DVOA Prefers 49ers in Close Final Four
NFL Divisional - The San Francisco 49ers continue to lead the NFL in our weighted DVOA ratings after a win over Dallas that was bigger than the final score indicates. After adjusting for the quality of the Cowboys, the 49ers ended up with 61% DVOA for the game, putting them close to the dominating victories by Philadelphia (74%) and Cincinnati (78%) earlier in the weekend.
That big Cincinnati victory moves the Bengals up and the Bills down, which puts the next three teams after San Francisco very close together. Kansas City and Cincinnati, in fact, are now tied in weighted DVOA before they face each other next Sunday in the AFC Conference Championship.
This year's NFL final four contains four of the top five teams from the regular season and four of the top five teams in weighted DVOA including the playoffs. The Buffalo Bills are the odd man out. That means that there has still never been a year where the top four teams in DVOA all made it to the conference championships. That's part of the fun of the NFL playoffs! In fact, there's never even been a year where the top two teams in each conference ended up in the two conference championship games, even if those teams weren't the top four teams overall.
It's sad to see Buffalo go, as it is surprisingly rare in recent years for the No. 1 team during the regular season to actually win the Super Bowl. (It last happened with the 2016 Patriots.) DVOA got some criticism over this during the weekend, but it wasn't alone. Pretty much every advanced metric had the Bills as the No. 1 or No. 2 team going into the playoffs. They were No. 1 before this weekend by regular-season DVOA, ESPN FPI, and Pro Football Reference's Simple Rating System. They were tied for No. 1 with Kansas City in PFF ELO. They were second to San Francisco in Pythagorean Wins and weighted DVOA, and they were second to Kansas City in 538 ELO. The Bills were also higher than the Bengals in EPA/play on both offense and defense during the regular season. Sometimes, you just get outplayed... and as we pointed out in the Monday livestream, outcoached as well.
The final four teams are very balanced according to the Vegas odds, with all four given roughly 25% chance to win it all. At BetMGM, for example, they've got the Chiefs at +250, Bengals at +260, Eagles at +270, and 49ers at +280. Our playoff odds simulation prefers the NFC teams but also comes out very close, with all four teams listed between 21.2% and 29.4%. I was asked this morning on Twitter whether this is the most balanced final four we've ever seen by DVOA. The answer is no, but it's a very balanced final four depending on which version of DVOA you want to look at.
Based on regular-season DVOA, this is the most balanced final four in nearly 25 years. I went back and looked at the standard deviation of DVOA for the last four teams in each season, all the way back to 1981. The most balanced year of all was 1997, where all four teams finished the regular season between 26% and 30% in DVOA. Then came 1986, 1998, and 2022 is fourth. Here's a look at the closest final fours since 1981 by regular-season DVOA, with the eventual Super Bowl champion of each year in bold:
|Closest Final Fours by Regular-Season DVOA, 1981-2022|
Which final was was the least close, you might ask? The answer is 1991. Not only was Washington the best team ever measured by DVOA, but the other three teams included Buffalo (fourth), Denver (13th), and Detroit (15th).
We can also look at which final four was closest going into the conference championships by looking at weighted DVOA. These ratings include the first two rounds of the playoffs and discount early-season games. Since they include fewer games, they naturally have a wider spread and higher standard deviation. I don't have these ratings as handy as regular-season DVOA, but I did go back and look at the last dozen years. Because San Francisco has pulled out to a big lead in weighted DVOA with its fantastic late-season performance (not to mention dropping Week 1 out of the formula), the 2022 season is not the closest final four by weighted DVOA, even in just the last dozen years. Instead, the closest final four of the last dozen years was 2018. The 2022 final four comes out third in this group.
|All Final Fours by Weighted DVOA after Divisional Round, 2011-2022|
As always, the following rules apply for postseason DVOA ratings:
- All 32 teams are ranked, whether they made the playoffs or not.
- Teams are ranked in order of weighted DVOA, not total season DVOA. Since weighted DVOA is meant to lower the strength of older games, these ratings do not include Weeks 1-5, and Weeks 6-13 are somewhat discounted.
