Bills defense

DVOA Week 5: Bills Continue Ascent to Greatness

Buffalo's dominating victory over the Kansas City Chiefs on Sunday night keeps them comfortably in the top spot of the Football Outsiders DVOA ratings after five weeks. With current opponent adjustments, Buffalo comes out with 85% DVOA for the game, which makes it the fourth-biggest victory of the season so far. Buffalo's win over Houston the week before was, of course, the biggest victory of the season.

Dallas and Arizona remain second and third with wins last week. Incidentally, Arizona's current DVOA of 31.3% is very close to the average of all 5-0 teams going back to 1983, which is 33.1%. The Cardinals are neither particularly strong nor particularly weak for an undefeated team after five weeks. Since 1983, 64 of 70 teams (91%) that started 5-0 made the postseason, so things are looking pretty good for Arizona right now despite the overall strength of their division. Our playoff odds simulation puts them in the postseason 90% of the time. The narrative of the NFC West as the strongest division in the game is going to take a hit in future weeks now that Russell Wilson is out for at least a month, but for now the NFC West still comes out very strong in DVOA. The Los Angeles Rams remain fifth in DVOA through five weeks, and even though they currently have losing records, Seattle and San Francisco rank ninth and tenth in DVOA.

Between Arizona and the rest of the NFC West, you'll find that Tampa Bay jumps up from ninth to fourth after walloping Miami. Cleveland, New Orleans, and Baltimore finish out the top ten, ranking sixth through eighth.

Returning to the No. 1 Buffalo Bills, Buffalo's DVOA this week (49.6%) is basically the same as their DVOA last week (49.8%) but most teams with ultra-high DVOA early in the year are going to see their ratings drop with each successive game added to the sample. Buffalo did not, and so the Bills climb onto our list of the best teams ever tracked by DVOA going back to 1983. A reminder that these tables are done with DVOA as it would have looked at the time; in other words, the opponent adjustments after Week 5 in past years are 50% strength, just like the opponent adjustments now. Buffalo has played an easy schedule so far, but so did most of the teams on this list.

Best Total DVOA
Through 5 Games, 1983-2021
Year Team W-L DVOA
1991 WAS 5-0 71.5%
2007 NE 5-0 68.7%
1999 STL 5-0 59.8%
2006 CHI 5-0 57.5%
2019 NE 5-0 56.4%
1996 GB 4-1 55.5%
2015 ARI 4-1 54.3%
2009 NO 5-0 53.0%
2021 BUF 4-1 49.6%
2013 DEN 5-0 49.6%
2007 IND 5-0 47.7%
2001 PHI 3-2 47.4%
1998 DEN 5-0 47.4%

What happened to other teams that started as hot as the Bills? Buffalo is the 36th team in history to have a DVOA over 40% after five games. Of the other 35 teams:

  • 8 won the Super Bowl
  • 9 lost the Super Bowl
  • 13 won their division but did not make the Super Bowl
  • 2 won wild cards but did not make the Super Bowl
  • 3 missed the playoffs (2003 Buccaneers, 2005 Chargers, and 2009 Giants)

One thing in Buffalo's favor as Super Bowl favorites is the strength of their odds of getting the No. 1 seed and the only first-round bye in the AFC. We have the Bills getting the top seed 56% of simulations. This isn't just because Buffalo has been so good this year and also leads the NFL in our DAVE ratings that combine 2021 performance and our preseason projections. The Bills also have a very, very easy schedule of remaining games. Buffalo's 12 remaining opponents have an average DVOA of -13.7%. They have only two games remaining against teams that rank in the top half of DVOA, trips to New Orleans (Week 12) and Tampa Bay (Week 14). Buffalo is on pace to have a schedule rating of -13.5% which would be the third-easiest schedule on record, trailing only the 1999 Rams and the 1991 Bills.

Buffalo's offense improved from 16th to 12th this week, but of course the Bills are cruising early thanks primarily to stellar defense. Buffalo's defense is now the second best in DVOA history through five games, trailing only the 2019 New England Patriots.

Best Defensive DVOA
Through 5 Games, 1983-2021
Year Team W-L DVOA
2019 NE 5-0 -48.3%
2021 BUF 4-1 -43.2%
2019 SF 5-0 -41.9%
1991 PHI 3-2 -41.1%
1991 NO 5-0 -39.7%
2002 TB 4-1 -37.2%
1984 CHI 3-2 -36.9%
1997 SF 4-1 -36.4%
1996 GB 4-1 -35.8%
2011 BAL 4-1 -34.4%
1991 WAS 5-0 -33.1%
2008 BAL 2-3 -32.9%

The flip side of the Buffalo defense is the Kansas City defense, which has been horrendous through the first five games. The Chiefs now are the eighth-worst defense ever tracked by DVOA through five games.

Worst Defensive DVOA
Through 5 Games, 1983-2021
Year Team W-L DVOA
2019 MIA 0-5 35.8%
1998 WAS 0-5 35.7%
2006 HOU 1-4 32.4%
1998 PHI 0-5 31.3%
1992 NE 0-5 31.2%
2001 ARI 2-3 30.1%
1985 BUF 0-5 29.5%
2021 KC 2-3 29.3%
2017 NE 3-2 28.2%
2010 BUF 0-5 27.8%
2012 BUF 2-3 27.8%
2018 TB 2-3 27.7%

And yet, I think the idea that the Chiefs are no longer a good team has been overstated. The Chiefs are still 14th in DVOA despite a losing record. Even after a poor game against the Bills, their offense still ranks No. 1 on the year, ahead of the Buccaneers and Rams. Some of the issue here is one of schedule strength early. Kansas City has played the hardest schedule so far according to average DVOA of opponent. It's also the No. 4 schedule of opposing defenses and the No. 2 schedule of opposing offenses. So both sides of the ball don't look as good as they really are because of the level of competition. The Chiefs, for example, rank only fifth in unadjusted offensive VOA -- it's when we add the adjustments for schedule and fumble recovery luck that they move into the top spot.

