Week 8 DVOA Ratings
A near upset victory for the New York Giants knocks the Tampa Bay Buccaneers down a few percentage points, but they're still No. 1 in the Football Outsiders DVOA ratings this week. Tampa Bay had a big enough lead that one average game was nowhere near enough to drop them from the top spot. It was Tampa's second-worst game of the year (behind the Week 1 loss to New Orleans) and the first game of the year where the Buccaneers had a positive (i.e. worse than average) defensive rating.
Pittsburgh is up a spot at No. 2, passing the Baltimore Ravens after beating them on Sunday. Kansas City and Indianapolis stay at four and five, and then Seattle moves up to sixth after a big win over San Francisco. Seattle had 50.0% DVOA for the game, which is the Seahawks' best single-game rating of the year and also any team's best single-game rating for Week 8. San Francisco drops four spots to No. 12 after the loss.
The Los Angeles Rams drop three spots to No. 10 after losing to Miami, although you might be surprised that their DVOA for the game wasn't that bad, just -14.4%. Of course, the Rams outgained the Dolphins by over 300 yards. On a per play basis, the Rams had 5.3 yards per play and held the Dolphins to just 3.0 yards per play.
Miami is now up into the top ten, moving up four spots with the win over the Rams. The Dolphins had their lowest offensive rating of the year at -31.5%, not an auspicious debut for Tua Tagovailoa -- again, 3.0 yards per play is very bad -- but the Dolphins didn't need him to play well.
On the other hand, Miami had -26.2% defensive DVOA for the game, the Dolphins' third straight game under -25%. One year after they were one of the worst defenses ever measured by DVOA, the Dolphins rank 11th in defense through eight weeks. Miami also had 36.9% special teams DVOA, the best single special teams game of the season so far for anyone. The 88-yard punt return by Jakeem Grant was huge and Grant also had a 45-yard kickoff return. Matt Haack also pinned the Rams inside the 10 four times and boomed another punt 63 yards from the Miami 15 to the Rams' 22. Miami now ranks second in special teams for the season, trailing only Baltimore.
You might notice that DAVE has Miami much lower than DVOA. The Dolphins are down at 15th in DAVE, which combines in-season performance with preseason projection to get a more accurate predictive metric. The reason is that the projection used in DAVE has been changed to use Tagovailoa as the Miami quarterback instead of Ryan Fitzpatrick. Despite the rookie performances of Joe Burrow and Justin Herbert, the expectation remains that rookies will not be very good, so the quarterback switch drags Miami down in DAVE.
Buffalo is only 14th in DVOA despite a 6-2 record, which means Miami has the highest rating of any team in the AFC East. Nonetheless, in part due to Tagovailoa's low projection, our playoff odds only give Miami a 14% chance to win the division. However, the win over the Rams doubled Miami's chances of making the playoffs, which are now up to 37.6%. Miami has the easiest remaining schedule in the league according to average DVOA of opponent, which is a bit of a surprise since the Dolphins have already played one of their two Jets games. Below Miami, the Patriots are a dismal 26th in DVOA with just 8.2% chance of making the postseason. And the Jets fell to -45.2% DVOA after they got slaughtered by the Chiefs. The gap between the Jets and No. 31 Jacksonville is now 20 percentage points, larger than the gap between No. 31 Jacksonville and the No. 20 Houston Texans.
Speaking of the New York Jets, here's an update on all the quirky odds I started tracking last week, which starts with the odds of the Jets going winless:
- New York Jets go 0-16: 8.8% (up from 6.7%)
- Pittsburgh goes 16-0: 4.9% (up from 1.7%)
- Entire NFC West makes the playoffs: 2.4% (down from 4.8%)
- Entire NFC West has a winning record: 17.2% (down from 38.4%)
- Entire NFC East has a losing record: 62.2% (down from 63.1%)
- Entire NFC East is 6-9-1 or worse: 26.2% (down from 26.4%)
- Entire NFC East is 5-10-1 or worse: 3.7% (down from 3.8%)
The NFC West's odds of fielding four winning teams really plummet this week. The Rams losing a non-division game is a big part of that, but so is a change in San Francisco's DAVE. Just as I've changed the projection part of Miami's DAVE to reflect Tagovailoa as the quarterback, I've also changed the projection part of San Francisco's DAVE to reflect the loss of both Jimmy Garoppolo and George Kittle. That change plus the loss to Seattle drops the 49ers' playoff odds in half this week, down to just 21.3%.
