Bills Among Familiar Faces Atop Week 1 DVOA
NFL Week 1 - The Buffalo Bills opened the 2022 NFL season with a convincing 31-10 opening night victory over the Los Angeles Rams, and they also open the season on top of the Football Outsiders DVOA ratings.
The most notable thing about the top of our ratings after one week is how much they resemble our projections from the preseason. The Rams are obviously a big exception, but otherwise most of the top teams in our projections had a strong Week 1. No. 2 Baltimore? Second in the preseason projections. No. 3 Los Angeles Chargers? Fourth in the preseason projections. Tampa Bay and Kansas City fill out the top five in DVOA, and they were both in the top seven of the preseason projections.
(Of course, though I'm calling our main metric DVOA here, it is actually VOA because there are no opponent adjustments right now. We do not apply opponent adjustments until after Week 4, so in Weeks 1-3 DVOA and VOA are the same thing.)
The similarity between Week 1 and our preseason forecast means there are not a lot of moves in our DAVE ratings. These are the ratings that combine our preseason projection with the results of early games to give us a better prediction of how each team will rank at the end of the year. For those who don't know the story, this metric is called DAVE as a reaction to criticism that our stats are too much alphabet soup. I mean, who can argue with a guy named Dave? (Technically, it stands for "DVOA Adjusted for Variation Early.") In these week's DAVE, the preseason forecast still counts for 93% of the rating.
Still, Week 1 did do some shifting to DAVE. Note that I've made a couple of changes in the projection part of DAVE, which for Dallas and Pittsburgh now reflect that Dak Prescott and T.J. Watt will be out half the season. There are no major climbers, although Minnesota and Cleveland each went up three spots. The big drops are the Rams, the Cowboys, and the Cardinals. The Rams went from third in the preseason projections to 10th in DAVE. The Cowboys, in part because we're now expecting Cooper Rush to start for half the season, drop from 12th in the preseason to 19th in DAVE. The Cardinals went from 21st in the preseason projections to 29th in DAVE. Oof!
There's not a lot to talk about with DVOA after just one game, so last year I used the Week 1 DVOA commentary to introduce a new formula called Post-Game Win Expectancy (PGWE). This year, I'm going to introduce a new and improved PGWE formula.
Here's the basic idea behind PGWE: How often should we expect each team to win an NFL game given how the two teams played overall? We all know there are close games where the "wrong team wins," or at least it seems like the wrong team wins. This is a measurement of that. It's an idea stolen from Bill Connelly, who does something similar for college football, although our PGWE works a bit differently from his.
The original PGWE accounted solely for VOA splits (DVOA without the opponent adjustments). The new PGWE adds in two new variables:
1) Which team ran more plays, and how many more? Efficiency stats (such as DVOA) are more predictive than volume, but volume plays an important role in who wins a particular game.
2) Which team had more penalties, and how many more? Penalties aren't as predictive as the yardage from passes and runs, but of course they play a role in wins and losses. Our penalty count includes both declined and offsetting penalties.
For a good example of the difference between the new PGWE and the old PGWE, let's look at one of last year's strangest games: Tennessee's 33-30 overtime win over Seattle back in Week 2. This is the 2021 game with the biggest difference between the new formula and the old one. The two teams had similar days on the ground but Seattle was much more efficient through the air, with 68.8% pass offense VOA compared to 39.2% for Tennessee. Overall, the Seahawks outgained the Titans per play, 7.6 to 6.4. So the old formula gave the Seahawks a PGWE of 79%.
However, this game was a really good example of last year's wacky Seattle offense that consisted of almost nothing but deep bombs and three-and-outs. So Seattle was more efficient on a per-play basis but ran only 52 plays compared to 82 plays for the Titans! The Seahawks also committed 12 penalties compared to just five for the Titans. Add in those stats with the new PGWE formula, and now we're listing Seattle with a PGWE of only 20%. If the VOA difference isn't too large, the team with more plays and fewer penalties is more likely to win. Makes sense, right?
In general, the new system of PGWE does a better job of correlation with actual winners and losers. The old system had a correlation of .79 with wins during the 2020-2021 seasons. The new system has a correlation of .82, so a little bit better. For the seven games from 2021 with the biggest gap between old and new PGWE, the new PGWE shifts things in the correct direction. But as you're about to see, switching from the old system to the new one doesn't always make things more accurate.