- The ratings listed do not include the adjustments used in the ratings for our playoff odds report. Right now, Philadelphia is adjusted to account for the two Gardner Minshew games and Kansas City is adjusted to try to account for the Patrick Mahomes injury.
- Only weighted DVOA is listed for offense, defense, and special teams. Total DVOA is also listed, but adding 10 games to a 271-game sample won't change things very much.
- Teams are treated as having a bye week in any week where they did not play. Since most teams haven't played in two weeks, that means some of the ratings for non-playoff teams can start getting a little unreliable. Really, this is only to be used for playoff teams, the other teams are just there for ranking comparison purposes.
- DVOA, as always, takes a long-term view of an NFL team's performance. That means that the games of the last two weeks are just two games among many, so teams may be listed below other teams that they have beaten in the playoffs.
* * * * *
To save people some time, we remind everyone to put their angry troll hatred into the official zlionsfan angry troll hatred Mad Libs form:
<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>
If you are new to our website, you can read the explanation of how DVOA is figured here. As always, positive numbers represent more points so DEFENSE is better when it is NEGATIVE.
Teams in yellow are still alive in the playoffs. Teams in gray lost this past weekend.
Click here for a look at full-season DVOA with offensive and defensive splits.
Here's a look at the single-game DVOA ratings for the divisional round. Kansas City comes out essentially tied with Jacksonville after applying opponent adjustments, but had the higher rating before that. However, the Kansas City-Jacksonville game was much closer than the other three games this weekend. Yes, the San Francisco game with Dallas had a close final score, but DVOA does not think the game was close at all. The 49ers had more yards per play, a higher success rate on offense, better third-down efficiency, and fewer turnovers. The 49ers ended up with 98.5% in our Post-Game Win Expectancy formula.
59 comments, Last at 26 Jan 2023, 11:04am
#11 by Vincent Verhei // Jan 23, 2023 - 4:13pm
Both teams allowed 20.2 points per game. The Eagles allowed fewer yards and forced more turnovers, but the Commanders allowed the fewest first downs in the league. Eagles defense also played the second-easiest schedule in the league, Washington played the 10th-hardest. It's pretty close.
#20 by andrew // Jan 23, 2023 - 5:16pm
Among other things Washington allowed what points they did despite having to face the powerful Eagles offense twice, while the Eagles allowed as many as they did despite getting the relatively anemic Commander offense twice.
#13 by Oncorhynchus // Jan 23, 2023 - 4:25pm
Yeah, the problem with weighted DVOA is that it doesn't do a great job of keeping track of injuries. Dallas's defense started hot but dropped to slightly above average towards the end of year because of season ending injuries to the secondary. It's no coincidence that the player on a lot of the 49ers big plays was DaRon Bland.
The Eagles defense has a similar story with their starting slot CB Avonte Maddox missing 11 games, including all but 2 of the final 9 regular season games and the Giants. Safety Chauncy Gardner-Johnson missed the last 6 games of the regular season, but moved into the slot CB for the divisional game against the Giants. Josh Sweat was also out for the final Giants game of the regular season. Those absences are huge and they're responsible for a lot of the drag on weighted DVOA.
The 49ers also had the injury bug, but a lot of those (on defense) were earlier in the year.
Eagles look to have 21 of their 22 starters playing next weekend. Total DVOA is probably a better measure of their defense's strength than DVOA (and their offense given injuries to Goedert, Hurts and Johnson in the final 6 weeks).
#30 by Oncorhynchus // Jan 23, 2023 - 9:08pm
Yeah, I mean, everybody has injuries. I'm not making an excuse for the Eagles poor play (such as it is), I'm pointing out that DVOA isn't necessarily predictive if there are big changes in the roster. That's why they adjust their playoff predictions to account for QBs being out - but not other star players.
The Eagles that played the last 6 weeks of the season are not the same Eagles that will be playing next week. They look like they'll be starting a UDFA rookie at safety, but other than that they'll have all their starters back.