The odds for the NFC East to all have losing records barely change and they won't really change until we get non-division games. This week won't move things much since the Eagles are on bye, Washington plays the Giants, and the simulator already thinks that Dallas will probably lose to Pittsburgh.
A few other random notes to finish up the week's commentary:
- Think the season has been bad for Atlanta so far? The Falcons have the hardest remaining schedule in the league. Atlanta still has to play both games with Tampa Bay and New Orleans, plus they get a trip to Kansas City. But since the Falcons don't completely suck, we only give them a 34% chance of earning a top five pick in the draft. They're stuck in limbo, especially if they beat Denver this week, that will make it tough to draft one of the top quarterbacks.
- Speaking of teams that want to draft a top quarterback, Jacksonville has the second-hardest remaining schedule in the league. Carolina is third.
- The New York Giants have played the league's toughest schedule of opposing defenses so far this year. The Minnesota Vikings have played the toughest schedule of opposing offenses.
- You wouldn't think of a team that loses in heartbreaking fashion as consistent, but the Los Angeles Chargers are consistent because they lose close games in heartbreaking fashion over and over. So the Chargers have the lowest game-to-game variance in the league this year. All of their single-game DVOA ratings fall between -20% and 20%. New Orleans has also been consistent this season. Cleveland and San Francisco are the least consistent teams in the league, which should be no surprise.
- Baltimore now ranks second in run offense DVOA but only 24th in pass offense. Their running game has been more efficient than their passing game this season. The same is true for Minnesota (third running, 22nd passing).
* * * * *
Football Outsiders playoff odds, snap counts, and the FO+ database are now all updated through Week 8. In addition, we have added more data to the FO+ database. The "DVOA as of a Specific Week" view is now updated with the current version of DVOA back to 1999. This also means we have run/pass splits updated as of a specific week going back to 1999, so you can learn, for example, that Washington led the NFL in both passing and rushing through Week 8 of 1999, ahead of even the Greatest Show on Turf Rams.
* * * * *
These are the Football Outsiders team efficiency ratings through eight 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 80% 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 50% preseason forecast for teams with seven games played and 45% preseason forecast for teams with eight 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).
76 comments, Last at 11 Nov 2020, 5:00pm
#1 by Aaron Brooks G… // Nov 03, 2020 - 6:04pm
- Baltimore now ranks second in run offense DVOA but only 24th in pass offense. Their running game has been more efficient than their passing game this season. The same is true for Minnesota (third running, 22nd passing).
New England too, unsurprisingly.
And then basically the bottom 5. The Raiders are comically poorly rated at rushing, despite seeming pretty able to move the chains that way. Have they fallen off as of late?
#2 by 41WhiskyMike // Nov 03, 2020 - 6:22pm
Aaron, with an understanding that MIA DAVE has dropped slightly due to the change in preseason projection, can you explain how we go from a Total DVOA of 14.1% on the Team Efficiency page to a WEI DVOA of -10.8% on the playoff odds page?
#6 by Aaron Schatz // Nov 03, 2020 - 6:50pm
"Starting with Week 7, the playoff odds simulation rescales DAVE ratings to match the standard deviation of ratings over the average full season in order to get more accurate results. The DAVE ratings listed below are the rescaled ratings, not the ratings you will find on the DVOA stats pages."
Also, it should say DAVE instead of WEI DVOA in the tables, I'll go fix that.
#4 by Lost Ti-Cats Fan // Nov 03, 2020 - 6:30pm
"the simulator already thinks that Dallas will probably lose to Pittsburgh"
Pffff. Copy cat simulator.
Moving away from the "dog bites man" news, pray tell in how many wacky universes does the simulator think Dallas will beat Pittsburgh?
#12 by p_cj // Nov 03, 2020 - 9:02pm
I've said it before and I'll say it again. As a Steelers fan I have seen it too many times before...this is EXACTLY the kind of game they lose.
Let's see if Tomlin keeps them motivated enough not to look past Dallas.
#37 by NYChem // Nov 04, 2020 - 12:28pm
Steelers are 6-1 ATS this year so far though. And the TN 3 point win was their closest game of the year so far. 3 of the 7 wins were by multiple possessions (>8 pts).
