Week 3 DVOA Ratings
by Aaron Schatz
The three remaining undefeated teams in the NFL are also the top three teams in DVOA through Week 3: the Los Angeles Rams, the surprising Miami Dolphins, and the unstoppable Kansas City Chiefs. Then there's a big gap before we get to the rest of the league, where things are a bit more scrambled.
You'll notice a handful of 1-2 teams interspersed with the 2-1 teams in the top half of DVOA, and even an 0-3 team. The New York Jets are sitting at No. 7 with the No. 2 defense and special teams in the league. This is almost all still the effect of that huge Week 1 victory over Detroit, which had 90.9% DVOA, the highest rating of any single game this year. Then the Jets go -57.9% for their loss to Miami and 0.1% for their loss to Cleveland. They'll likely keep falling in future weeks as that Week 1 game becomes less and less of their resume. As for the Houston Texans, I discussed their first two games last week. They had their worst game of the year against the New York Giants but DVOA didn't rate it that badly (-10.4%) and so they are still sitting at No. 11, with a rating just a little lower than a week ago.
Let's talk instead about those top three teams, starting with the Kansas City Chiefs. Their offense is a juggernaut right now, blowing away all kinds of records. Because the overall offensive environment around the league is so high compared to the past, they are not blowing away the record for the best offensive DVOA we've ever measured after three weeks. But they're up there.
|BEST OFFENSIVE DVOA
THROUGH WEEK 3, 1989-2018
|WORST OFFENSIVE DVOA
THROUGH WEEK 3, 1989-2018
If Josh Rosen can play better than Sam Bradford, we will only be running one of these two tables in future weeks.
For the most part, those top offenses are teams that kept scoring with top offenses all year. The glaring exception is the 2011 Buffalo Bills and a younger version of Fitzmagic, where Ryan Fitzpatrick turned back into a pumpkin after three weeks. That team averaged 37.7 points over the first three games, then just 19.9 points the rest of the way, finishing 6-10. Tampa Bay Buccaneers, beware!
As for the Chiefs, they also have one of the top ten special teams DVOA ratings ever tracked through Week 3, but I'm going to wait a while before we start updating a table on that every week because this early, the sample size of special teams is so small. It's basically just a list of "which teams had the best single games between Week 1 and Week 3."
What about the Miami Dolphins? While everyone is paying attention to Tampa Bay and the crazy offensive numbers put up by Fitzmagic, what's going on down in south Florida is almost as wild. Miami is not only No. 2 in DVOA; they're also No. 5 in our DAVE ratings, which are still 65 percent made up of our preseason projections. That means the Dolphins now have a higher DAVE rating than the New England Patriots. Combine that with a two-game lead in the division, and our playoff odds report now puts Miami in the playoffs 82 percent of the time. They win the division 67.5 percent of the time, and they win the Super Bowl 9.6 percent of the time, which ranks third behind Los Angeles and Kansas City!
I will fully admit that this sounds completely insane. It is very unlikely that the Dolphins are really now just a 9-to-1 shot to win the Super Bowl. If I may publicly ponder my own methods, this makes me feel like it would be a good offseason project to examine and recalculate the percentages of the preseason projection that go into DAVE. In fact, many statistical prognosticators have found that you get better results by including preseason projections as part of your ratings all the way up until the end of the season. We may need to consider doing the same, perhaps creating something like DAVE or "predictive DVOA" that combines past performance with future projection all season long.
You may also be saying to yourself, "Well, of course, the Dolphins look too high because they aren't including the opponent adjustments yet. Just wait until they factor in the easy schedule that Miami started with." This is true ... but it is even more true of the Los Angeles Rams. If you're going to discount the Dolphins because of their schedule, you have to discount the Rams too.
I put together some numbers to figure out how easy each team's schedule has been so far. I looked at the average opponent rating with both DVOA (Week 1-3 performance only) and DAVE (which incorporates our preseason projections). And as great as the Rams are -- I happen to think they're the best team in the NFL right now -- they've played a very easy schedule no matter how you look at it. Oakland and Arizona are two bad teams we expect to stay bad. The Chargers have been mediocre, and our projections had them as mediocre. The Rams' past schedule ranked 31st in the league either way we measure. The only team with an easier schedule is the team the Rams are playing on Thursday night, the Minnesota Vikings, whose schedule ranks 32nd either way.
Miami's schedule has been easy as well: 24th based on DVOA of opponents, 27th based on DAVE. But what really stands out is Miami's schedule going forward. Based on current DVOA of their remaining 13 opponents, Miami has the easiest remaining schedule in the NFL. Based on current DAVE ratings -- which most importantly gives the New England Patriots a lot more credit -- Miami's remaining schedule still ranks 31st in the league.