And here's the thing: SF defense is not good at CB. That's why they're ranked 27th or some such by deep pass DVOA. They weren't great at CB even before losing those guys. The rest of defense is quite good though and super well-coached. But you can't honestly tell me they'd be just as good on defense if they lost any one of Fred Warner, Nick Bosa, Dre Greenlaw or Talanoa "Troy Polamalu" Hufanga. DVOA wouldn't know if one of those guy tore an ACL trying to climb a greased lamppost in Center City - but betting markets would sure as hell move on that news.
I don't know if the guys the Eagles temporarily lost are as good individually as any of those 4. But I think it does explain why you can't just look at that number and think the Eagles were slumping.
#31 by Will Allen // Jan 23, 2023 - 11:32pm
Yep, the Bucs odds of beating the Chiefs in the S.B. improved immensely, when Mike Remmers started at tackle. Evaluating depth charts is still in the Stone Age, at least among sports media. Then again, what were the indicators that the Bengals oline, after signifcant injuries, were going to dominate the game Sunday?
#33 by rh1no // Jan 24, 2023 - 1:34am
The indicators for a quality performance by the Bengals' offensive line were subtle, but noticable:
- Jackson Carman played left tackle for Clemson in college, but was forced into playing guard for the Bengals during last year's playoff run. Obviously, he was terrible. Against the Bills, though, he replaced Jonah Williams at left tackle, returning to his bread-and-butter.
- Hakeem Adeniji played left tackle for Kansas. Like Carman, he was also forced into playing guard -- POORLY! -- during Cincinnati's ill-fated 2021 Super Bowl season. Adeniji replaced La'el Collins at right tackle beginning with the Bengals' season finale against the Ravens; not his preferred position, but still a more comfortable position for him than guard.
- Max Scharping was a second-round pick who started at guard for the Houston Texans for the better part of three seasons, including multiple playoff games. Obviously, the Texans are the worst-run team in the league, so it's not necessarily a feather in Scharping's cap to have played for them. On the other hand, the Texans are the worst-run team in the league, so it isn't unreasonable to think that they made a mistake in evaluating and developing Scharping's talent.
In a different thread here on FO, I commented:
"I might be able to talk myself into this motley crew being 'more than the sum of its parts' if they had any time during the regular season to develop into a cohesive unit."
Ultimately, though, I concluded that Buffalo's stout defense was likely to dismantle Cincinnati's patchwork offensive line by taking a page out of the Rams' playbook and blitzing until the Bengals broke. So I can't say I predicted the outcome of the game ... nobody outside the Bengals locker room expected the domination we saw on the gridiron on Sunday.
But if you looked hard and squinted your eyes, you could find a few reasons to believe that the 2022-2023 offensive line would outperform the 2021-2022 unit, despite some of the same players appearing on both rosters.
#47 by bravehoptoad // Jan 24, 2023 - 1:24pm
I think the 49ers were pretty good at CB before those injuries. The last time Jason Verrett was playing, his PFF score was 77.6. Emmanuel Mosely had a PFF score of 70.9 before he got injured. Charvarious Ward sits at an 83.2.
Deommodore Lenoir, their current starting CB opposite Ward? 55.6.
I mean, the SF defense is pretty good, but before their CB injuries, they were pushing historic levels of good. Like, through the first 4 games, they'd allowed 3.8 yards per play, which would have ranked #6 this century, and they were getting better.
#3 by rh1no // Jan 23, 2023 - 3:29pm
How does that Bengals win over the Bills stack up to other playoff wins? This was definitely the single best performance from the Bengals that I've ever seen, and a 78% opponent-adjusted DVOA seems pretty impressive overall.
#5 by KnotMe // Jan 23, 2023 - 3:46pm
The Bengals were 78%, Eagles were 71 % and Niners 61%. I would assume the Chiefs victory was worse since they only won by a single score.
Kinda weird that Cincy has the lowest SB odds. Mahomes injury is hard to put in however.
