Week 4 DVOA Ratings
This week, we introduce opponent adjustments into our DVOA formula for the first time this season. Add that to another game's worth of data -- every game means a lot when there have only been four of them -- and this week's DVOA ratings look pretty scrambled compared to last week's ratings. We have a new team at No. 1, and it's probably a surprise: Tom Brady and the Tampa Bay Buccaneers are now on top.
The Buccaneers are winning with balance. The offense is good (seventh) and the defense is better (second). The offense is particularly interesting because DVOA likes the Bucs so much more than other offensive statistics. The Bucs are just 17th in yards per play at 5.87 and their 46% success rate ranks 18th. They aren't particularly low in turnovers. Their drive stats aren't impressive: 27th in yards per drive, 14th in points per drive. One big reason they are excelling is that they rank fifth in offensive DVOA on third and fourth downs, but that doesn't fully explain why their DVOA is so much different from the other stats. They get a small boost from opponent adjustments, but that doesn't explain it either. Six DPIs help. But overall, DVOA seems to believe this is an offense that is more than the sum of its parts.
The defense is easier to decipher. The Bucs have allowed a 39% success rate, better than every defense except Pittsburgh. They rank fourth in yards per play allowed at 5.12. So only Indianapolis has a better defensive DVOA than the Bucs so far, and the Colts have allowed just 4.52 yards per play (also the best in the league).
Tampa Bay is a surprise No. 1 in DVOA this week but the Bucs don't have a huge lead. While the media may leap to crown a new king of the hill every week, the fact is that there really isn't a king of the hill this year. Kansas City isn't going to finish 16-0, Green Bay isn't unstoppable, and all the undefeated teams have weaknesses. No team is dominating its opponents week after week, and that's reflected in the DVOA ratings. Tampa Bay barely gets over 30%, finishing the week at 30.1%. Usually at this point of the season, the best team is at least over 40%. Only once in the 36-year history of DVOA has there been no team over 30% after Week 4: 1994, when the Seahawks were No. 1 at 27.6% at this point. The Bucs are just a couple of percentage points ahead of No. 2 Kansas City, and the Chiefs are just one point ahead of No. 3 Baltimore, and Indianapolis and Green Bay are also above 25%. The 2020 season has a lot of very good teams so far but nobody who is having a legendary start like last year's Patriots or the 2018 Rams.
We're seeing the same effect on the other end of the DVOA ratings. The New York Giants are currently last at -32.4% DVOA. Their offense is a dumpster fire, but their defense is above average, and the Giants have had only one really huge loss (to San Francisco in Week 3). Only once in the last 36 years was the last place team at this point higher than the 2020 Giants: the 2016 Jets at -31.1%. Usually the worst team at this point is down around -50%. Last year's Dolphins were the worst team ever after Week 4, at -90.9%.
The lack of outliers extends to individual units. You may have seen a lot of stats around the Internet about how stellar the Aaron Rodgers and the Green Bay Packers offense have been this season. For example, right now the Packers are averaging 3.97 points per drive which would blow away the all-time record of 3.37 set by the 2007 Patriots. It's very impressive, and the Packers are easily No. 1 in offensive DVOA so far this year. Yet their 30.4% rating doesn't even rank within the top dozen offenses of the last decade in terms of DVOA through Week 4. Why? Because lots of teams are putting up impressive offensive numbers this season. Overall scoring around the league is way up, as are other offensive numbers, and DVOA is normalized so that every year averages exactly zero.
To show you how extreme this is, take a look at "raw VOA," which is VOA before we apply the normalization variables for each season, and compare that to VOA with normalization applied:
#Packers are doing some amazing things on offense, but they are less amazing when you consider how high offensive levels are around the league.
In "RAW" offensive VOA, no normalization for the season, Wk 1-4:
07 NE 54.4%
20 GB 52.1%
07 NE 53.9%
20 GB 30.9%
— Aaron Schatz 🏈 (@FO_ASchatz) October 6, 2020
With the opponent adjustments now added, most of the teams at the top of the DVOA ratings take a bit of a hit. The exception, the one team that goes up from VOA to DVOA, is Kansas City. However, I should note that this does expose a weakness in the way we apply opponent adjustments. Opponent adjustments are based on how well a team plays all year, and of course sometimes a team plays very differently when it has a lot of injuries and in particular a backup quarterback. Overall, the Patriots have been an average offense this season, but that's split into 20.7% with a healthy Cam Newton and -43.7% without Newton. So the Chiefs get very little adjustment for dominating the Patriots' backup quarterbacks. This is an effect that tends to become a lot less important once we've got more games under our belts, but it's worth noting especially early on.