There's never been a team like the Chiefs that was so good on offense and so bad on defense this early in the season, but I did go looking for similar teams that were not as extreme. I started by looking for other teams that had offensive and defensive DVOA both over 25% after five games. There's one such team: the 2013 San Diego Chargers, who ranked third in offense and last in defense after five games. The Chargers started 2-3, like the Chiefs. They went on to finish 9-7 and made the playoffs, upsetting the Bengals in the wild-card round and then losing to Peyton Manning and the Denver Broncos in the divisional round. The Chargers finished that season second in offensive DVOA and last in defensive DVOA, although their defensive rating improved from 27.3% after Week 5 to 19.3% at the end of the year.

If we want more comparisons, we have to expand our pool of teams. So I went looking for similar teams by rank: teams that ranked in the top three for offense and the bottom three for defense after five games. Here's our list, along with a look at where their offense and defense ended the year.

Great Offense, Awful Defense Through 5 Weeks, 1983-2021
Year Team W-L DVOA Rk Off Rk Def Rk Final
W-L
Final
Off
Rk Final
Def
Rk
1984 DET 1-4 1.7% 17 26.8% 3 21.2% 27 4-11-1 -2.6% 14 7.7% 20
1986 MIA 1-4 -9.2% 19 21.6% 3 22.7% 27 8-8 28.5% 1 23.0% 27
1987 MIA 3-2 9.9% 9 27.7% 1 18.1% 26 7-5 23.0% 2 24.8% 28
1988 PHX 3-2 14.3% 8 34.2% 2 17.5% 28 7-9 6.6% 5 7.1% 22
1989 GB 3-2 3.8% 14 24.7% 2 20.3% 28 10-6 9.1% 6 9.3% 22
1991 DAL 3-2 1.4% 14 23.7% 2 22.9% 28 11-5 18.0% 4 12.0% 24
1996 BAL 2-3 -9.5% 20 21.8% 2 24.4% 30 4-12 22.8% 1 19.7% 30
2002 KC 3-2 22.1% 7 45.9% 1 23.5% 32 8-8 35.4% 1 15.1% 31
2011 NE 4-1 22.7% 4 37.8% 1 15.6% 30 13-3 32.6% 3 15.2% 30
2013 SD 2-3 -4.9% 19 26.4% 3 27.3% 32 9-7 23.6% 2 19.3% 32
2017 NE 3-2 -2.3% 21 23.9% 2 28.2% 32 13-3 27.9% 1 12.0% 31
2021 KC 2-3 3.2% 14 28.9% 1 29.3% 32 -- -- -- -- --

Well, that certainly is a strange mix of teams, with some forgotten teams from the 80s and a couple of Dan Marino squads, the Cowboys in the early days of the triplets, then one of the powerful early 2000s Chiefs teams, and finally two recent Patriots teams that made and lost the Super Bowl. What you may notice about this list, however, is that nearly every one of these teams improved on defense over the final three-fourths of the season. Offense is a mix, with some of the teams improving, some declining, and some staying roughly the same. But almost every defense improves, except those 80s Miami defenses.

This falls in line with something we often talk about, which is that offense is more predictive and consistent than defense. That's especially true between seasons, but it is also true within seasons. If you are the Kansas City Chiefs, you would rather be the league's best offense and worst defense than the other way around. It's likely, especially given their past history, that the Chiefs will continue to have one of the top offenses in the league for the rest of the season. It's also likely that their defense will at least somewhat improve and not be quite this bad over the next three months.

If you want to find a team that is great on defense and terrible on offense rather than the other way around, we've got a couple of those teams going this season. They aren't quite as extreme as the Chiefs. Carolina, which falls to 17th overall in DVOA this week, is currently second on defense behind Buffalo but only 25th on offense. Chicago, which is down at 22nd overall, ranks fifth on defense but 31st on offense, ahead of only the New York Jets.

* * * * *

Football Outsiders playoff odds, snap counts, and the FO+ database are now all updated through Week 5.

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, and picks against the spread.

* * * * *

Here is the Football Outsiders Top 16 through five weeks of 2021, 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 for opponent and performance indoors and consider all fumbles, kept or lost, as equal value. Because it is early in the season, opponent adjustments are currently at 50% strength. 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 65% preseason forecast and 35% actual performance. It is not currently adjusted for any backup quarterbacks.

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>

RK TEAM TOTAL
DVOA
LAST
WEEK
TOTAL
DAVE
RANK W-L OFF.
DVOA
OFF.
RANK
DEF.
DVOA
DEF.
RANK
S.T.
DVOA
S.T.
RANK
1 BUF 49.6% 1 24.3% 1 4-1 5.8% 12 -43.2% 1 0.5% 16
2 DAL 33.9% 2 15.5% 3 4-1 22.3% 4 -10.3% 6 1.2% 12
3 ARI 31.3% 3 10.3% 9 5-0 12.2% 8 -14.5% 4 4.6% 3
4 TB 26.3% 9 22.7% 2 4-1 27.0% 2 -1.1% 17 -1.7% 23
5 LAR 23.2% 5 13.3% 6 4-1 26.4% 3 -2.3% 14 -5.6% 30
6 CLE 22.6% 4 9.5% 10 3-2 14.7% 6 -3.7% 12 4.2% 4
7 NO 17.7% 7 9.2% 11 3-2 0.8% 14 -15.8% 3 1.1% 13
8 BAL 12.7% 6 13.9% 5 4-1 11.8% 9 6.0% 22 6.9% 2
9 SEA 12.7% 11 11.1% 7 2-3 19.7% 5 8.6% 25 1.6% 10
10 SF 7.7% 13 7.5% 12 2-3 7.3% 11 -1.7% 16 -1.2% 20
11 LAC 7.3% 14 -0.5% 17 4-1 13.4% 7 1.4% 18 -4.7% 29
12 CIN 6.6% 10 -2.3% 19 3-2 -3.3% 16 -8.2% 7 1.6% 9
13 MIN 4.3% 16 4.8% 13 2-3 -3.5% 17 -7.1% 8 0.7% 15
14 KC 3.2% 8 14.6% 4 2-3 28.9% 1 29.3% 32 3.6% 5
15 GB 2.8% 17 10.8% 8 4-1 10.4% 10 2.8% 20 -4.7% 28
16 PHI 1.6% 20 -6.3% 21 2-3 -3.8% 19 -3.9% 11 1.5% 11

Click here for the full table.