And if you're looking for rules, they always score 26, 27, 28, or 38 points, and their opponents always score 7, 16, 21, 24, or 29 points.
#51 by Perfundle // Nov 04, 2020 - 4:15pm
Maybe it's a reference to their tendency to play down to their opponent when they're a huge favorite over the years.
Since 2012, they're 3-8 against the spread as at least 10-point favorites, with five 3-point wins.
#40 by Richie // Nov 04, 2020 - 1:42pm
Dallas is a 13.5 point underdog. Going back to 2010, underdogs of 12 to 17 points won 14 out of 136 games (10.3%). The last team to pull it off were the Atlanta Falcons last year at New Orleans as 13.5 underdogs. But Atlanta had a real QB when they did it, and I think that probably matters.
Also, being a 13.5 road underdog is probably a little different than being a 13.5 home underdog.
So if we look at just home underdogs since 2010, there were 18 teams underdogs of 12 to 17 points. Only one (6%) won the game outright. Again, it was the Saints falling victim. This time it was to the Rams in 2011. And the Rams had AJ Feeley at QB, so that might be a little more comparable to DeNucci. Anyway, recent history (with a small sample size) says the Cowboys have somewhere between a 1-in-10 and 1-in-20 chance of pulling off the upset.
#9 by DIVISION // Nov 03, 2020 - 8:10pm
There's more to life than the suburbs, Man.
For the Giants to win the division, they would have to have a quality QB capable of making those basic touch throws. The difference between Danny Dimes and Dwayne Haskins is that one organization is being patient with a project and the other has given up. I literally haven't seen good QB play from either.
Even when they were going for that 2-point conversion, I knew that they wouldn't win in OT. Didn't even come to that because the referee didn't even have enough respect for the Giants to stay with the call on the field of PI.
#8 by Raiderjoe // Nov 03, 2020 - 7:57pm
Dvoa drunk again. No way Raiders are 24th vest tewm. Played Btowns. Taught them a lesson aboit what it is lime facing a real twam. Maybe Buccaneers and Steelers are vbetter rhan Raiders. Possibly Cheifs tooo but Raiders beat them heaf to jead.
#15 by LionInAZ // Nov 03, 2020 - 10:12pm
An overwhelming need to be insulting is also an indication for counseling, Divot.
Raiderjoe is an icon here because he knows way more football history than most here. Noobs like you should show some respect before dribbling nonsense.
#38 by scraps // Nov 04, 2020 - 1:13pm
meaning radio-talk-land, puffing-out-chest, my team’s accomplishments reflects on me, etc, etc, et goddam cetera. People have pointed out to him several times that this is not the place for doing that idiocy, but he keeps on doing it. I don’t think he’s going to stop, and other than an official warning, I would think the best reaction at this point is to stop responding, which of course is somewhat hard when he offers up such fat pitches that it’s hard to not swing. It would be one thing if he offered something, anything at all of value. I’m trying to not hold the Cardinals against him; god knows they’ve had enough problems.
#39 by theslothook // Nov 04, 2020 - 1:36pm
He's certainly not the first poster who comports himself in this way. I have found they usually grow out of it or lose interest and move on. I agree, his posts are engaging enough merit a response, even if would all be better off not doing so.
#73 by bravehoptoad // Nov 06, 2020 - 5:09pm
True enough, he's an acquired taste.
Remember when there were all these "Joe" knockoffs...Broncosjoe and Clotsjoe and Steelersjoe &c &c? No one could do the fumble-fingered typing as well as the original, but that knockoff-phase went on so long that some newbies never figured out Raiderjoe was the first.
#18 by RickD // Nov 03, 2020 - 11:49pm
Playing the NFC East and AFC South. Oooh!
Challenging games left: rematch with Ravens, Bills, and Colts. Though I guess they cannot really look past divisional games. Bengals are starting to look better on offense.
I can't really knock them for having the #2 schedule instead of the #1 schedule. They miss the Pats and Chiefs but get the Bills and Broncos. One up, one down. I'll believe they can beat the Chiefs when I see it.
Of course, they might not have to.