The Patriots' poor performance drops them from 12th to 17th in DVOA, one of a number of teams to make big moves below the top ten. Cleveland moves up seven spots, Minnesota moves up 11 spots, and Carolina moves up nine. On the other hand, Arizona falls seven spots and Atlanta falls six spots.
Besides opponent adjustments, this week is also the season debut for the second weekly table that includes past and future schedule ratings. As usual early in the season, because our opponent adjustments are only at 40 percent strength, teams high in DVOA generally get listed with easy schedules and teams low in DVOA get listed with hard schedules. However, there are a few teams that stand out, both because of their schedule so far and because of how things will change going forward.
You are surely not shocked to know that Houston has played the hardest schedule so far by average DVOA of opponent, by a significant margin. The Chargers are second and Minnesota is third. New Orleans is the one team in the DVOA top 12 that has played a top-10 schedule so far (seventh).
The easiest schedule so far is also probably not shocking; it belongs to the San Francisco 49ers, who dismantled both the Jets and Giants. But things are going to get much more difficult from here on out. Based on current DVOA ratings, San Francisco has the No. 2 hardest remaining schedule in the league. The only team with a harder remaining slate of opponents? The Chicago Bears, who have played the No. 24 slate of opponents so far. Other teams whose schedules get harder from here on out include Cincinnati, Pittsburgh, and Tennessee.
The easiest remaining schedule based on current DVOA ratings belongs to the Dallas Cowboys, who have played a top-ten schedule in difficulty so far. And if you're wondering why the Chiefs have such a huge lead to win Super Bowl LV in our playoff odds simulation, one reason is that the Chiefs have the second-easiest remaining schedule based on current DVOA. Other teams where it gets easier from here on out include the Chargers (from No. 2 to No. 30!), the New York Giants, New Orleans, and Miami.
How do opponent adjustments affect the player stats so far? Here are players with particularly strong differences between DYAR and YAR:
- The biggest, most obvious difference belongs to Melvin Gordon, who jumps all the way from No. 22 to No. 2 in rushing DYAR when we add in opponent adjustments. Gordon has played the run defenses ranked 1 (PIT), 2 (TB), 7 (NYJ), and 22 (TEN) so far this year. The game against Pittsburgh has the srongest effect because Gordon averaged 3.7 yards per carry with 37% success rate while all other running backs (primarily Saquon Barkley and David Johnson) have combined for 1.5 yards per carry with 21% success rate against the Steelers. The fact that Pittsburgh has played only three games makes the opponent adjustment a bit bigger than it would be otherwise.
- Other running backs with hard schedules so far include Frank Gore, Ronald Jones, and Antonio Gibson. Easier schedules so far for Devin Singletary, Chris Carson, Alvin Kamara, and Austin Ekeler.
- As often happens early on, the worst quarterbacks so far (Carson Wentz and Daniel Jones) come out with the biggest boosts from opponent adjustments. However, Justin Herbert and Deshaun Watson have also played tougher schedules so far.
- Russell Wilson and Aaron Rodgers take a penalty from opponent adjustments, but so do Mitchell Trubisky and Joe Burrow. And Josh Allen, currently No. 1 in passing DYAR with very tiny lead on Rodgers, takes only a slight penalty from opponent adjustments.
- The opponent adjustments for wide receivers and tight ends are pretty small this year. Marquise Brown gets the biggest bump up for wide receivers, while Tim Patrick of Denver has the biggest penalty. For tight ends, Hunter Henry gets the biggest boost while George Kittle gets the biggest penalty, which you already knew if you read today's Quick Reads.
* * * * *
* * * * *
These are the Football Outsiders team efficiency ratings through four weeks of 2020, 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 40% strength; they will increase 10% every week through Week 10.
DAVE is a formula which combines our preseason forecast with current DVOA to get a more accurate projection of how a team will play the rest of the season. DAVE is currently 70% preseason forecast for teams with four games played and 78% preseason forecast for teams with three games played. This week's listed DAVE is based solely on projections with starting quarterbacks, even if that starting quarterback may miss the next week or two.
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).