Here's are PGWE numbers for Week 1 of the 2022 season. Houston-Indianapolis was a tie, of course, but Houston had the slightly higher PGWE as shown below. There are some surprising results here:
|Post-Game Win Expectancy, Week 1 2022|
As you can see, not all close games are created equal when it comes to PGWE. There are some surprises on the bottom here, certainly, but let me start by looking at the surprise at the top: the Chargers ended up with a 99% PGWE despite only beating the Raiders by five points. It's kind of remarkable that the Raiders stayed in this game despite the Chargers having a massive advantage in the passing game. The Chargers had 60.6% VOA passing compared to -36.5% for the Raiders. The Chargers were getting 8.1 net yards per pass with the Raiders at 6.0 net yards per pass, plus the Raiders had three interceptions and three fumbles on passing plays: two on sacks, one on a reception. The Raiders got a bit lucky by recovering all three of those fumbles. The Chargers also ran a few more plays than the Raiders, although they had one additional penalty. On this one, VOA is the main reason why PGWE loves the Chargers and felt they beat the Raiders by a lot more than the final score indicates.
On the other hand, we've got the Giants' last-second win over the Titans. This was also a close game, but in this one PGWE feels that the losing team outplayed the winning team. Again, plays and penalties were similar between the two teams so VOA ratings are causing this result. The total gap between the teams is not very large. The Titans had 12.5% offensive VOA and the Giants had 4.6% offensive VOA. The Giants were slightly better on special teams, as well. However, historically when building the PGWE system I found that pass VOA tends to be about three times as important as run VOA when it comes to predicting who won a game. It wasn't in this game, certainly, but that's the reason why PGWE says we would be much more likely to expect the Titans to win this game based on how the teams played. Tennessee had 78.9% pass VOA and -44.5% run VOA. The Giants, on the other hand, had -12.9% pass VOA and 32.6% run VOA. Yeah, that's some difference there! It was a big comeback week for Saquon Barkley.
The Giants ended up with the lowest VOA rating of any of the winning teams in Week 1. We controversially projected the Giants as the worst team in the league, and their Week 1 win certainly makes it a lot less likely that the Giants will be picking No. 1 overall. Still, the Giants were not particularly good in Week 1, and there's no reason to believe that they're suddenly an above-average team based on their Week 1 performance.
I said that switching from the old system to the new one is not always more accurate, and a good example of that comes in the Pittsburgh upset of Cincinnati (also covered here in Any Given Sunday). Pittsburgh had -17.0% offensive VOA while Cincinnati had -33.0% offensive VOA. The Steelers also had better special teams, even though both teams lose value for those missed field goals in overtime. Put those two things together, and the old system predicted a Pittsburgh win 86% of the time. However, the new system actually says the Bengals were the more likely team to win this game with a 61% PGWE. Why? The Bengals ran 94 plays to just 59 plays for the Steelers, and the Steelers had nine penalties with just five for the Bengals! Given those two stats, the eventual Pittsburgh victory was a bit of a surprise.
We look forward to bringing you more PGWE numbers in this column and other places throughout the season; if you bought Football Outsiders Almanac 2022, you know these numbers were also added to the table in each chapter that looked back at each team's 2021 performance on a week-by-week basis.
* * * * *
Football Outsiders playoff odds are updated through Week 1. The annual stats pages are now updated with 2022 data, although some of that data can be kind of sketchy after just one week (in particular the offensive line and defensive line pages). Snap counts and the FO+ DVOA database are also now fully updated through Week 1.
A reminder that all our free stats pages, including DVOA and player position stats, require registration to view. This is not a paywall! You only need to register (for free) and then log in to the site to view these pages. While you're at it, you can get a seven-day trial of FO+ and check out the FO+ features like a deeper DVOA database, weekly fantasy projections, fantasy football research tools, Derrik Klassen's new All-32 game preview column, and picks against the spread.
* * * * *
These is the Football Outsiders Top 16 through one week of 2022, 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.) Please note that there are no opponent adjustments in DVOA until after Week 4, which is why it is listed here as VOA.
OFFENSE and DEFENSE VOA are adjusted for performance indoors and 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.
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 93% preseason forecast and 7% actual performance. DAVE ratings for Dallas and Cleveland are based on projections for the entire rest of the season, including games with both starting and backup quarterbacks. DAVE for Pittsburgh is based on a projection that has T.J. Watt returning at midseason.
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>