#66 by RickD // Nov 05, 2020 - 9:39am
- All-American at Colorado
-Oxford allowed him to delay his scholarship to play in NFL
-led NFL in rushing in 1938
-went to Oxford in 1939 to study
-returned to US in 1940 after war started, two more years in NFL, led NFL in rushing again in 1940
-joined Naval Intelligence in 1941
-finished law school after war
-Deputy AG under Bobby Kennedy
-named to SCOTUS in 1962
-non-ideological justice (remember them?) served until 1993
#23 by David // Nov 04, 2020 - 4:21am
20% is ridiculously high - if it's as high as five per cent, I'd be amazed
Some things to consider - there are a lot of ways to approach this as an intellectual exercise, so these are just some random thoughts, really
First, let's think about some of the priors (bayesian type approach) - there is only one US president at any one time, and that single person serves for either four or eight years - this means that in the next forty years (Brady's projected lifespan), there will be between five and ten more US presidents. Let's be generous and say ten. Absent any other information, what are the odds that Brady will be one of those ten people - given that the pool of candidates is the approximately 300 million US citizens?
Okay, that's probably a bit unfair, let's narrow the pool of candidates down a bit - what can we say about those next ten presidents? Well, with the glaring exception of Trump, every president for the last 120 years has been a politician - specifically, all the presidents have held an elected office - either in congress, the senate, or as a governor. So, let's say that our actual pool of presidents is anyone in congress (535) or the senate (100) or a governor (50), or anyone who will be in those roles over the next forty years. Average terms are difficult to work out across all of these, so let's go with a pessimistic (for people in the roles) average of four years. That's a pool of around 7000 people. Once again, even if we assume that Brady gets to be one of those people (see below for more on this), 10 in 7000 is better than before, but it still isn't good.
But what about that outlier - maybe that's our route to getting Tom in the white house - he never stands as a political candidate, and bypasses that bigger pool. The odds still don't look great, since the only thing we really know here is that this has only worked once in the last 30 or so presidential tries. Let's be generous, and assume that no-one else is going to try this route - Tom's odds of achieving this (based on history) is still only 1 in 30.
So, based on the above, I'd put Tom's odds at less than 5% by the most generous of all calculations. This doesn't account for any more subjective analysis. In short, if you offered 1 in 5 as actual odds, I'd bite your hand off.
#42 by Richie // Nov 04, 2020 - 1:48pm
It's an interesting thought. Is Tom Brady more commonly loved or hated by NFL fans? I would think hated. Especially now that many Patriots fans have inexplicably turned on him.
We love voting celebrities into office. Do people overlook their sports-hate of Tom Brady to vote for him, or does his NFL villain status work against him in the ballot box?
#43 by theslothook // Nov 04, 2020 - 2:01pm
If he was hated, he wouldn't be in commercials. A lot of casual fans + spouses probably like Tom Brady. My wife who reviles football certainly knows who he is and if nothing else he would seem like a novelty as a candidate.
Its interesting to compare Brady with Jordan. Jordan seems to be universally beloved by everyone in the sport including all of the teams he defeated. Maybe that's just a quirk of how things work - Kobe received the same universal adoration even prior to his death while Lebron is a polarizing figure way before the BLM and China controversies.
In any case, I don't know if actual fans would vote against Brady just because they didn't like him as a player. I've rooted against Brady for a large chunk of my life, but I wouldn't have a problem voting for him as a person. He probably seems like a chill dude off the field too.
#46 by Noahrk // Nov 04, 2020 - 3:44pm
This notion that all anyone has to do to run a country successfully is be good at anything else in life and be popular may only be a symptom of something more deeply wrong, but it's going to end up running you guys' country to the ground. Not my problem, but that's my opinion.
I assume some of these people did end up being competent politicians despite a late start, but it seems to be getting out of hand. Personally, I'd hold fame against a politician rather than in his favor.
I wouldn't want Brady managing even the apartment building where I live unless he had some strong qualifications other than being great at reading defenses.
#47 by theslothook // Nov 04, 2020 - 3:51pm
Trying to hold back my cynicism, I would say that what makes a good politician doesnt necessarily make that person a good President. Politicians are about forming coalitions, balancing tradeoffs between constituencies desires and the broader political machine. Basically, knowing how to play the game.
A good President needs to hire a smart staff of hopefully non partisan leaning advisors and think hard about the pros and cons of various policies.