#28 by Noahrk // Jan 23, 2023 - 8:06pm
They do have an adjustment in the playoff odds page. I like that they have a three different sets of ratings and it's up to the observer to decide which is best or how to weigh them depending on the situation.
#10 by rh1no // Jan 23, 2023 - 4:12pm
The Bengals had a few very inconsistent games against inferior opponents (looking at you, NE & TB), which I imagine brings down their efficiency ratings. So I'm not concerned that DVOA has then as the least likely Super Bowl champion in a pool of closely-matched contenders because they havent played as consistently well on a per-drive basis as KC, SF, or PHI.
Subjectively, though, it seems like the NFC teams have benefited from playing patty-cake schedules while the Bengals have had to find their footing against better teams, leading to a wider swing in performances. I thought that running through the gauntlet of their 2022 schedule would catch up with them eventually, but after watching them dismantle the Bills, I think they've only gotten stronger as the season has gone on. Their climb up the DVOA rankings seems to reflect this.
#4 by andrew // Jan 23, 2023 - 3:43pm
Is there an analystics thing for something like Rock-Paper-Scissors? What I mean is, in theory Rock Paper and Scissors are equal, but Rock is ideally suited to face Scissors.
The Bengals are pretty close to the Chiefs, but they seem to be ideally suited for facing them? that kind of thing? Could you do some kind of similarity score for how their positions line up and and gauge how they did versus similar opponents (including them) and thereby attempt to quantify how well they are versus a given team?
#9 by turbohappy // Jan 23, 2023 - 4:10pm
This is very much a thing we see in football all the time, but it seems quite tough to analyze. It often comes down to finding success in certain personnel packages or formations or whatever and using them much more in that matchup than they would against other teams. Which seems hard to predict based upon other games.
#12 by Kaepernicus // Jan 23, 2023 - 4:13pm
Was the Giants vs. Vikings game last week the worst combined defense in playoff history? A 27th and 31st ranking is insane and the film definitely backed it up. It seems really unlikely that 2 teams with defenses that bad would ever make the playoffs, especially in the same conference.
#17 by Aaron Brooks G… // Jan 23, 2023 - 4:51pm
In terms of ranking, maybe.
In terms of worst combined defense, both KC-Houston and NE-Tennessee from 2020 were worse from a combined DVOA perspective. NE-Tennessee from 2017 was also worse. NE was the worst defense in the league for much of that year, because Matt Patricia. This is how you can gain 613 yards in the Super Bowl and lose to a backup QB.
#42 by Kaepernicus // Jan 24, 2023 - 10:55am
How can a guy be that unlikable, terrible at personnel, terrible at his primary job(DC), and still find work? He and Lombardi will always confuse me. Nepotism is weird. I guess Belichick and Farentz are good friends and Farentz's OC situation is probably the worst version of nepotism I have ever seen in major football.
#22 by ih8california // Jan 23, 2023 - 5:25pm
Is DVOA weighted between offense, defense, and ST based on average snap counts? Seems weird that if you get a big number in ST it count the same as a big number on Off/Def when it is much lower impact.
That is, when getting the "TOT" number is that simply OFF + DEF + ST or is a weight already factored in?
#29 by LionInAZ // Jan 23, 2023 - 8:09pm
This has been covered many times in the past. Total DVOA includes weightings for offense, defense and ST.
It's all covered in the Explanation of DVOA that newbies never read before commenting. Snark intentional.
#51 by Tutenkharnage // Jan 24, 2023 - 6:19pm
Once upon a time, a team’s total DVOA consisted of three parts offense, three parts defense, and one part special teams. As the league shifted more and more heavily in the offense’s favor, this ratio changed to 4:3:1, and I believe that’s where it still sits today.
Note that this recipe for calculating a team’s overall DVOA isn’t what is meant by “weighted DVOA.” That term refers to reducing the effect of early-season games as the season goes on.
#32 by Will Allen // Jan 23, 2023 - 11:42pm
I'm fine with either the Eagles or Niners winning, but if the Niners really beat the Eagles soundly, that might drive the Vikings even lower in the rankings, which would be kind of funny and enjoyable.