Comments

79 comments, Last at 14 Oct 2021, 2:55pm

1 The Chiefs have also had…

The Chiefs have also had injuries to key personnel on their defense - Frank Clark and Charvarius Ward have missed three games, Tyrann Mathieu and Chris Jones one game apiece. If they can get all four of those guys on the field at once, the defense will likely look a lot better.

56 So have the Cowboys...

So much so that at the chargers they started Parsons at Defensive End, a position he last played in high school.  Demarcus Lawrence out with a broken foot, what has to be the league lead in visits to the Covid Protocol team (Gregory, Neal, Anae, I guess that is Texas for ya!), Wilson, Armstrong, Bernard, Gallimore, Hill.  5 of those were starters last season, though after last season's defensive performance you could only go up.  Lucky for them they seem to have hit on several rookies.

2 I have to admit I was kind…

I have to admit I was kind of expecting the early 2000s Chiefs to show up on this list of worst defenses over the first five games. Maybe my memory is being colored by the times, but that iteration felt even worse than this one.

This Chiefs offense pairs interestingly with it's defense. It's big play nature means it's not the kind of offense that will hide its defense well. On the other hand it's big play nature means it's never out of a game no matter how big a hole the defense digs them in.

6 The gap between KC and…

The gap between KC and Seattle in defensive DVOA is pretty dramatic out of the top 15 teams. You would think they'd regress to the mean, at least, over the course of the season. Then again, they're one more injury away from Jacksonville territory for the rest of the year. It's a good thing they're through most of the brutal part of their schedule.

Their offense should be fine. You can only do what the Bucs/Bills did to them with a specific set of personnel, and not many teams have that. I would expect to see a lot of 2 deep man under from teams with good D-lines, though.

3 The AFC is definitely having…

The AFC is definitely having a down year, with only 6 out of the top 16 teams and a poopfecta at the bottom of HOU, NYJ, MIA, and JAX. I was surprised to see that 2-3 ATL has somehow managed to play worse than those four.

It's clearly BUF in their own tier at the top and I would be absolutely shocked if they don't make the AFCCG, at least. 

Below that, you've got some frisky but flawed teams with CLE, LAC, BAL, and KC. I'm not sure if I buy that CIN can compete with those teams but we'll see. The Browns lost this week but I think they should actually be encouraged by that game. The offense held up even when both starting tackles were out, Baker looked a lot better, and the defense only collapsed because of all the injuries. They've got good special teams and if they can get healthy then I wouldn't be surprised to see them make a strong playoff run.

11 Yeah trying to take advantage of that betting wise

Just looked earlier and this week's games for AFC versus NFC teams but I think there was only two games one of which was Detroit home against Cincinnati. Thought this might be a good spot to look for Detroit was a little surprised to see their only getting three especially considering how heavily the Bengals were bet last week against the Packers but a lot of that must have had to do with the Green Bay injuries

12 Yeah trying to take advantage of that betting wise

Just looked earlier and this week's games for AFC versus NFC teams but I think there was only two games one of which was Detroit home against Cincinnati. Thought this might be a good spot to look for Detroit was a little surprised to see their only getting three especially considering how heavily the Bengals were bet last week against the Packers but a lot of that must have had to do with the Green Bay injuries

4 Rush DVOA

Interesting nugget: there are only three teams that have managed to produce a positive rushing DVOA this year. See if you can guess which three!

26 I only got Cleveland without…

In reply to by Drunken5yearold

I only got Cleveland without looking. I'll leave the other 2 to drive hits to the stats page (#2 will shock you! LOL ).

Related to that tidbit only 2 defenses have allowed positive rushing value this year. Can you guess those?

I'll even help. Green Bay at -2.4% rushing defense is not one of them, though they are ranked 29th. 30th is also -2.4% so were looking at some differences in the hundreths for that.

 

I guess DVOA is trying to say that passing has more value or something....

48 Oh, I absolutely knew one…

Oh, I absolutely knew one was the Chargers. Staley's defense is designed to trade off the run defense to defend the pass better, and we've already faced two great running teams in Dallas and Cleveland. I'm actually a bit surprised it's not worse, but they did have a pretty good game defending the Raiders.

The other team was KC and shouldn't be a surprise either. Man, is that KC defense bad!

50 V LV

Yep managed to avoid any big rushing plays and they only faced 15 non qb runs which helped

5 I was stunned by how well…

I was stunned by how well the Bills' defense played Kelce, and also the number of kill shots the offense took.

Then I read this, and what I saw from formations and personnel made more sense: https://readoptional.substack.com/p/bills-josh-allen-daboll-offense-scheme

Still, kudos to Leslie Frazier for getting the team to stay disciplined and make KC matriculate the ball down the field while Daboll was essentially doing the opposite. Rumors of KC's demise are exaggerated, though - I highly suspect these two teams will meet again in January.

7 Yep, exactly this

On Sunday night, the Bills frequently lined up in formations they almost never use. They knew what KC would do to try to slow down what they expected the Bills to do, and so they simply pulled a Belichick and changed things specifically for this game. To some extent, Frazier did the same thing on defense, eliminating the blitz entirely to maximize coverage, letting Hyde and Poyer stay deep to eliminate Hill as a deep threat, and going after Kelce early and often to wear him out and frustrate him. It was a really great game plan on both sides of the ball, but it's the kind of thing you can't pull off without the personnel to do it, much like teams can't just go out and annihilate the Dolphins 35-0 every week, whip the Texans 40-0, etc.

PS. Thanks for that article. Definitely gonna send it around to my friends :)

13 Thanks for sharing

I thought this is what Buffalo should've done in the AFCCG, if only to get KC out of playing dime. That Mathieu is that much more effective with six defensive backs is something I didn't know, but it doesn't surprise me. And going dime allows the defense to set the terms for what the officials will call vis-a-vis PI and (especially) defensive holding.

That Buffalo was able to so effortlessly move to such heavy packages can't help but make me wonder if they didn't intentionally hold back vs. Pittsburgh in order to further set up KC for a personnel/formation ambush like this. Moss was inactive that game. The Bills ran more 10 and 00 personnel than any team in the league in Week 1, and with the latter, they ran it more than the rest of the league combined. And Knox's role has grown considerably. Shoot, I'm not sure if Cole Beasley broke 20 snaps vs. KC.

