Week 10 DVOA Ratings
Other advanced analytics may disagree, but the Football Outsiders DVOA ratings continue to drive the New Orleans Saints bandwagon this week. The Saints end up with 52.1% DVOA for their 27-13 win over San Francisco, which is the second-highest single game of the Saints' season (behind last week's dismantling of Tampa Bay) and the second-highest single game of Week 10 (behind Las Vegas).
The high rating is a bit of a surprise for a sloppy game without a lot of yardage on either side, but this rating was built mostly with defense and special teams. The Saints offense put up an average DVOA despite having just 4.6 yards per play because they were moving the ball consistently, albeit without explosive plays. The Saints had a 53% success rate on plays compared to 41% for the 49ers. The Saints defense limited the 49ers to 4.3 yards per play with two interceptions. The running game was particularly bad, with just 49 yards on 25 carries; we'll come back to that in a few paragraphs. On special teams, a good day from Wil Lutz and a big kickoff return from Deonte Harris overcame two fumbles on punt returns. (Remember that our adjustments for playing indoors give a bit of a penalty to offense and a bit of a bonus to defense.)
Of course, the injury to Drew Brees will put a wrench into the Saints' Super Bowl hopes, especially since we don't know how long Brees will be out. Incorporating an estimate for the injury into our playoff odds simulation, we end up with the Saints dropping to fourth in Super Bowl odds.
The new No. 1 team in Super Bowl odds is the Pittsburgh Steelers, just barely over the Kansas City Chiefs. The Steelers played the third-best single game of the week, with a 51.9% DVOA that's basically tied with the Saints. This moved them ahead of the Chiefs and up to third in DVOA. The Steelers are still behind Kansas City in our DAVE ratings that incorporate some of our preseason data, but the Steelers have the one-game lead in the standings and an easier remaining schedule: 23rd by average DVOA of opponent compared to sixth for Kansas City. We now give Pittsburgh an 8.5% chance of finishing the regular season undefeated, although it's worth noting that they are far from a dominant team and would not be favored to finish up that perfect regular season with a Super Bowl title. Even in simulations where they go 16-0, Pittsburgh makes the Super Bowl only 42% of the time and wins the Super Bowl only 21% of the time.
The big winner of the week was Las Vegas, with 63.3% DVOA despite heavy opponent adjustments for playing a bad Denver team. The Raiders were floating around as one of the "record better than DVOA" teams, similar to Chicago and Cleveland, but the win over Denver lifted the Raiders eight spots from 22nd to 14th. The biggest drops of the week by good teams were Baltimore (from 24.5% to 17.2%, though they only dropped one place from fifth to sixth) and Tennessee (from 8.9% and 12th to 2.2% and 16th). Tennessee's drop made room for teams such as Buffalo, Minnesota, and Las Vegas to move up.
Returning to the Saints, I wanted to write a little bit about run defense. Shutting down the 49ers' running game this week moved the Saints into first place in run defense this year. Vince Verhei noticed while writing this week's Quick Reads just how good the Saints run defense has been. You may recall that this year we moved to a new version of DVOA, version 7.3, where the changes included counting quarterback scrambles as pass plays. Run defense is now an even more accurate look at how teams stop planned running plays.
Right now, we have the new version of DVOA run for every year back to 1999. And in that time period, the current Saints run defense DVOA of -37.3% would be the second-best run defense measured, trailing only the 2000 Baltimore Ravens at -38.7%.
It's not just the Saints, either. Right now, six different defenses have run defense DVOA of -28.0% or better. If the season ended today, six of the top 25 run defenses of the last 22 years would be playing in 2020. That's a lot. Are run defenses really that good this year?
OK, probably not. There's actually two things going on here and if we adjust for them, we can get a better idea of how good this year's Saints run defense really is compared to run defenses of the past.
