by Aaron Schatz
The New England Patriots are back on top of the Football Outsiders DVOA ratings this week, as the San Francisco 49ers fall back one spot to No. 2. The Patriots have 56.7% DVOA for their Thursday night win over the Giants, which is their fifth game this season with single-game DVOA over 50% according to current opponent adjustments. Philadelphia and San Francisco have two such games, and no other team has more than one. The only exception on the Patriots' schedule was Week 4 against Buffalo, where they have -5.3% DVOA because of a dismal offensive performance that day. The Patriots offense has actually been fine other than that one game. They also have not been excellent except in Week 1 against Pittsburgh. But altogether, including the great Pittsburgh game and the awful Buffalo game gives New England the No. 8 offense in DVOA.
All those games over 50% give the Patriots a total DVOA over 50%, something only five other teams have ever done after six games of the season. The Pats move up one spot on the all-time total DVOA table.
|BEST TOTAL DVOA
THROUGH 6 GAMES, 1986-2019
San Francisco had its worst offensive performance of the year against the Los Angeles Rams on Sunday, bad enough to kick the 49ers out of the top spot in our ratings and kick them off the "all-time best total DVOA" table. However, it does not kick them off the "all-time best defense table." In fact, looking at the top defenses in DVOA history through five games, the top two are teams playing right now: New England and San Francisco. The Patriots do fall behind the 1991 New Orleans Saints in the all-time list of defenses through six games because the Saints had an incredible performance in Week 7 of 1991 where they forced six fumbles and picked off Philadelphia quarterbacks five times.
|BEST DEFENSIVE DVOA
THROUGH 5 GAMES, 1986-2019
|BEST DEFENSIVE DVOA
THROUGH 6 GAMES, 1986-2019
The San Francisco-Los Angeles game was one of a number of games this week that ended with surprising DVOA ratings that didn't necessarily match the final score. Because of DVOA ratings, the Rams actually ended up with the higher rating for this game, 16.9% to just 3.0% for the 49ers. Without opponent adjustments, the 49ers have the better rating, 3.1% to -13.5%, but that's still closer than you might expect for a 20-7 game that San Francisco controlled throughout. The reason the rating was close is that the Rams offense was pretty good at staying on schedule on first and second downs. The Rams had a success rate of 60% on first down and 56% on second down. By comparison, the 49ers offense only had a success rate of 39% on first down and 44% on second down.
But ugh, third and fourth down. The Rams were a mind-boggling 0-for-15 converting third and fourth down opportunities on Sunday, including 0-for-6 in short yardage (1-2 yards to go). San Francisco converted 50% of their third and fourth downs, although the 49ers also had their two turnovers on third downs. Anyway, the DVOA formula looks at the Rams and their success rate on first and second down and says "this offense was bad, but not horrible" despite the terrible performance on third down. Once you add in the adjustment for the 49ers defense being so strong, the Rams' offensive rating for this game was -13.0%, not that far below average.
Another game this week where the DVOA doesn't quite match the final score was Monday night's Green Bay-Detroit thriller. Even though that game came down to the final play, the Packers come away with 40.0% DVOA for the game and move up two spots to No. 6 in DVOA, while Detroit has -1.3% DVOA for the game and moves down three spots to No. 14. Although the score of the game was close, the Packers did outgain the Lions on average, 6.5 yards to 5.4 yards. DVOA also discounts the two very long Detroit passes a little bit because really deep passes like that are harder to replicate than a series of medium-length passes that gradually move the ball down the field. The Packers also rank high in DVOA right now because they have played the No. 3 toughest schedule so far.