I think to beat Buffalo, somebody's gonna have to call Daboll's formation bluffs. I don't think Buffalo will get to a 50-50 run/pass split, ever, with Josh Allen, but if you're going to play your base defense, you need two all-world corners who don't need safety help to cover Diggs or Sanders. No small task indeed.

17 In the AFCCG, Moss was out…

In the AFCCG, Moss was out injured, Diggs, Beasley, and Brown were all hurt, and Chris Jones was not. Then when Singletary and Knox had some drops, McDermott benched them for TJ Yeldon and Lee Smith. We were rolling out a much worse squad on offense and Jones was utterly destroying the interior of the OL. Scheme could have helped, but we were losing most of our 1v1 match ups. 

8 Gotta wonder about the expected win formula

I know what it says about how it's calculated, but I'm bewildered that the Bills lost a relatively close game in week 1, absolutely smoked their next four opponents, and have 2.8 expected wins. Really? You sure? :)

74 Estimated wins isn't an …

Estimated wins isn't an "overrated/underrated" thing. It's descriptive. Tells you what happened. Doesn't tell you anything more about the team, intrinsically. Just what happened to them.

The Bills are #1 in DVOA, so how'd they lose a game? Because the one game where it was close late, they kinda stunk up the joint. If they *keep* stinking up the joint in close games, they'll lose more than you'd expect for that DVOA.

But continuing to stink in those situations *and* maintaining that #1 DVOA isn't likely. Most likely the close game stuff will pull up, because it's smaller sample size.

33 I think I might understand the problem here

According to the stats page, 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."

Buffalo's defense has been fine in the red zone this year. The first-quarter offense has been spotty more often than not, but it hasn't mattered. And the second half has been close for all of, like, week 1. That's pretty much it. So if that's the case, their 2H+CLOSE performance probably doesn't look great. But I don't know how the formula works, so I'm just guessing.

73 Estimated wins isn't…

Estimated wins isn't predictive, it's descriptive. Helps to tell why a team might have more or less wins than you'd expect. The relative performance of a team in close games is basically random: high leverage (so leads to wins/losses) but not predictive.

77 I think you're making my point for me

The Bills absolutely destroyed four teams this year. How the hell would you have expected them to win fewer than three games based on their play in those five games? Let's put this in poker terms: you keep getting in at about 80% to win, and you go 4-1. That's about expected. If you kept getting your money in good, under no circumstances would your EV dip below ... well, below EV, which is what 2.8 expected wins is saying. It's basically say that a lot of the butt Buffalo has kicked to build those leads ... well, it just doesn't count. That sounds like a methodology problem to me. I get that it's descriptive; I'm saying that it did a lousy job describing their performance over the first five games.

79 The Bills absolutely…

The Bills absolutely destroyed four teams this year. How the hell would you have expected them to win fewer than three games based on their play in those five games?

Because their schedule is anomalously bad. If they had faced stronger teams and still put up those same splits (and again, those splits are not predictive, so you would not expect that) they likely would've lost more.

I'm saying that it did a lousy job describing their performance over the first five games.

Why? They played teams ranked 14, 19, 21, 28, 30 and the smallest gap between them and their opponents was 46.4% in DVOA. That's a huge gap, the equivalent of the difference between #1 and #24 last year. You absolutely would've expected them to win all of those games.

Instead, they lost one, and that one loss is basically what contributes the "close/second half." So if they had played like that against an average schedule (which, again, they wouldn't, because it's descriptive) you would've expected more losses.

Yes, the estimated wins looks like a big outlier for the Bills from what you'd expect, but so does the Steelers game itself. Which is really all it's telling you. It's like it's saying "hey, these guys played like crap in certain leveraged situations, they probably lost a game or two they shouldn't have."

(Another way you could think about it is that the loss to the Steelers is so out of character it's almost like more than one loss).

9 Those Marino Dolphins are…

Those Marino Dolphins are exactly who this years Chiefs were reminding me of but I couldn't put a finger on it.

 

I still think the 96 Packers is the best analog to this years Bills team. I got into that in last weeks DVOA article. I went a bit on the Favre/Allen comp though I didn't point out that while they were both young, clearly talented and improving that they both would still make bonehead plays mixed in with their amazing something out of nothing plays. While their growth was in different areas it was still growth that mattered. 

I pointed out that they are both under 5th year coaches that are part of franchise turn arounds from a decade or more of missing the playoffs to being on the cusp of super bowl contenders in the 4th year. Packers won it in 96, Bills have a great shot this year.

Both teams have dominate defenses mostly based on the defensive line play. Sure there are other good defenders but that was the core of the defense. They are doing it a bit differently but not as much as thinking oh the Packers where White and Jones and some JAGs that year. White only had 8.5 sacks that year. Jones only had 5.0. Dotson had 5.5. Gabe Wilkins had 3, Big Gilbert Brown even had 1. The rotation was not as deep as Buffalo, but they had more of a rotation and more contributors than was usual at the time.

Bills have solid young corners and a great veteran safety. Packers had solid young corners and a great veteran safety.

Already talked about young QB's who made lots of great plays and a few bonehead plays that had shown continual improvement. While Favre couldn't run as well as Allen, he did have at least 2 rushing TD every year from 94 - 96. Ran for 180 - 210 yards at ~4.6 ypc most of his early seasons, defenses did care about his scambling ability though to a much lesser extent he was a Mahomes/Rodgers scrambler, not a Young/Allen scrambler.

Both teams have a couple of OK but nothing special RB (you will not convince me that Levens or Bennet were ever special at running back) who were involved in a couple plays a game each in the passing game.

It's possible that's why I like watching the Bills this year, they bring back happy memories.

14 New story idea ...

I'm enthralled by the absolute suckiness of Buffalo's schedule. I don't think it's getting enough attention. I'm aware that there's the potential for their schedule to improve, perception-wise, but it's gonna involve Houston taking near-misses and making them wins (a la NE), or Pittsburgh to continue to hit some deep shots like it did against Denver.