On the second point, I don't think Tom Brady is any more or less qualified to do that than the politicians themselves. If I may be even more cynical, a lot of the politicians who are famous and have a lot of following clamor for economic policies that are incredibly incoherent.
There was a time that political service was viewed like the army. You devoted your time to it out of service to the country, but it was never meant to be a full time profession.
#62 by The Ancient Mariner // Nov 04, 2020 - 7:22pm
I would bet the sort of delegation, big-picture thinking and overarching leadership required to be an effective QB would be good training for the sort of delegation, big-picture thinking and overarching leadership required to be a good president. I wouldn't want someone with no other qualifications, but Brady's record as an NFL QB might well be a good start.
#67 by Richie // Nov 05, 2020 - 1:52pm
Hmm. I don't really think of QB's as delegators. I think of them as selfish, wanting to do it all. Guys like Peyton Manning and Dan Marino are famous control freaks who wanted to pass the ball on every play, aren't they? That sounds like a personality that doesn't want to delegate.
Off the top of my head, Jim Harbaugh is the only former NFL QB I can think of who became a successful head coach. (I feel like I'm missing somebody obvious.) And even he seems to have some control issues.
#54 by Aaron Brooks G… // Nov 04, 2020 - 4:28pm
The available political systems seem to reward appearances, bloodshed, cronyism, and being excreted from the proper vagina with the proper sequence.
We tried competence with the Tyranny system, but it didn't work out.
#76 by GwillyGecko // Nov 11, 2020 - 4:47pm
Eisenhower didn't hold any elected office before the Presidency, but Brady's not a five star general either.
Taft did not hold elected office before the Presidency either, but was Secretary of War, Governor-General of the Philippines, and Court of Appeals Judge
Hoover did not hold elected office either, he went from Secretary of Commerce to President
All of these are within the last 120 years, but not exactly contemporary haha
#27 by Perfundle // Nov 04, 2020 - 9:49am
The Dolphins have an extreme split in their defense, with the 3rd-best passing defense but 32nd rushing defense. The Rams fell behind quite quickly and couldn't really exploit that, but with Miami's offense looking weak under Tua, teams should be pounding it on the ground against them.
The Chiefs' defense has a similar split, but their offense is too good for that strategy (though the Chargers nearly beat them with it).
#33 by Noahrk // Nov 04, 2020 - 11:43am
The only realistic way the Dolphins make the playoffs is if Tua plays a whole lot better than he did last Sunday or if they travel back in time in order to stop themselves from benching Fitz. The future, however, does look promising. It's becoming clearer and clearer that Flores is no ordinary Belichik disciple. There was vast in-season improvement last season and there has been again this year.
#56 by p_cj // Nov 04, 2020 - 4:52pm
Does DVOA overly weigh beating up on bad teams vs competitive games against good ones? How much do opponent adjustments really adjust?
For example, if the Steelers trounce the Cowboys this week would it increase their ranking more than barely beating the Titans/Ravens? Seems like wrestling a bear should "count" more than kicking a puppy.
This is not to say that I expect the Steelers to do so, they have a well documented history of underperforming in these types of games.
#57 by theslothook // Nov 04, 2020 - 5:07pm
I believe its predictive. Stomping bad teams is more indicative of team quality than I suppose narrow squeakers over good teams.
I understand the logic against this view, but also remember that in a game thats nip and tuck, the result of victory is much more impacted by randomness and high leverage plays that can go either way. In other words, a stomp is pretty definitive about what kind of team you are. A narrow victory is more open to interpretation.
#59 by Vincent Verhei // Nov 04, 2020 - 5:49pm
There is a non-adjusted VOA column in the second table so you can compare how the schedule has affected each team. The Steelers are second in both categories, but much closer to the rest of the pack when you take their past schedule into account.
However, while we haven't updated this in a looooooooooooooong time (probably should get to that in the offseason), historically the teams that have been best at kicking puppies (as you put it) are those that go on to win championships.
#75 by Dales // Nov 06, 2020 - 6:58pm
"This week won't move things much since the Eagles are on bye, Washington plays the Giants, and the simulator already thinks that Dallas will probably lose to Pittsburgh."
But what if Dallas wins? They actually might! Bad teams do occasionally beat good teams. I think I've read that here a few times :)