Let's start with the list of the top 12 run defenses since 1999, full season, with a couple of 2020 teams added in:
|Best Run Defense DVOA, 1999-2020|
|*through Week 10|
You may have noticed that the majority of these teams come from just the last few years. Over half of the teams listed come from just 2014 or later. There's a reason for this, and it has to do with how DVOA is calculated. All plays, both passes and runs, get thrown into team DVOA. The league average for all plays is set to 0% every year. However, remember that passes are always more efficient than runs. So overall, the league average DVOA for passes is going to be positive every year while the league average DVOA for runs is going to be negative.
And this difference between passes and runs has gotten larger in recent years. Twenty years ago, the average run defense DVOA was around -5%. This started to go down around 2007 and reached its nadir in 2015, when the average run defense DVOA was -13.9%. For 2020 so far, the average run defense DVOA is -11.2%.
So instead of comparing run defense DVOA as compared to all plays, let's look at run defense DVOA compared only to other runs. We'll do this by subtracting each team's run defense DVOA from the league average for that season. Now we get this list of top dozen run defenses instead:
|Best Run Defense DVOA vs. NFL Average, 1999-2020|
|YEAR||TEAM||DVOA||NFL AVG||vs. AVG|
|*through Week 10|
Now only four of the top dozen teams have played since 2014. You can see with this table just how special that 2000 Ravens run defense was, as they really stand out compared to other teams since 1999. I've written a few times about the fact that the 2000 Ravens are not the greatest defense ever. They did not even lead the league in defensive DVOA during the regular season; Tennessee did. But the Ravens definitely might be the greatest run defense ever.
But there's another thing to consider if we want to compare the Saints to historical run defenses. The Saints have only played nine games this year. All the other defenses on that table played 16 games. As we know, extreme stats tend to (though they don't always) regress towards the mean with a bigger sample size. So of course, the best run defenses after Week 10 are not the same as the best run defenses after Week 17. No run defense has ever been better than -40% in DVOA through an entire season. It turns out that five different run defenses since 1999 were that good through Week 10, including two Ravens teams (but not the 2000 Ravens) and last year's Tampa Bay Buccaneers.
There's a new FO+ feature that allows subscribers to see both offensive and defensive DVOA through a specific week in a past season, split into pass and run. We can use that to look at historical run defenses through Week 10 only. We can compare those ratings to the NFL average in run defense for each year through Week 10, and then see where the Saints stand among first-half defenses in recent memory.
|Best Run Defense DVOA vs. NFL Average
as of Week 10, 1999-2020
|YEAR||TEAM||DVOA||NFL AVG||vs. AVG|
That's a better look at where the Saints really stand among the best run defenses of the last 22 years. You'll notice that the 2000 Ravens aren't at the top of this list. I said that extreme stats tend to regress towards the mean, but they don't always. The 2000 Ravens run defense was great through the first half of the season and then actually got better in the second half of the season, which is why they're on top of the full-season lists above but not this list.
As for the other run defenses of 2020, they also come out lower when we compare them to both the NFL average in run defense and historical teams measured specifically through Week 10. The 2020 Buccaneers would be 21st on this list. The 2020 Ravens and Colts would be 32nd and 33rd.
* * * * *
We've made another change in our playoff odds this week, with a new, even more accurate equation to predict each team's chance of winning the remaining games. With this new equation, we'll no longer need to do the "scaling of DAVE to standard deviation" that we started a couple of weeks ago. That means the DAVE rating (later weighted DVOA) that is published on the playoff odds page will match the ratings on our regular DVOA pages.
With the new equation, the Jets (who had the week off) see their odds of finishing 0-16 go up from 11.7% to 14.9%. As for the odds I've been running of all the NFC East teams having various dismal records, we're having a technical problem with our query tool today but I'll try to post those odds here once we get things cleared up.
UPDATE: We finally got the query tool working properly.
- Odds the entire NFC East is 7-8-1 or worse: 82.8%
- Odds the entire NFC East is 6-9-1 or worse: 45.5%
- Odds the entire NFC East is 5-10-1 or worse: 9.3%
* * * * *
These are the Football Outsiders team efficiency ratings through 10 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.
Opponent adjustments are now at full strength.
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 36% preseason forecast for teams with nine games played and 24% preseason forecast for teams with 10 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).