intent and statistical evaluation.

114 Re: Week 4 DVOA Ratings

Yeah, JAX bit on some play-action big time on early downs last week. If they never ran on early downs, that probably wouldn't work.

When they have gone run-run-pass late in games, they have been unsuccessful. DVOA is showing that.

Most of the time they are not, and the passing game DVOA on first and second down is quite good. If they passed every down, it stands to reason that it would be less successful and more predictable. That said, they could probably pass _more_ -- and they certainly did against JAX.

112 Re: Week 4 DVOA Ratings

Downs are also not truly in isolation.

They got a TD on first down off a play-action that probably would not have worked had they not been doing those 1 and 2 yard runs previously.

As noted earlier here, there can be strange skews between run/pass dvoa due to how they are coupled. Running into the pile every now and then sets up play action passes. Conversely, making the defense respect the pass keeps 8 guys out of the box.

EDIT: The TD referenced was not the Seattle game.

115 Re: Week 4 DVOA Ratings

So passing on first down should be more successful than passing on third and long when the opponent knows you're going to pass, right? Right now San Diego is passing only 46.1% of the time on first down in the first half (second half ignored because they were usually ahead by then), even though they have the worst YPC on first down at 2.03 and the 3rd-best YPA at 10.85. Doing something so futile for 53.9% of the time is not "every now and then."

130 Re: Week 4 DVOA Ratings

The only point is you can't just look at those numbers in isolation because they are coupled.

If they pass 100% of the time on first down it won't be for 10.85 YPA. Even the poor runs into the pile can help make the better plays less predictable and more effective.

I'm not arguing that they would not have been more successful if they passed more -- they certainly would have had more success with more early down passes given the results. However prospectively they can not yet give up on the running game so early in the season. They need to try and make it work better, and see if they can solve some of their problems for longer term effectiveness through the rest of the season. If they can get the o-line playing at a mediocre level, and Mathews back, they should be able to get a decent running game back.

134 Re: Week 4 DVOA Ratings

A earlier article showed that play-action on early downs works regardless of how often the team runs it. They certainly don't need to run it more than they throw it given the personnel they have; once they get Mathews back it's a different story, but they certainly made it more difficult on themselves by running so much. Also, they wouldn't need to be better than 10.85 YPA; they just need to be better than the 6.14 YPP they're at for runs and passes combined.

66 Re: Week 4 DVOA Ratings

The pass offence is very, very good. The running game has been awful.

Defensively, they've had significantly upgraded play in the secondary which has improved the overall play from *abysmal* last year to *average* this.

That's a good mix in today's NFL, so the team's prospects currently look good. But they are relying on outstanding play from it's QB. Of course that can be sustained, but is obviously more fragile than if the good play was distributed more evenly across the team. If Rivers misses time, or his play declines, given the divisional opposition, the team is sunk.