Is Buffalo's schedule typical for a team with the 32nd-rated schedule? The gap between it and the 31st rated schedule? Or, like their defensive performance, is it unsustainable because of sample size? Some of the teams Buffalo plays do play each other, and the dominance it showed against Miami can be watered down if the Fins register a pair of 14-point wins vs. the Jets.

To take a stab at answering my own question, as a Bills fan, I do remember the 1991 season, but looking back, I didn't realize how bad the schedule was. Four playoff teams from the previous year finished .500 or worse, led by Cincy (9-7 to 3-13). The Colts went 1-15. Tampa and GB won 7 games combined. With JAX, NYJ (two games) and ATL on the schedule, among others, it seems possible that the 2021 Bills could give a run to the 1999 Rams for the weakest schedule ever.

19 The schedule is pathetic…

The schedule is pathetic despite having to play the 1st place teams from last year. Partially, that's because their own division sucks the suck out of suck and partially because, well, WFT and NFC South.

I have no idea why they got the Jags. I wonder who their interim coach will be by then, since I assume Meyer's not lasting much longer?

21 We play both South divisions…

We play both South divisions this year based on our normal rotation, and everyone except KC, Tampa, Pitt, and Tennessee are changing QBs, Head Coaches, or both. And two of those four are changing OCs.

24 Bills on pace for 3rd easiest schedule ever

I was wondering what, if any effect that an easy schedule has on a playoff team.  Then looking at the only two teams that had easier schedules, one lost the Super Bowl and the other won the Super Bowl.

I figured that after cakewalking through the season with such wins as 35-0 and 40-0 that there may be some shock to play some talented teams in the playoffs.

However, although it is a very small sample size, it seems that a team may actually benefit in the playoffs by playing a creampuff schedule.  I would love to see a top 10 list of easy schedules, we already know the top 2 and a projected number 3.  Maybe the other 7-8 teams are NE 6 times in that awful AFC East.  We know how they did in the playoffs.

I would expect the Bills to make the playoffs at this point as division winners even if they rested Josh Allen for the playoffs starting right now.

31 "Shock"

It's categorically not true. In fact, crushing cupcakes is one of the ways you know which teams are legitimate contenders. FO spells this out every year in the Almanac.

Championship teams are generally defined by their ability to dominate inferior opponents, not their ability to win close games.
Football games are often decided by just one or two plays: a missed field goal, a bouncing fumble, the subjective spot of an official on fourth-and-1. One missed assignment by a cornerback or one slightly askew pass that bounces off a receiver’s hands and into those of a defensive back 5 yards away and the game could be over. In a blowout, however, one lucky bounce isn’t going to change things. Championship teams—in both professional and college football—typically beat their good opponents convincingly and destroy the cupcakes on the schedule.

Case in point: The Bills annihilated three consecutive opponents, then were forced to play a "real team" in a hostile environment on SNF ... and promptly crushed that team, too.

I don't think FO has ever explained why this is, except that good teams are good, but I think one ancillary benefit of winning a lot of blowouts is that it's less physically and mentally demanding on the players if the game is effectively over with 10–20 minutes to go. Reducing that level of wear and tear certainly can't hurt, but it might be helping.

43 Yes you are correct about blowing out inferior opponents

In reply to by Tutenkharnage

That is a known fact, I am just wondering if it would help to play more good teams regardless of whether or not you win or lose those games.   The Bills may end up with a plus minus that is among best ever with the cupcake schedule if they are truly as good as they have shown.

With regards to the pythagorean (plus minus) Bill James the great baseball founder of baseball analytics has shown that plus minus is a better predictor of future performance than record.  Baseball analytics has the advantage of a large sample size of games. 

The reason that blowing out teams is a great indicator is simple, do you buy Baltimore’s 4-1 record over Buffalo?  Of course not, Baltimore nearly lost to KC, Detroit and Indy, clearly DVOA would indicate that if the teams traded schedules the next 5 games, Baltimore could not win two games 35-0 or 40-0, whereby Buffalo may blow out anyone.

32 New story idea?

In reply to by IdahoPat

A question I want answered (I asked it a couple days ago) but what the success rate is after timeouts! I seem to only see timeouts used mostly where the timeout user fails on the next play anyway. 

Sorry, dont know where else to put this lol

20 That's a great question and…

That's a great question and I can't find anything on it. Would love to see a new study or an update to an old one I missed.

My foggy memory on anything that had been done was that the strength of the offense still had the strongest correlation to playoff success. It's not that defense didn't matter but that as long the defense wasn't obviously flawed in some aspect that as long as was at least some value (like 2%) it was good enough and the better the offensive DVOA the better your chance for playoff success. But there didn't seem to be a set of if you have a great defense this is a good enough offense. It's not that something like that doesn't exists just that the data couldn't tease it out. Probably because offense mattered more so there weren't enough great defenses with success to find that minimum.

That could be flawed memory though as I'm having trouble find ANY of the historical correlation studies that have happened on here. Well I can find drafting success studies for some reason, but nothing else I wanted to find.

29 LaFleur's Packers Continue the Strange Trend

This is not a critique of DVOA, which correctly identified the Packers' deficiencies in 2019 and also TB's superiority last year. But the fact remains that under LaFleur they keep beating teams on the road (CIN/SF this season), having better W/L records than those teams after they beat them, yet are rated beneath them in DVOA.

The biggest reason is the LaFleur blowout. The man is 30-7-- either the best or 2nd best record at this point in a coaching career in the history of the league (Seifert inheriting Montana/Walsh/Rice/Young is right there with him). But the 7 losses have included blowout warm weather defeats to: SF in 2019 (and again in the playoffs), LA Chargers in 2019, TB in 2020 and, most recently in Jacksonville, to NO, in 2021. Most of his defeats have been blowouts-- and only Brady throwing three 2nd half INTs in the NFC Championship at Lambeau prevented another one.

And yet the team has an outstanding record in close games under LaFleur, and also at home (16-2 as of this writing)....  I frankly don't know what exactly to make of him yet. The in-game decisions Sunday were uber conservative in a couple of instances-- yet his team kept creating chance after chance to win. (The notion that the Bengals "outplayed" GB is laughable-- they stayed in the game because of a two man pass rush caused by confusion (Packers thought they heard a call for a certain type of play-- they were mistaken) leading to the Chase 70 yard scoring play at the end of the first half--  the Packers failure to cash in on two huge plays (Adams pass; Jones run) in the second half due to the play calling and two makeable Crosby misses kept the game close-- GB, as DVOA acknowledged, was clearly the better team)  Rodgers has bailed him out several times, but then again the coach deserves some credit for the play of his QB. 