Combine that 3-0 record with the easy remaining schedule, and you can see that Miami's excellent playoff odds are not just a product of DAVE overestimating what their early performance means for their chances of winning games in the future. To show this, I put together an additional playoff simulation featuring every team's records after Week 3, except this time I based the remaining wins and losses solely on our preseason projections. Past performance meant absolutely nothing in this simulation except for actual wins and losses. Remember, Miami's preseason DVOA projection ranked 25th in the league. New England's preseason DVOA projection ranked second.
In this simulation, Miami still makes the playoffs 55 percent of the time. They still win the AFC East 35 percent of the time. Their Super Bowl odds are much lower, as they only win the Super Bowl 2.7 percent of the time. But that's a huge change in playoff odds based solely on a two-game lead in the standings. In the preseason simulation, Miami only made the playoffs 29.5 percent of the time. Even assuming they are the same quality of team that we assumed in the preseason, Miami's odds of making the playoffs have nearly doubled because the Dolphins got off to a 3-0 start while the rest of the AFC East is 1-2. In this "what if" projection using solely the preseason forecast, only one other team even came close to improving its playoff odds this much thanks to the first three weeks of the season -- and that team exceeded the Dolphins. We listed Kansas City before the season with a 34.8 percent chance of making the playoffs. In a simulation that still assumes that Kansas City is a slightly below-average team, like we had in the preseason projections, the Chiefs would now have a 65.6 percent chance of making the playoffs thanks to their 3-0 start.
Here's the entire table of schedules with both past and future schedules based on average DVOA and DAVE rating of opponents. It's sortable!
Looking at the first three games, the Tampa Bay Buccaneers have by far the largest gap between schedule rank by current DVOA (3.0%, 14th) and schedule rank by DAVE (11.3%, 1st). That's a gap of 13 places in rank; no other team has a similar gap of more than seven places in rank in either direction. New Orleans, Philadelphia, and Pittsburgh have all undershot their preseason projections at this point, in part because of how well the Buccaneers played against them. Right now, the Bucs are just 20th in DVOA despite their 2-1 record. This past schedule rank by DAVE suggests that Tampa Bay's first three games will rise in our ratings as the opponent adjustments are put in, starting next week, and as their first three opponents (probably) begin to play closer to the level we expected before the season.
If we look based on the difference in actual rating between past schedule DVOA and DAVE, we get a bunch of teams who've played the Packers and Patriots so far: Chicago, Washington, Minnesota, Houston, and Jacksonville.
The team with the biggest gap in the other direction for the schedule so far, based on ranks, is the Los Angeles Chargers, whose schedule ranks sixth by current DVOA (11.3%) and 13th by DAVE (1.7%). Based on the ratings themselves, the biggest gap belongs to the Oakland Raiders, but their No. 1 toughest schedule based on DVOA only drops to the No. 2 toughest schedule by DAVE. Miami, Denver, and yes, even the Rams are all outperforming their preseason projections.
Now let's look at upcoming schedules for the rest of the season.
Before the season started, we had the New England Patriots projected with the easiest schedule in the league. Not anymore. However, it makes a big difference if the Dolphins and Jets are closer to the teams they've been through three games or closer to the teams that we thought they were before the season. The Patriots' remaining schedule is ranked 10th based on current DVOA ratings, but it drops to 23rd if we use DAVE ratings.
Other teams whose remaining schedules look easier if we use DAVE ratings include Houston, Denver, and Cleveland. The Browns still have the No. 2 toughest remaining schedule with DAVE ratings, but it's a lot easier than their remaining schedule using past DVOA because of Baltimore and Cincinnati outperforming their projections right now.
On the other side are the Carolina Panthers. Their schedule ranked 23rd if we use current DVOA ratings, but improves to 10th if we use current DAVE ratings. A big part of that is their December Saints-Falcons-Saints stretch, but they have the Steelers and Eagles on their schedule too.
Other teams whose remaining schedules look harder if we use DAVE ratings include Washington and Baltimore.
* * * * *
All stats pages should now be updated through Week 3, including snap counts and playoff odds. The FO Premium DVOA database is also updated through Week 3. We're still experimenting with our tables, the sticky headers and the sortable functionality, so you'll find some of them are sortable while there are others we need to fiddle with before the sortable functionality works. We're getting to it all gradually and we hope everyone is enjoying the changes. Feel free to comment about them in the discussion thread.
* * * * *
These are the Football Outsiders team efficiency ratings through three weeks of 2018, 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.
Please note that there are no opponent adjustments in DVOA until after Week 4. (It's still listed as DVOA instead of VOA because I don't feel like going through and changing all the tables manually.) In addition, our second weekly table which includes schedule strength, variation, and Estimated Wins will appear beginning after Week 4.
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 65 percent of DAVE. The projection part of DAVE has been lowered to account for quarterback changes in both Arizona and San Francisco going forward. (Yes, Sam Bradford has been awful so far, but a veteran quarterback such as Bradford will still project to be better than a rookie such as Josh Rosen over a full season.)
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>