DVOA is a value per play metric, not a total value metric, and it doesn't have anything in it to give value to time of possession. That means that every so often, we'll end up with a game such as Houston at Kansas City. The Chiefs come out with 36.3% DVOA for the game and stay at No. 3 overall, while the Texans come out with -3.9% DVOA and drop from 10th to 13th. Kansas City outgained the Texans 6.6 yards per play to 5.7 yards per play and had a plus-1 turnover margin. However, Houston ran 83 plays to just 47 for the Chiefs! It's rare to have a game with this much of an imbalance in the number of plays run by each team. But DVOA looks at that result and says, Kansas City is still the more efficient team, and we would expect them to be the better team in the long run based on what we saw in this game. Kansas City didn't have many "middle of the road" drives in this game where they gained a few yards and then punted. They had two three-and-outs, and a drive that gained 5 yards followed by a field goal, and a drive where they fumbled the ball away on the first play. But once they got going, they had three drives with touchdowns, plus two drives that gained 34 and 36 yards on six plays apiece.
The final game that had a surprising DVOA result was the Tua Bowl between Miami and Washington. Although the Dolphins almost won with a final two-point conversion, Washington dominated in terms of DVOA. Well, as much as you can dominate when both teams are in the negative: -9.8% DVOA for Washington, and -54.9% DVOA for Miami. Washington had more yards per play (5.4 to 3.9) and Miami had two interceptions and two fumbles while Washington had zero interceptions and one fumble. (None of the three fumbles ended up as turnovers.) Four different Miami drives went backwards, while the only Washington "drive" that went backwards was the final kneeldown.
A -54.9% rating is still the best game of the season for the Dolphins but it keeps them comfortably in first place as the worst team we've ever tracked in 34 years of DVOA ratings. Once again, they continue to be among the worst teams ever on both offense and defense. Here's a look at those tables. Ratings for the 1987 Atlanta Falcons do not include the strike games from Week 4-6.
|WORST TOTAL DVOA
THROUGH 5 GAMES, 1986-2019
|WORST OFFENSIVE DVOA
THROUGH 5 GAMES, 1986-2019
|WORST DEFENSIVE DVOA
THROUGH 5 GAMES, 1986-2019
Yes, that is the 2017 New England Patriots on the list of worst defenses ever through five games, just two years before a team with most of the same personnel would be the best defense ever tracked through five games. The Patriots that year allowed over 400 yards of offense in their first six games of the season, and only had five takeaways in their first five games. You might remember that as Stephon Gilmore's first year, and he seemed to have no idea what his assignments were in the early games leading to blatant missed coverages. Go back in time, here's an article about how bad things were. Here's another one. This is not an attempt to subject anyone to "Old Takes Exposed." It really was that bad. Here's what I wrote about it at the time. Suffice it to say, the Patriots straightened things out in the long run, although this same defense went on to allow 41 points and over 500 yards in a Super Bowl loss.
Moving back to the top of this week's DVOA ratings, you may notice the extreme imbalance between the two conferences. Although the Patriots are No. 1, there are only two other AFC teams in the top 12: Kansas City at No. 3 and Baltimore at No. 9. That means there are 9 NFC teams in the top dozen. With Houston at No. 13, that also gives us 11 NFC teams in the top 15.
Right now the median DVOA in the NFC is 9.6%, and the median rank is 11.5. Over in the AFC, the median DVOA is -2.2%, and the median rank is 20. (The gap between the conferences is roughly the same if you prefer mean to median: 5.3% for the NFC, -6.5% for the AFC.)
Is this the greatest imbalance ever between the two conferences? Nope, but it's pretty close if it continues like this for the rest of the year.
Here are the six years with the strongest imbalance between the conferences, based on the difference between the median AFC DVOA and the median NFC DVOA. (I used median instead of mean so that a single really good or really bad team couldn't overinfluence things.) Each year is listed with the median DVOA and median rank for each conference.