He will have to come up very big in a playoff run-- and this may or may not be his last opportunity-- to be regarded anywhere near his W/L record-- will the injuries to Smith and Alexander-- both of which could be season-ending-- prevent that?

35 Flip flop

2019 - frauds

2020 - legit

2021 - frauds?

Although I believe most of 2021 can be chalked up to injuries (more so than 19) and other such bad luck (ie NO getting to choose where and, apparently, they chose really wisely). Maybe I'm delusional but it's still early

44 My take

I think he does as well as McCarthy at coming up with a game plan, and I think McCarthy was above average at that. I think he may be NFL average at adapting that plan to adjust for in game circumstances, I say might because I think McCarthy was awful at that, and since 75% of my viewing is Packers I know my sense of that is off. I think he is excellent at getting the team prepared to execute. 

I think he struggles if the team isn't executing. This is different that changing the game plan. The game plan attack the deep corners. The other team is cover 2 to prevent that. He will switch attacking the middle. The blocking is scheme is designed to handle outside pass rush. The opponent is twisting and attacking the middle of the line. He can adjust to that. But where he fails is when the plan is to block outside pressure, the other team is applying outside pressure and it's getting home. Now admittedly that is a problem for every coach and their nightmare. They are set-up to do something or stop something and still can't do it. McCarthy never really adjusted the plan if things weren't working, his response was to throw in wrinkles or try things to get that plan to work. That could actually deal with failed execution a little better. LaFleur has his schemes and if they don't work they don't work and he seems to get a bit panicky with high leverage decisions in those situations and that tends to lead to the blowouts.

His basic offensive game plan regardless of the primary focus is always deliberate though. Under him the Packers have been the slowest or close to the slowest paced team with or without a lead in both halfs. It clearly works most of time going 13-3, 1-1; 13-3, 1-1; 4-1. As you mentioned 30-7 regular season through 2 5/17th seasons and 2-2 in the playoffs is not bad. It's not ideal (the playoff losses) but it works. It's very different than McCarthy who tried to change pace of play depending on the situation. I think if it works it could mess with a defense. Because it doesn't matter what just happened, the Packers are going to take 32 seconds a play every play. The one exception is the "2:00 minute drill" offense which I think is generally "Here's your 3 plays go get em Aaron!"  I know in 2019 Rodgers was not fully onboard with the slower paced play and he bristled a bit which is part of why he actually was limited on his audibles.

I think part of that offensive style is to try and be a little more QB neutral. You've got time to survey the defense, you've got lots of motion options to help with reads, we've got designs to get people open deep so we can still score quickly. Just stay calm take your time, run the play.

The extra dependence on Adams this year is worrisome. Adams is targeted on I think 38% of all Rodgers targets vs the 28% average in the past 3 years. The other WR are way down. Usually the next 3 WR in targets will have more targets than Adams does. This year he's still nearly double all the other WR combined 61 to 38 (the next 3 have 40 but 2 of those targets were from Love). The running backs and TE aren't being targeted more than usual. It really is more so than ever Rodgers and Adams. I know the other WR are nothing special but they had their strengths and the scheme was usually solid enough to help with their issues. But they have done nothing through the first 5 games. Yeah Cobb has had a few splash plays but 6 targets is his season high. Adams season low was the disaster vs NO and he had 7. While healthy last year there were games Adams wasn't the top target.

 

As to the DVOA. You have some of. The blowouts do suppress it. But as Aaron has pointed out keeping all that data still makes the model more predictive. The defensive issues, especially not being able to stop anyone in the redzone hurts too. Being prone to big plays doesn't help either. They never really make an opponent feel like they are out of the game. So it's rare that you see another team panicking and just abandoning their game plan. I think getting of plan would hurt mosts teams DVOA as you are probably trying to do things you haven't practiced quite as much. I'm not talking about improvising on a busted play, I'm talking the whole we practiced deep curls a bunch but they aren't working so we are going to throw a bunch more slants that we only practiced a couple times. It's not a huge effect and it would be hard to measure but I imagine it would be there. So if the other team is always running their best offense that's going to hurt your defense rating even more.

 

So I'm sold on LaFleur at this point. I think he's an upgrade of McCarthy in nearly every category a coach needs to be good at. I think he's working with less offensive talent than the McCarthy Packers tended to have and getting better results with it. I think he can adjust the scheme to the talent he has. I've been impressed by what they've done with the OL with a bunch of guys thanks to all the injuries. Obviously Rodgers makes it look even better. If the front office would open the purse strings a bit more on paying coaches I think we could get more out of the defense, which actually might have more talent than most of the McCarthy era defenses. I actually think a good DC could be worth +/- 5% DVOA. So if you have talent that the average DC would get -5% DVOA out of a bad DC could suppress that to 0% and a good one improve it to -10%.  The effect may be larger or may not exist at all it would be basically impossible to study. I think the majority of special teams issues is coaching and not talent as well. Though the talent isn't great, but it's not bottom of the league talent and it's getting bottom of the league results.

 

 

57 I'm also starting to get…

In reply to by DisplacedPackerFan

I'm also starting to get concerned about their lack of targets to receivers not named Adams. On one hand they're doing a really nice job moving Davante all over the place to make it hard for defenses to double team him, and they've also been getting him open in creative ways against zone coverage. The quarters beater out of a 4x1 set they used on Sunday to get him wide open for a 34 yard gain on 3rd and 12 was a thing of beauty. So they've managed so far...

But yeah, the Packers got so many easy yards last year in the short flats from their secondary receivers on bootlegs and mesh. Defenses have keyed on that big time so far this year - they're making sure Tonyan doesn't get easy releases into the flat on play action (big reason why his production is down so far) and the deep zone defenders are driving downhill on drag routes. I'd love to see them scheme a few targets downfield for the rest of the group the way they have for Adams. One, they should be able to take advantage of the attention that Adams is going to get every down. (And they have done this, just not a ton so far.) Two, unfortunately Davante takes a lot of hits and tends to miss at least a couple of games every year - you want to have guys ready.