#44 by Oncorhynchus // Jan 24, 2023 - 11:57am
Week 1 is a mirage. Week 1 is an invitation to overreaction.
If Trey Lance sucks because Week 1 is predictive of future performance, then ...
- Tom Brady and the Buccaneers are locks for the NFCCG.
- Mitch Trubisky and the Steelers are going to be slugging it out with the Ravens for the AFC North title.
- Kirk Cousins is going to be MVP.
- Carson Wentz and the Commanders might actually work out.
Also in that Week 1 game, the 49ers offense gained more yard and their defense held their opponent to fewer yards than either unit achieved in the Divisional game against the Cowboys. The difference is -1 vs +1 turnovers. Let's hold off on declarations of sucking.
#45 by serutan // Jan 24, 2023 - 12:43pm
"The Bills suck in weather" seems pretty predictable.
I know this horse has been beaten into subatomic particles, but it's still weird to contemplate that the Bengals looked better adapted to Buffalo weather than the Bills.
#49 by JMcNally // Jan 24, 2023 - 3:36pm
Yeah. Cincy resident here. Would never want to compare the worst of what we get vs. Buffalo, especially after everything they went through this winter. But we do have a winter, and Sunday's game was nothing residents don't see a dozen times a year.
#52 by Aaron Brooks G… // Jan 25, 2023 - 9:39am
Buffalo gets a shitload of snow because of where it is on the lake, but it also doesn't get as cold, because of where it is on the lake.
\GB, Chicago, and Detroit are on the wrong side of the lake to get the same benefit
\\Detroit and Cleveland don't have as many super-cold games because those occur in the playoffs.
#54 by Mike B. In Va // Jan 25, 2023 - 9:54am
I don't actually think the weather had much to do with it, which I know goes against the grain. The defense looked equally as bad against Cincy in the Hamlin non-game and close to as bad against Miami. The system is super talent-reliant up the middle and key pieces of that talent were missing. Plus, don't play freakin' zone in the snow!
#41 by dmstorm22 // Jan 24, 2023 - 10:41am
I forgot what sport it was around where I saw this, but there's the concept of the Playoff's ultimate loser, so who loses to the team that loses to the team that loses to the eventual champion.
So for last year it would be the 2021 Patriots (who lost to the Bills, who lost to the Chiefs, who lost to the Bengals).
The year before it was the 2020 Titans (who lost to the Ravens, who lost to the Bills, who lost to the Chiefs)
The year before it was the 2019 Eagles (Seahawks --> Packers --> 49ers)
Teams alive for the distinction this year:
- Vikings (would require SF losing to either AFC team)
- Buccaneers (PHI losing SB)
- Dolphins (KC loses SB)
- Chargers (CIN loses SB)
#34 by Mike B. In Va // Jan 24, 2023 - 7:17am
Comparing this year to last year's Bills DVOA rings true - they've been sliding downhill since before the Hamlin game, and last year's team was definitely better at this point of the season. Seems that DVOA is capturing the fact that Dorsey cannot yet create a varied NFL offense and you can't rely on Allen being a magician every game.
And, ahem, The Bills defense is clearly ranked too high because have you seen Leslie Frazier in the playoffs, bro?
#55 by pigskin.lover // Jan 25, 2023 - 12:38pm
Any chance we have an idea of where PHI would likely fall if we take out the games without Hurts? Curious as to how much that hinders their Weighted DVOA.
They obvs had the pseudo bye against the Giants where they played their JV team, so it almost feels right to pull that one, but wondering if their two losses against Dallas and New Orleans (and potentially the Giants game), change the rating much
#56 by MC2 // Jan 25, 2023 - 2:14pm
It probably makes a pretty big difference, since the Playoff Odds Report (which is adjusted to account for the Minshew games) has the Eagles as a slight favorite against the 49ers, despite having a weighted DVOA that is 17% lower (pretty sure home field advantage isn't worth 17%).