AFC median 16.1% DVOA (9.5)
NFC median -8.2% DVOA (20.5)
The year with by far the biggest conference imbalance, from the early days of Football Outsiders. Only one of the top 11 teams in DVOA that year was an NFC team: Philadelphia, which finished sixth. The Eagles would have finished higher, but they had already clinched the No. 1 seed in the conference with two weeks to go and sat their starters for two games. Only four teams in the NFC had winning records in 2004, which led to two 8-8 wild cards. The St. Louis Rams were a wild card despite being outscored by 73 points and finishing 30th in DVOA, making them the worst playoff team since 1986 according to our numbers. Over in the AFC there were four teams that went at least 12-4 and only four teams worse than 7-9. This imbalance stayed strong in both 2005 and 2006, which would be the next two years on this list if I wanted to write about more than six different seasons.
NFC median 3.8% DVOA (13.5)
AFC median -10.6% DVOA (22.5)
The imbalance in 2012 was stronger in DVOA than it was in the standings. The AFC still had six different teams with at least 10 wins, although that included the 11-5 Colts who were actually outscored by 30 points and finished 25th in DVOA. Seven of the top ten teams in DVOA were NFC teams, but more impressive was how much the NFC dominated the middle-of-the-road teams. On the bottom of the league, only two of the 12 worst teams by DVOA were NFC teams. Only three NFC teams had records worse than 7-9, and the Bears (sixth in DVOA) missed the playoffs by tiebreaker at 10-6.
NFC median 9.4% DVOA (11.5)
AFC median -4.9% DVOA (16.5)
It's a bit surprising there aren't more of these late '80s/early '90s years that stand out for conference imbalance, because that's what the NFL was known for back then: the domination of the NFC, at least when it came to the Super Bowl. In 1989, eight of the 14 NFC teams had winning records with seven of them at 10-6 or better. Over in the AFC, the Denver Broncos were the only team with double-digit wins at 11-5, and only six teams had winning records although another four teams went 8-8. In the DVOA ratings, eight of the top dozen teams were NFC teams.
AFC median 4.9% DVOA (13)
NFC median -8.1% DVOA (17)
Weirdly, the tables turned just one year later. With four different teams at 9-7, the AFC ended up with eight out of 14 teams finishing with winning records, including three different teams at 12-4 or better. Over in the NFC, only five teams finished with winning records. This was the first year with three wild cards, and one of them went to 8-8 New Orleans, which finished just 22nd in DVOA. Although the top two teams that year came from the NFC East, the Giants and Eagles, nine of the top 15 teams were AFC teams.
AFC median 2.7% DVOA (14)
NFC median -9.4% DVOA (20)
OK, this is a weird one. It didn't look like there was any imbalance at the top of the standings. Both conferences had 7 out of 15 teams with winning records. The difference was at the bottom. In the AFC, you had the 1-15 Jets and the 4-12 Ravens and then everyone else was 7-9 or better. Meanwhile, the NFC had six different teams that went 6-10 or worse. So in the DVOA ratings, the NFC had four of the top five teams and six of the top nine ... but then the AFC had every team between 10th and 18th. So only two of the ten worst teams in the league that year were AFC teams.
NFC median 2.3% DVOA (14.5)
AFC median -8.2% DVOA (20)
There was a strong imbalance between the conferences just two years ago, as only four teams in the AFC went 10-6 or better. Although the top ten teams in DVOA were evenly split between the conferences, 11 of the top 17 teams were NFC squads, with only three NFC teams in the bottom ten.
* * * * *
* * * * *
These are the Football Outsiders team efficiency ratings through six weeks of 2019, measured by our proprietary Defense-adjusted Value Over Average (DVOA) system that breaks down every single play and compares a team's performance to the league average based on situation in order to determine value over average. (Explained further here.)
OFFENSE and DEFENSE DVOA are adjusted to consider all fumbles, kept or lost, as equal value. SPECIAL TEAMS DVOA is adjusted for type of stadium (warm, cold, dome, Denver) and week of season. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
Because it is early in the season, opponent adjustments are only at 60 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 35 percent of DAVE for most teams (45 percent for teams with just five 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>
- 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).