34 I don't agree with the…

I don't agree with the article's interpretation of the "great offense, bad defense through 5 weeks" table. It looks like the five most recent teams with this type of start finished with a bottom 3 defense, and the earlier ones all still finished in the bottom 10. That doesn't seem to support the notion that most defenses that have started poorly have improved over the course of the season. 

I'm surprised none of the mid-2010s Saints teams or the 2000 Rams didn't make that table. 

41 I laughed when I saw the…

I laughed when I saw the 2006 Bears with the 4th best DVOA ever through five games. I remember telling my older brother that year that the Bears were beating teams as badly as they did in '85, and he almost bit my head off. And the '85 team is nowhere to be found on that list---or even on the best defenses through five games. People forget that their first game in '85 was a close, high-scoring win over Tampa Bay, and they didn't start to really curb-stomp people until the middle of the season.

42 Does DVOA account for 4th down decisions?

As a Chargers fan, it's struck me this year that their per-play stats might not be much better, but their offense as a whole is more effective because they're going for it so often on 4th down. Does this factor into DVOA at all? Obviously the success/failure of those plays would count, but if we're using DVOA as a model to predict future victories, the fact that the Chargers are *giving themselves more plays* would affect their win-loss rate in a way that per-play efficiency doesn't cover.

Probably especially so for the Chargers since their STs are always atrocious so the more they can avoid using them the better...

46 I so agree with you arandom

I made this comment in open discussion while watching the Chargers game.  I expect DVOA to not like all of the failed plays that caused the 4th downs. Staley’s brilliant coaching is getting the most out of statistics that are not tremendous.

Before teams would ever go on 4th down it was an analytic saying that the key to running a successful offense is not your 3rd down conversion rate.  The key to a successful offense is not getting to 3rd down in the first place.  Clearly needing a bunch of 4th down conversions is hardly ideal.  

DVOA does not take into account coaching decisions to the best of my knowledge, but I would rather Aaron address that question.

 

55 Nope

In reply to by bedfordp

You're right, that DVOA does not give extra credit for trying fourth downs. It's something that's on my list of things to look into in the future, but it is something that makes the Chargers more successful than DVOA would otherwise indicate. On the other hand, one of these days they're going to actually fail on one of these fourth downs instead of picking them all up.

49 Trade for kicker

Has there ever been a high profile trade for a place kicker. I think the Chargers should trade for Younghoe Koo. Should massively help that ST rank of 29. He has a great technique and hasn’t missed a kick all year and surely ATL would (should) be interested. May make all the difference - most common result is a 3 point win

53 The Chargers had Younghoe…

In reply to by bedfordp

The Chargers had Younghoe Koo and cut him a month into the 2017 season. Who knows, maybe its water under the bridge now, but it would be funny if they tried to get him back.

58 Trading for a kicker

In reply to by bedfordp

I tried to look for kicker trades, I think that you idea is brilliant for the Chargers at the right price.  What would the cost be however?  That is unknown.  To me is seems that kickers values are unknown.

The only trades at all that I can find are:

1. The Tampa Bay Buccaneers drafted Aguayo with the 59th overall pick in the second round of the 2016 NFL Draft. They acquired the pick in a trade with the Kansas City Chiefs, sending their third and fourth-round selections.  Aguayo lasted 1 year for Tampa, going 4 for 11 from the 40 and beyond.

2.  Baltimore traded Kaare Vedvik to the Minnesota Vikings in 2019 for a fifth-round pick in the 2020 draft.  Vedvik, then proceeded to not make the team as Dan Bailey won the job.  Vedvik has one career NFL kick and he missed it.  Baltimore turned the 5th round pick into Calais Campbell.  Baltimore tries each pre season to replicate this, and they had a shot until this years preseason kicker had a dud of a 3rd game.

3.  This year Carolina traded a conditional seventh-round pick to the New York Giants for kicker and punter Ryan Santoso. The Giants will receive the pick if Santoso is on the Panthers’ roster for at least two regular-season games. Santoso has already been released by the Panthers.

The biggest draft selection was Sebastian Janikowski taken with the 17th pick in the first round.  He did kick for 17 years for the Raiders, so although an unorthodox selection, it was hardly a bust.  Yes, surely the Chargers would have won more games in a 17 year period with a kicker of Janikowski's caliber, but to use a first round pick is highly questionable.  Hindsight always works as we know.

Its only three trades, clearly won by the team that traded the kicker in two, with one being irrelevant , but I love your concept for the Chargers.

I am the Falcons, you know that I am listening, what are you willing to give up for Koo?

59 Justin Tucker went undrafted ...

... and Janikowski was a first-round pick. I don't think the NFL will ever get it right when it comes to assessing kicking talents, and it's probably one of the more overlooked aspects in what makes the league so ultra-competitive year after year.

61 Kicker Value

Assessing kicker value is extremely difficult, because they only play a few plays per game (especially now with the vast majority of kickoffs being touchbacks), but they are such high leverage plays.  The difference between getting 3 points and being able to kickoff and most likely give your opponent the ball at the 25, and getting 0 points, and giving your opponent the ball at the sport of your miss is clearly huge.  This is to say nothing of the importance of extra points, which are also a lot more important than the average fan would probably realize.  

The problem is that it seems like teams often don't know who the good kickers are until they already have had success in the NFL.  If you are going to use valuable draft capital on a kicker, you had BETTER get it right, and I'm not sure there is enough of a track record of this actually happening.  You mentioned the Janikowski pick, and while he was a pretty good kicker for a long time, I think most people would agree it was a terrible pick, if for no other reason than it was completely unnecessary.  The Raiders could have almost assuredly gotten him several rounds later.  To this point, with the benefit of hindsight, I think you could argue very easily that Justin Tucker has provided 1st round value to the Ravens, but it would have still been a terrible pick because they didn't and wouldn't have needed to pick him that high.

This is why the original thought of trading for an established kicker is so interesting.  It's almost never done, but it seems that rebuilding teams with quality kickers would want the extra draft pick, and contenders with a shaky kicker should be jumping at the opportunity.  Just have to figure out what the correct value should be.

63 I would argue that…

In reply to by Mark H

I would argue that Janikowski did provide good value for the Raiders first round selection. I think if I was building a team and was offered a guarantee of 15+ seasons of solidly above average kicking, with the certainty that provides, I would gladly expend a first round pick. The problem is that projecting kicking performance is so damned difficult, even for seasoned NFL veterans. For college kickers it seems to be nigh on impossible. And borderline average kickers can be acquired for nothing. It's a pretty terrible bet. 

To that end it is interesting to speculate what Justin Tucker is worth in a trade. Even though he has been excellent now for a number of seasons, there remains a sneaking suspicion that regression is due at some stage soon, just because he is such an outlier. Would/should the Chargers trade a first for him? I suspect not, but I don't know.

67 Remember

1st round contracts are only 5 years. After that you're paying for it. And if you know he's "solidly above average" he'll ask for it in his contract.

I'm a pretty staunch on no drafting kickers (or ST in general) due to articles like this (wish they would update it, wish a lot of articles updated every year or so, alas).

And the ultimate value of ST in general is like 9 snaps a game. That's just a such a small % of plays that, if invested in wisely (and since there's evidence that kickers are truly a coin flip in the draft, for whatever reason) it's better off spending whatever pick on a guy that could possibly blossom into an actual starter. Maybe that pick becomes a starting guard and gets you that extra couple yards to where the difference in kicker doesn't matter. Not to mention all the other plays he could make an impact. 

Trading for established kickers is more acceptable. Although a 1st is surely too much for the reasons you stated and even a 2nd (unless you're REALLY set at everything else and are that good, nay, elite of a team). Maybe I'll give up a 3rd but even a day 2 pick feels rich and shortsighted. Even that 5th round pick for the Ravens K the Vikes got a couple years back, and he had some good preseason tape,... amounted to him not even making the final 53 and is out of the league now. Special teams is just so fickle. I would honestly just rotate them once their contract is up. Find the cheapest guys and have em battle it out seems like the way to go. Invest that other money in a LB6 that perhaps becomes a key role player due to injury or whatever. 

70 Vital

In reply to by ImNewAroundThe…

LB6 is never going to have the game in the line in what more than 1 occasion in his career

The kicker will have the game on the line a few times a year

i think in a $BN industry you should be able to have guys who can get a handle on kicker technique, couple this with some mental psychology you should improve your chance somewhat 

My two cents worth is the kicking action needs to be simple set up quick to the ball and repeatable. I like the way Koo does this

75 This was key

In reply to by bedfordp

"And the ultimate value of ST in general is like 9 snaps a game. That's just a such a small % of plays that, if invested in wisely (and since there's evidence that kickers are truly a coin flip in the draft, for whatever reason) it's better off spending whatever pick on a guy that could possibly blossom into an actual starter. Maybe that pick becomes a starting guard and gets you that extra couple yards to where the difference in kicker doesn't matter. Not to mention all the other plays he could make an impact."

You're playing more than one LB. And frequently. Injuries happen. Especially at that position. Contracts don't always line up perfectly. But kickers are still being put out there due to lack of offensive failure in some regard. And most kicks aren't that hard at this level. So for the most part kickers aren't super distinguishable inside a certain range that they usually end up kicking in. 

I said don't draft them, obviously you have to roster one. But you bring up Koo, who was undrafted. A case in point. 

71 Oh sure I wasn't suggesting…

In reply to by ImNewAroundThe…

Oh sure I wasn't suggesting Janikowski was a good pick. It was a terrible pick for all the reasons outlined above. But if you knew for certain his career was going to pan out the way it did, that's easily worth a 1st round pick (and probably more). Sustained above average kicking performance is very valuable. But it seems almost impossible to project. 

76 Yeah

Perhaps, but are you willing to pay for such a player after 8 years instead of just drafting a more impactful, every down player, like Keith Bulluck (also a 1x pro bowler like Seabass)? 

Thinking in guarantees isn't reality so I'm not seeing the benefit of discussing it too much. If the draft guaranteed anything, well that changes... a lot, which just distorts thinking since the position we're talking about, well, is the exact opposite and K is like the position that's furthest from a guarantee. 

Kinda like this boards discussion on a "robopunter," it's moot point I don't find particularly helpful irl when it comes to team building philosophy. 

64 Point well taken on Janikowski

In reply to by Mark H

Even in hindsight of his career you are correct, why on earth take him in round 1, he will surely be there for Oakland later in the draft.

 

69 Koo

He is on a 1 year $920k deal and then becomes an UFA. They would need to trade and extend him but he hasn’t earned much money yet so surely would be interested in that. From ATL perspective even if they really rate him they surely should be looking at draft capital as they need a rebuild.

Koo made 37-of-39 field goal attempts (94.9 percent), including 8-of-8 from at least 50 yards and 33 extra point attempts for a career-high 144 points in 15 games last season to earn Pro Bowl honors for the first time in his career. Koo's 37 made field goals led the NFL and surpassed Matt Bryant (35 in 2016, 2017) for the most made field goals in a single season in franchise history. From Week 4-15, Koo made 27 consecutive field goals – the longest such streak in a single season in team history and the second-longest streak in franchise history overall. Last season, Koo also became the first player to make at least eight field goals of 50-or-more yards without a miss in a season in team history.

Would they trade for a 5th? 

 

52 Tough division

The NFC West dropped five net points in DVOA ranking after battling itself last week, and given the injuries sustained to AZ, SF and Seattle, probably will end up dropping even more.

60 London games

Great job by the NFL scheduling the bottom 4 teams by DVOA for a London round robin on back-to-back weekends. The loser should get relegated 

72 Living in London I watched…

Living in London I watched last weeks games (and know a few people who went). It was indeed mostly dreadful, although my main takeaway was that Atlanta probably are not going to stay at the very bottom of the DVOA (the Jets are).

As for this week: it sounds as though Tua is going to suit up, so at least we get to see a couple of good, young QB prospects. 

78 Atlanta surprise

I did not expect them to rank last in DVOA, figuring that spot was reserved for the Jets or Jaguars. Did they score lower than the Jets in DVOA for last week's game despite winning? Just curious.